summaryrefslogtreecommitdiff
path: root/site/docs/1.5.0/api/python/_modules/pyspark/rdd.html
blob: 94f50f803a184077ee38065385f8c7a082c46030 (plain) (blame) 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
2143
2144
2145
2146
2147
2148
2149
2150
2151
2152
2153
2154
2155
2156
2157
2158
2159
2160
2161
2162
2163
2164
2165
2166
2167
2168
2169
2170
2171
2172
2173
2174
2175
2176
2177
2178
2179
2180
2181
2182
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202
2203
2204
2205
2206
2207
2208
2209
2210
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
2225
2226
2227
2228
2229
2230
2231
2232
2233
2234
2235
2236
2237
2238
2239
2240
2241
2242
2243
2244
2245
2246
2247
2248
2249
2250
2251
2252
2253
2254
2255
2256
2257
2258
2259
2260
2261
2262
2263
2264
2265
2266
2267
2268
2269
2270
2271
2272
2273
2274
2275
2276
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
2289
2290
2291
2292
2293
2294
2295
2296
2297
2298
2299
2300
2301
2302
2303
2304
2305
2306
2307
2308
2309
2310
2311
2312
2313
2314
2315
2316
2317
2318
2319
2320
2321
2322
2323
2324
2325
2326
2327
2328
2329
2330
2331
2332
2333
2334
2335
2336
2337
2338
2339
2340
2341
2342
2343
2344
2345
2346
2347
2348
2349
2350
2351
2352
2353
2354
2355
2356
2357
2358
2359
2360
2361
2362
2363
2364
2365
2366
2367
2368
2369
2370
2371
2372
2373
2374
2375
2376
2377
2378
2379
2380
2381
2382
2383
2384
2385
2386
2387
2388
2389
2390
2391
2392
2393
2394
2395
2396
2397
2398
2399
2400
2401
2402
2403
2404
2405
2406
2407
2408
2409
2410
2411
2412
2413
2414
2415
2416
2417
2418
2419
2420
2421
2422
2423
2424
2425
2426
2427
2428
2429
2430
2431
2432
2433
2434
2435
2436
2437
2438
2439
2440
2441
2442
2443
2444
2445
2446
2447
2448
2449
2450
2451
2452
2453
2454
2455
2456
2457
2458
2459
2460
2461
2462
2463
2464
2465
2466
2467
2468
2469
2470
2471
2472
2473
2474
2475
2476
2477
2478
2479
2480
2481
2482
2483
2484
2485
2486
2487
2488
2489
2490
2491
2492
2493
2494
2495
2496
2497
2498
2499
2500
2501
2502
2503
2504
2505

<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
  "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">


<html xmlns="http://www.w3.org/1999/xhtml">
  <head>
    <meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
    
    <title>pyspark.rdd &mdash; PySpark master documentation</title>
    
    <link rel="stylesheet" href="../../_static/nature.css" type="text/css" />
    <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
    
    <script type="text/javascript">
      var DOCUMENTATION_OPTIONS = {
        URL_ROOT:    '../../',
        VERSION:     'master',
        COLLAPSE_INDEX: false,
        FILE_SUFFIX: '.html',
        HAS_SOURCE:  true
      };
    </script>
    <script type="text/javascript" src="../../_static/jquery.js"></script>
    <script type="text/javascript" src="../../_static/underscore.js"></script>
    <script type="text/javascript" src="../../_static/doctools.js"></script>
    <link rel="top" title="PySpark master documentation" href="../../index.html" />
    <link rel="up" title="Module code" href="../index.html" /> 
  </head>
  <body role="document">
    <div class="related" role="navigation" aria-label="related navigation">
      <h3>Navigation</h3>
      <ul>
        <li class="nav-item nav-item-0"><a href="../../index.html">PySpark master documentation</a> &raquo;</li>
          <li class="nav-item nav-item-1"><a href="../index.html" accesskey="U">Module code</a> &raquo;</li> 
      </ul>
    </div>  

    <div class="document">
      <div class="documentwrapper">
        <div class="bodywrapper">
          <div class="body" role="main">
            
  <h1>Source code for pyspark.rdd</h1><div class="highlight"><pre>
<span class="c">#</span>
<span class="c"># Licensed to the Apache Software Foundation (ASF) under one or more</span>
<span class="c"># contributor license agreements.  See the NOTICE file distributed with</span>
<span class="c"># this work for additional information regarding copyright ownership.</span>
<span class="c"># The ASF licenses this file to You under the Apache License, Version 2.0</span>
<span class="c"># (the &quot;License&quot;); you may not use this file except in compliance with</span>
<span class="c"># the License.  You may obtain a copy of the License at</span>
<span class="c">#</span>
<span class="c">#    http://www.apache.org/licenses/LICENSE-2.0</span>
<span class="c">#</span>
<span class="c"># Unless required by applicable law or agreed to in writing, software</span>
<span class="c"># distributed under the License is distributed on an &quot;AS IS&quot; BASIS,</span>
<span class="c"># WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.</span>
<span class="c"># See the License for the specific language governing permissions and</span>
<span class="c"># limitations under the License.</span>
<span class="c">#</span>

<span class="kn">import</span> <span class="nn">copy</span>
<span class="kn">import</span> <span class="nn">sys</span>
<span class="kn">import</span> <span class="nn">os</span>
<span class="kn">import</span> <span class="nn">re</span>
<span class="kn">import</span> <span class="nn">operator</span>
<span class="kn">import</span> <span class="nn">shlex</span>
<span class="kn">import</span> <span class="nn">warnings</span>
<span class="kn">import</span> <span class="nn">heapq</span>
<span class="kn">import</span> <span class="nn">bisect</span>
<span class="kn">import</span> <span class="nn">random</span>
<span class="kn">import</span> <span class="nn">socket</span>
<span class="kn">from</span> <span class="nn">subprocess</span> <span class="kn">import</span> <span class="n">Popen</span><span class="p">,</span> <span class="n">PIPE</span>
<span class="kn">from</span> <span class="nn">tempfile</span> <span class="kn">import</span> <span class="n">NamedTemporaryFile</span>
<span class="kn">from</span> <span class="nn">threading</span> <span class="kn">import</span> <span class="n">Thread</span>
<span class="kn">from</span> <span class="nn">collections</span> <span class="kn">import</span> <span class="n">defaultdict</span>
<span class="kn">from</span> <span class="nn">itertools</span> <span class="kn">import</span> <span class="n">chain</span>
<span class="kn">from</span> <span class="nn">functools</span> <span class="kn">import</span> <span class="nb">reduce</span>
<span class="kn">from</span> <span class="nn">math</span> <span class="kn">import</span> <span class="n">sqrt</span><span class="p">,</span> <span class="n">log</span><span class="p">,</span> <span class="n">isinf</span><span class="p">,</span> <span class="n">isnan</span><span class="p">,</span> <span class="nb">pow</span><span class="p">,</span> <span class="n">ceil</span>

<span class="k">if</span> <span class="n">sys</span><span class="o">.</span><span class="n">version</span> <span class="o">&gt;</span> <span class="s">&#39;3&#39;</span><span class="p">:</span>
    <span class="nb">basestring</span> <span class="o">=</span> <span class="nb">unicode</span> <span class="o">=</span> <span class="nb">str</span>
<span class="k">else</span><span class="p">:</span>
    <span class="kn">from</span> <span class="nn">itertools</span> <span class="kn">import</span> <span class="n">imap</span> <span class="k">as</span> <span class="nb">map</span><span class="p">,</span> <span class="n">ifilter</span> <span class="k">as</span> <span class="nb">filter</span>

<span class="kn">from</span> <span class="nn">pyspark.serializers</span> <span class="kn">import</span> <span class="n">NoOpSerializer</span><span class="p">,</span> <span class="n">CartesianDeserializer</span><span class="p">,</span> \
    <span class="n">BatchedSerializer</span><span class="p">,</span> <span class="n">CloudPickleSerializer</span><span class="p">,</span> <span class="n">PairDeserializer</span><span class="p">,</span> \
    <span class="n">PickleSerializer</span><span class="p">,</span> <span class="n">pack_long</span><span class="p">,</span> <span class="n">AutoBatchedSerializer</span>
<span class="kn">from</span> <span class="nn">pyspark.join</span> <span class="kn">import</span> <span class="n">python_join</span><span class="p">,</span> <span class="n">python_left_outer_join</span><span class="p">,</span> \
    <span class="n">python_right_outer_join</span><span class="p">,</span> <span class="n">python_full_outer_join</span><span class="p">,</span> <span class="n">python_cogroup</span>
<span class="kn">from</span> <span class="nn">pyspark.statcounter</span> <span class="kn">import</span> <span class="n">StatCounter</span>
<span class="kn">from</span> <span class="nn">pyspark.rddsampler</span> <span class="kn">import</span> <span class="n">RDDSampler</span><span class="p">,</span> <span class="n">RDDRangeSampler</span><span class="p">,</span> <span class="n">RDDStratifiedSampler</span>
<span class="kn">from</span> <span class="nn">pyspark.storagelevel</span> <span class="kn">import</span> <span class="n">StorageLevel</span>
<span class="kn">from</span> <span class="nn">pyspark.resultiterable</span> <span class="kn">import</span> <span class="n">ResultIterable</span>
<span class="kn">from</span> <span class="nn">pyspark.shuffle</span> <span class="kn">import</span> <span class="n">Aggregator</span><span class="p">,</span> <span class="n">InMemoryMerger</span><span class="p">,</span> <span class="n">ExternalMerger</span><span class="p">,</span> \
    <span class="n">get_used_memory</span><span class="p">,</span> <span class="n">ExternalSorter</span><span class="p">,</span> <span class="n">ExternalGroupBy</span>
<span class="kn">from</span> <span class="nn">pyspark.traceback_utils</span> <span class="kn">import</span> <span class="n">SCCallSiteSync</span>

<span class="kn">from</span> <span class="nn">py4j.java_collections</span> <span class="kn">import</span> <span class="n">ListConverter</span><span class="p">,</span> <span class="n">MapConverter</span>


<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span><span class="s">&quot;RDD&quot;</span><span class="p">]</span>


<span class="k">def</span> <span class="nf">portable_hash</span><span class="p">(</span><span class="n">x</span><span class="p">):</span>
    <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">    This function returns consistent hash code for builtin types, especially</span>
<span class="sd">    for None and tuple with None.</span>

<span class="sd">    The algorithm is similar to that one used by CPython 2.7</span>

<span class="sd">    &gt;&gt;&gt; portable_hash(None)</span>
<span class="sd">    0</span>
<span class="sd">    &gt;&gt;&gt; portable_hash((None, 1)) &amp; 0xffffffff</span>
<span class="sd">    219750521</span>
<span class="sd">    &quot;&quot;&quot;</span>
    <span class="k">if</span> <span class="n">sys</span><span class="o">.</span><span class="n">version</span> <span class="o">&gt;=</span> <span class="s">&#39;3.3&#39;</span> <span class="ow">and</span> <span class="s">&#39;PYTHONHASHSEED&#39;</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">os</span><span class="o">.</span><span class="n">environ</span><span class="p">:</span>
        <span class="k">raise</span> <span class="ne">Exception</span><span class="p">(</span><span class="s">&quot;Randomness of hash of string should be disabled via PYTHONHASHSEED&quot;</span><span class="p">)</span>

    <span class="k">if</span> <span class="n">x</span> <span class="ow">is</span> <span class="bp">None</span><span class="p">:</span>
        <span class="k">return</span> <span class="mi">0</span>
    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="nb">tuple</span><span class="p">):</span>
        <span class="n">h</span> <span class="o">=</span> <span class="mh">0x345678</span>
        <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="n">x</span><span class="p">:</span>
            <span class="n">h</span> <span class="o">^=</span> <span class="n">portable_hash</span><span class="p">(</span><span class="n">i</span><span class="p">)</span>
            <span class="n">h</span> <span class="o">*=</span> <span class="mi">1000003</span>
            <span class="n">h</span> <span class="o">&amp;=</span> <span class="n">sys</span><span class="o">.</span><span class="n">maxsize</span>
        <span class="n">h</span> <span class="o">^=</span> <span class="nb">len</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
        <span class="k">if</span> <span class="n">h</span> <span class="o">==</span> <span class="o">-</span><span class="mi">1</span><span class="p">:</span>
            <span class="n">h</span> <span class="o">=</span> <span class="o">-</span><span class="mi">2</span>
        <span class="k">return</span> <span class="n">h</span>
    <span class="k">return</span> <span class="nb">hash</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>


<span class="k">class</span> <span class="nc">BoundedFloat</span><span class="p">(</span><span class="nb">float</span><span class="p">):</span>
    <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">    Bounded value is generated by approximate job, with confidence and low</span>
<span class="sd">    bound and high bound.</span>

<span class="sd">    &gt;&gt;&gt; BoundedFloat(100.0, 0.95, 95.0, 105.0)</span>
<span class="sd">    100.0</span>
<span class="sd">    &quot;&quot;&quot;</span>
    <span class="k">def</span> <span class="nf">__new__</span><span class="p">(</span><span class="n">cls</span><span class="p">,</span> <span class="n">mean</span><span class="p">,</span> <span class="n">confidence</span><span class="p">,</span> <span class="n">low</span><span class="p">,</span> <span class="n">high</span><span class="p">):</span>
        <span class="n">obj</span> <span class="o">=</span> <span class="nb">float</span><span class="o">.</span><span class="n">__new__</span><span class="p">(</span><span class="n">cls</span><span class="p">,</span> <span class="n">mean</span><span class="p">)</span>
        <span class="n">obj</span><span class="o">.</span><span class="n">confidence</span> <span class="o">=</span> <span class="n">confidence</span>
        <span class="n">obj</span><span class="o">.</span><span class="n">low</span> <span class="o">=</span> <span class="n">low</span>
        <span class="n">obj</span><span class="o">.</span><span class="n">high</span> <span class="o">=</span> <span class="n">high</span>
        <span class="k">return</span> <span class="n">obj</span>


<span class="k">def</span> <span class="nf">_parse_memory</span><span class="p">(</span><span class="n">s</span><span class="p">):</span>
    <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">    Parse a memory string in the format supported by Java (e.g. 1g, 200m) and</span>
<span class="sd">    return the value in MB</span>

<span class="sd">    &gt;&gt;&gt; _parse_memory(&quot;256m&quot;)</span>
<span class="sd">    256</span>
<span class="sd">    &gt;&gt;&gt; _parse_memory(&quot;2g&quot;)</span>
<span class="sd">    2048</span>
<span class="sd">    &quot;&quot;&quot;</span>
    <span class="n">units</span> <span class="o">=</span> <span class="p">{</span><span class="s">&#39;g&#39;</span><span class="p">:</span> <span class="mi">1024</span><span class="p">,</span> <span class="s">&#39;m&#39;</span><span class="p">:</span> <span class="mi">1</span><span class="p">,</span> <span class="s">&#39;t&#39;</span><span class="p">:</span> <span class="mi">1</span> <span class="o">&lt;&lt;</span> <span class="mi">20</span><span class="p">,</span> <span class="s">&#39;k&#39;</span><span class="p">:</span> <span class="mf">1.0</span> <span class="o">/</span> <span class="mi">1024</span><span class="p">}</span>
    <span class="k">if</span> <span class="n">s</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">units</span><span class="p">:</span>
        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s">&quot;invalid format: &quot;</span> <span class="o">+</span> <span class="n">s</span><span class="p">)</span>
    <span class="k">return</span> <span class="nb">int</span><span class="p">(</span><span class="nb">float</span><span class="p">(</span><span class="n">s</span><span class="p">[:</span><span class="o">-</span><span class="mi">1</span><span class="p">])</span> <span class="o">*</span> <span class="n">units</span><span class="p">[</span><span class="n">s</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span><span class="o">.</span><span class="n">lower</span><span class="p">()])</span>


<span class="k">def</span> <span class="nf">_load_from_socket</span><span class="p">(</span><span class="n">port</span><span class="p">,</span> <span class="n">serializer</span><span class="p">):</span>
    <span class="n">sock</span> <span class="o">=</span> <span class="bp">None</span>
    <span class="c"># Support for both IPv4 and IPv6.</span>
    <span class="c"># On most of IPv6-ready systems, IPv6 will take precedence.</span>
    <span class="k">for</span> <span class="n">res</span> <span class="ow">in</span> <span class="n">socket</span><span class="o">.</span><span class="n">getaddrinfo</span><span class="p">(</span><span class="s">&quot;localhost&quot;</span><span class="p">,</span> <span class="n">port</span><span class="p">,</span> <span class="n">socket</span><span class="o">.</span><span class="n">AF_UNSPEC</span><span class="p">,</span> <span class="n">socket</span><span class="o">.</span><span class="n">SOCK_STREAM</span><span class="p">):</span>
        <span class="n">af</span><span class="p">,</span> <span class="n">socktype</span><span class="p">,</span> <span class="n">proto</span><span class="p">,</span> <span class="n">canonname</span><span class="p">,</span> <span class="n">sa</span> <span class="o">=</span> <span class="n">res</span>
        <span class="n">sock</span> <span class="o">=</span> <span class="n">socket</span><span class="o">.</span><span class="n">socket</span><span class="p">(</span><span class="n">af</span><span class="p">,</span> <span class="n">socktype</span><span class="p">,</span> <span class="n">proto</span><span class="p">)</span>
        <span class="k">try</span><span class="p">:</span>
            <span class="n">sock</span><span class="o">.</span><span class="n">settimeout</span><span class="p">(</span><span class="mi">3</span><span class="p">)</span>
            <span class="n">sock</span><span class="o">.</span><span class="n">connect</span><span class="p">(</span><span class="n">sa</span><span class="p">)</span>
        <span class="k">except</span> <span class="n">socket</span><span class="o">.</span><span class="n">error</span><span class="p">:</span>
            <span class="n">sock</span><span class="o">.</span><span class="n">close</span><span class="p">()</span>
            <span class="n">sock</span> <span class="o">=</span> <span class="bp">None</span>
            <span class="k">continue</span>
        <span class="k">break</span>
    <span class="k">if</span> <span class="ow">not</span> <span class="n">sock</span><span class="p">:</span>
        <span class="k">raise</span> <span class="ne">Exception</span><span class="p">(</span><span class="s">&quot;could not open socket&quot;</span><span class="p">)</span>
    <span class="k">try</span><span class="p">:</span>
        <span class="n">rf</span> <span class="o">=</span> <span class="n">sock</span><span class="o">.</span><span class="n">makefile</span><span class="p">(</span><span class="s">&quot;rb&quot;</span><span class="p">,</span> <span class="mi">65536</span><span class="p">)</span>
        <span class="k">for</span> <span class="n">item</span> <span class="ow">in</span> <span class="n">serializer</span><span class="o">.</span><span class="n">load_stream</span><span class="p">(</span><span class="n">rf</span><span class="p">):</span>
            <span class="k">yield</span> <span class="n">item</span>
    <span class="k">finally</span><span class="p">:</span>
        <span class="n">sock</span><span class="o">.</span><span class="n">close</span><span class="p">()</span>


<span class="k">def</span> <span class="nf">ignore_unicode_prefix</span><span class="p">(</span><span class="n">f</span><span class="p">):</span>
    <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">    Ignore the &#39;u&#39; prefix of string in doc tests, to make it works</span>
<span class="sd">    in both python 2 and 3</span>
<span class="sd">    &quot;&quot;&quot;</span>
    <span class="k">if</span> <span class="n">sys</span><span class="o">.</span><span class="n">version</span> <span class="o">&gt;=</span> <span class="s">&#39;3&#39;</span><span class="p">:</span>
        <span class="c"># the representation of unicode string in Python 3 does not have prefix &#39;u&#39;,</span>
        <span class="c"># so remove the prefix &#39;u&#39; for doc tests</span>
        <span class="n">literal_re</span> <span class="o">=</span> <span class="n">re</span><span class="o">.</span><span class="n">compile</span><span class="p">(</span><span class="s">r&quot;(\W|^)[uU]([&#39;])&quot;</span><span class="p">,</span> <span class="n">re</span><span class="o">.</span><span class="n">UNICODE</span><span class="p">)</span>
        <span class="n">f</span><span class="o">.</span><span class="n">__doc__</span> <span class="o">=</span> <span class="n">literal_re</span><span class="o">.</span><span class="n">sub</span><span class="p">(</span><span class="s">r&#39;\1\2&#39;</span><span class="p">,</span> <span class="n">f</span><span class="o">.</span><span class="n">__doc__</span><span class="p">)</span>
    <span class="k">return</span> <span class="n">f</span>


<span class="k">class</span> <span class="nc">Partitioner</span><span class="p">(</span><span class="nb">object</span><span class="p">):</span>
    <span class="k">def</span> <span class="nf">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">numPartitions</span><span class="p">,</span> <span class="n">partitionFunc</span><span class="p">):</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">numPartitions</span> <span class="o">=</span> <span class="n">numPartitions</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">partitionFunc</span> <span class="o">=</span> <span class="n">partitionFunc</span>

    <span class="k">def</span> <span class="nf">__eq__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">other</span><span class="p">):</span>
        <span class="k">return</span> <span class="p">(</span><span class="nb">isinstance</span><span class="p">(</span><span class="n">other</span><span class="p">,</span> <span class="n">Partitioner</span><span class="p">)</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">numPartitions</span> <span class="o">==</span> <span class="n">other</span><span class="o">.</span><span class="n">numPartitions</span>
                <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">partitionFunc</span> <span class="o">==</span> <span class="n">other</span><span class="o">.</span><span class="n">partitionFunc</span><span class="p">)</span>

    <span class="k">def</span> <span class="nf">__call__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">k</span><span class="p">):</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">partitionFunc</span><span class="p">(</span><span class="n">k</span><span class="p">)</span> <span class="o">%</span> <span class="bp">self</span><span class="o">.</span><span class="n">numPartitions</span>


<div class="viewcode-block" id="RDD"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD">[docs]</a><span class="k">class</span> <span class="nc">RDD</span><span class="p">(</span><span class="nb">object</span><span class="p">):</span>

    <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">    A Resilient Distributed Dataset (RDD), the basic abstraction in Spark.</span>
<span class="sd">    Represents an immutable, partitioned collection of elements that can be</span>
<span class="sd">    operated on in parallel.</span>
<span class="sd">    &quot;&quot;&quot;</span>

    <span class="k">def</span> <span class="nf">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">jrdd</span><span class="p">,</span> <span class="n">ctx</span><span class="p">,</span> <span class="n">jrdd_deserializer</span><span class="o">=</span><span class="n">AutoBatchedSerializer</span><span class="p">(</span><span class="n">PickleSerializer</span><span class="p">())):</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">_jrdd</span> <span class="o">=</span> <span class="n">jrdd</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">is_cached</span> <span class="o">=</span> <span class="bp">False</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">is_checkpointed</span> <span class="o">=</span> <span class="bp">False</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span> <span class="o">=</span> <span class="n">ctx</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">_jrdd_deserializer</span> <span class="o">=</span> <span class="n">jrdd_deserializer</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">_id</span> <span class="o">=</span> <span class="n">jrdd</span><span class="o">.</span><span class="n">id</span><span class="p">()</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">partitioner</span> <span class="o">=</span> <span class="bp">None</span>

    <span class="k">def</span> <span class="nf">_pickled</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_reserialize</span><span class="p">(</span><span class="n">AutoBatchedSerializer</span><span class="p">(</span><span class="n">PickleSerializer</span><span class="p">()))</span>

<div class="viewcode-block" id="RDD.id"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.id">[docs]</a>    <span class="k">def</span> <span class="nf">id</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        A unique ID for this RDD (within its SparkContext).</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_id</span>
</div>
    <span class="k">def</span> <span class="nf">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_jrdd</span><span class="o">.</span><span class="n">toString</span><span class="p">()</span>

    <span class="k">def</span> <span class="nf">__getnewargs__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="c"># This method is called when attempting to pickle an RDD, which is always an error:</span>
        <span class="k">raise</span> <span class="ne">Exception</span><span class="p">(</span>
            <span class="s">&quot;It appears that you are attempting to broadcast an RDD or reference an RDD from an &quot;</span>
            <span class="s">&quot;action or transformation. RDD transformations and actions can only be invoked by the &quot;</span>
            <span class="s">&quot;driver, not inside of other transformations; for example, &quot;</span>
            <span class="s">&quot;rdd1.map(lambda x: rdd2.values.count() * x) is invalid because the values &quot;</span>
            <span class="s">&quot;transformation and count action cannot be performed inside of the rdd1.map &quot;</span>
            <span class="s">&quot;transformation. For more information, see SPARK-5063.&quot;</span>
        <span class="p">)</span>

    <span class="nd">@property</span>
    <span class="k">def</span> <span class="nf">context</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        The L{SparkContext} that this RDD was created on.</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span>

<div class="viewcode-block" id="RDD.cache"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.cache">[docs]</a>    <span class="k">def</span> <span class="nf">cache</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Persist this RDD with the default storage level (C{MEMORY_ONLY_SER}).</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">is_cached</span> <span class="o">=</span> <span class="bp">True</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">persist</span><span class="p">(</span><span class="n">StorageLevel</span><span class="o">.</span><span class="n">MEMORY_ONLY_SER</span><span class="p">)</span>
        <span class="k">return</span> <span class="bp">self</span>
</div>
<div class="viewcode-block" id="RDD.persist"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.persist">[docs]</a>    <span class="k">def</span> <span class="nf">persist</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">storageLevel</span><span class="o">=</span><span class="n">StorageLevel</span><span class="o">.</span><span class="n">MEMORY_ONLY_SER</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Set this RDD&#39;s storage level to persist its values across operations</span>
<span class="sd">        after the first time it is computed. This can only be used to assign</span>
<span class="sd">        a new storage level if the RDD does not have a storage level set yet.</span>
<span class="sd">        If no storage level is specified defaults to (C{MEMORY_ONLY_SER}).</span>

<span class="sd">        &gt;&gt;&gt; rdd = sc.parallelize([&quot;b&quot;, &quot;a&quot;, &quot;c&quot;])</span>
<span class="sd">        &gt;&gt;&gt; rdd.persist().is_cached</span>
<span class="sd">        True</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">is_cached</span> <span class="o">=</span> <span class="bp">True</span>
        <span class="n">javaStorageLevel</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="o">.</span><span class="n">_getJavaStorageLevel</span><span class="p">(</span><span class="n">storageLevel</span><span class="p">)</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">_jrdd</span><span class="o">.</span><span class="n">persist</span><span class="p">(</span><span class="n">javaStorageLevel</span><span class="p">)</span>
        <span class="k">return</span> <span class="bp">self</span>
</div>
<div class="viewcode-block" id="RDD.unpersist"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.unpersist">[docs]</a>    <span class="k">def</span> <span class="nf">unpersist</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Mark the RDD as non-persistent, and remove all blocks for it from</span>
<span class="sd">        memory and disk.</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">is_cached</span> <span class="o">=</span> <span class="bp">False</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">_jrdd</span><span class="o">.</span><span class="n">unpersist</span><span class="p">()</span>
        <span class="k">return</span> <span class="bp">self</span>
</div>
<div class="viewcode-block" id="RDD.checkpoint"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.checkpoint">[docs]</a>    <span class="k">def</span> <span class="nf">checkpoint</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Mark this RDD for checkpointing. It will be saved to a file inside the</span>
<span class="sd">        checkpoint directory set with L{SparkContext.setCheckpointDir()} and</span>
<span class="sd">        all references to its parent RDDs will be removed. This function must</span>
<span class="sd">        be called before any job has been executed on this RDD. It is strongly</span>
<span class="sd">        recommended that this RDD is persisted in memory, otherwise saving it</span>
<span class="sd">        on a file will require recomputation.</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">is_checkpointed</span> <span class="o">=</span> <span class="bp">True</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">_jrdd</span><span class="o">.</span><span class="n">rdd</span><span class="p">()</span><span class="o">.</span><span class="n">checkpoint</span><span class="p">()</span>
</div>
<div class="viewcode-block" id="RDD.isCheckpointed"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.isCheckpointed">[docs]</a>    <span class="k">def</span> <span class="nf">isCheckpointed</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Return whether this RDD has been checkpointed or not</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_jrdd</span><span class="o">.</span><span class="n">rdd</span><span class="p">()</span><span class="o">.</span><span class="n">isCheckpointed</span><span class="p">()</span>
</div>
<div class="viewcode-block" id="RDD.getCheckpointFile"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.getCheckpointFile">[docs]</a>    <span class="k">def</span> <span class="nf">getCheckpointFile</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Gets the name of the file to which this RDD was checkpointed</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="n">checkpointFile</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_jrdd</span><span class="o">.</span><span class="n">rdd</span><span class="p">()</span><span class="o">.</span><span class="n">getCheckpointFile</span><span class="p">()</span>
        <span class="k">if</span> <span class="n">checkpointFile</span><span class="o">.</span><span class="n">isDefined</span><span class="p">():</span>
            <span class="k">return</span> <span class="n">checkpointFile</span><span class="o">.</span><span class="n">get</span><span class="p">()</span>
</div>
<div class="viewcode-block" id="RDD.map"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.map">[docs]</a>    <span class="k">def</span> <span class="nf">map</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">f</span><span class="p">,</span> <span class="n">preservesPartitioning</span><span class="o">=</span><span class="bp">False</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Return a new RDD by applying a function to each element of this RDD.</span>

<span class="sd">        &gt;&gt;&gt; rdd = sc.parallelize([&quot;b&quot;, &quot;a&quot;, &quot;c&quot;])</span>
<span class="sd">        &gt;&gt;&gt; sorted(rdd.map(lambda x: (x, 1)).collect())</span>
<span class="sd">        [(&#39;a&#39;, 1), (&#39;b&#39;, 1), (&#39;c&#39;, 1)]</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">def</span> <span class="nf">func</span><span class="p">(</span><span class="n">_</span><span class="p">,</span> <span class="n">iterator</span><span class="p">):</span>
            <span class="k">return</span> <span class="nb">map</span><span class="p">(</span><span class="n">f</span><span class="p">,</span> <span class="n">iterator</span><span class="p">)</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">mapPartitionsWithIndex</span><span class="p">(</span><span class="n">func</span><span class="p">,</span> <span class="n">preservesPartitioning</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.flatMap"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.flatMap">[docs]</a>    <span class="k">def</span> <span class="nf">flatMap</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">f</span><span class="p">,</span> <span class="n">preservesPartitioning</span><span class="o">=</span><span class="bp">False</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Return a new RDD by first applying a function to all elements of this</span>
<span class="sd">        RDD, and then flattening the results.</span>

<span class="sd">        &gt;&gt;&gt; rdd = sc.parallelize([2, 3, 4])</span>
<span class="sd">        &gt;&gt;&gt; sorted(rdd.flatMap(lambda x: range(1, x)).collect())</span>
<span class="sd">        [1, 1, 1, 2, 2, 3]</span>
<span class="sd">        &gt;&gt;&gt; sorted(rdd.flatMap(lambda x: [(x, x), (x, x)]).collect())</span>
<span class="sd">        [(2, 2), (2, 2), (3, 3), (3, 3), (4, 4), (4, 4)]</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">def</span> <span class="nf">func</span><span class="p">(</span><span class="n">s</span><span class="p">,</span> <span class="n">iterator</span><span class="p">):</span>
            <span class="k">return</span> <span class="n">chain</span><span class="o">.</span><span class="n">from_iterable</span><span class="p">(</span><span class="nb">map</span><span class="p">(</span><span class="n">f</span><span class="p">,</span> <span class="n">iterator</span><span class="p">))</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">mapPartitionsWithIndex</span><span class="p">(</span><span class="n">func</span><span class="p">,</span> <span class="n">preservesPartitioning</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.mapPartitions"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.mapPartitions">[docs]</a>    <span class="k">def</span> <span class="nf">mapPartitions</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">f</span><span class="p">,</span> <span class="n">preservesPartitioning</span><span class="o">=</span><span class="bp">False</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Return a new RDD by applying a function to each partition of this RDD.</span>

<span class="sd">        &gt;&gt;&gt; rdd = sc.parallelize([1, 2, 3, 4], 2)</span>
<span class="sd">        &gt;&gt;&gt; def f(iterator): yield sum(iterator)</span>
<span class="sd">        &gt;&gt;&gt; rdd.mapPartitions(f).collect()</span>
<span class="sd">        [3, 7]</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">def</span> <span class="nf">func</span><span class="p">(</span><span class="n">s</span><span class="p">,</span> <span class="n">iterator</span><span class="p">):</span>
            <span class="k">return</span> <span class="n">f</span><span class="p">(</span><span class="n">iterator</span><span class="p">)</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">mapPartitionsWithIndex</span><span class="p">(</span><span class="n">func</span><span class="p">,</span> <span class="n">preservesPartitioning</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.mapPartitionsWithIndex"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.mapPartitionsWithIndex">[docs]</a>    <span class="k">def</span> <span class="nf">mapPartitionsWithIndex</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">f</span><span class="p">,</span> <span class="n">preservesPartitioning</span><span class="o">=</span><span class="bp">False</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Return a new RDD by applying a function to each partition of this RDD,</span>
<span class="sd">        while tracking the index of the original partition.</span>

<span class="sd">        &gt;&gt;&gt; rdd = sc.parallelize([1, 2, 3, 4], 4)</span>
<span class="sd">        &gt;&gt;&gt; def f(splitIndex, iterator): yield splitIndex</span>
<span class="sd">        &gt;&gt;&gt; rdd.mapPartitionsWithIndex(f).sum()</span>
<span class="sd">        6</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">return</span> <span class="n">PipelinedRDD</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">f</span><span class="p">,</span> <span class="n">preservesPartitioning</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.mapPartitionsWithSplit"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.mapPartitionsWithSplit">[docs]</a>    <span class="k">def</span> <span class="nf">mapPartitionsWithSplit</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">f</span><span class="p">,</span> <span class="n">preservesPartitioning</span><span class="o">=</span><span class="bp">False</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Deprecated: use mapPartitionsWithIndex instead.</span>

<span class="sd">        Return a new RDD by applying a function to each partition of this RDD,</span>
<span class="sd">        while tracking the index of the original partition.</span>

<span class="sd">        &gt;&gt;&gt; rdd = sc.parallelize([1, 2, 3, 4], 4)</span>
<span class="sd">        &gt;&gt;&gt; def f(splitIndex, iterator): yield splitIndex</span>
<span class="sd">        &gt;&gt;&gt; rdd.mapPartitionsWithSplit(f).sum()</span>
<span class="sd">        6</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span><span class="s">&quot;mapPartitionsWithSplit is deprecated; &quot;</span>
                      <span class="s">&quot;use mapPartitionsWithIndex instead&quot;</span><span class="p">,</span> <span class="ne">DeprecationWarning</span><span class="p">,</span> <span class="n">stacklevel</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">mapPartitionsWithIndex</span><span class="p">(</span><span class="n">f</span><span class="p">,</span> <span class="n">preservesPartitioning</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.getNumPartitions"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.getNumPartitions">[docs]</a>    <span class="k">def</span> <span class="nf">getNumPartitions</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Returns the number of partitions in RDD</span>

<span class="sd">        &gt;&gt;&gt; rdd = sc.parallelize([1, 2, 3, 4], 2)</span>
<span class="sd">        &gt;&gt;&gt; rdd.getNumPartitions()</span>
<span class="sd">        2</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_jrdd</span><span class="o">.</span><span class="n">partitions</span><span class="p">()</span><span class="o">.</span><span class="n">size</span><span class="p">()</span>
</div>
<div class="viewcode-block" id="RDD.filter"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.filter">[docs]</a>    <span class="k">def</span> <span class="nf">filter</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">f</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Return a new RDD containing only the elements that satisfy a predicate.</span>

<span class="sd">        &gt;&gt;&gt; rdd = sc.parallelize([1, 2, 3, 4, 5])</span>
<span class="sd">        &gt;&gt;&gt; rdd.filter(lambda x: x % 2 == 0).collect()</span>
<span class="sd">        [2, 4]</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">def</span> <span class="nf">func</span><span class="p">(</span><span class="n">iterator</span><span class="p">):</span>
            <span class="k">return</span> <span class="nb">filter</span><span class="p">(</span><span class="n">f</span><span class="p">,</span> <span class="n">iterator</span><span class="p">)</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">mapPartitions</span><span class="p">(</span><span class="n">func</span><span class="p">,</span> <span class="bp">True</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.distinct"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.distinct">[docs]</a>    <span class="k">def</span> <span class="nf">distinct</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">numPartitions</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Return a new RDD containing the distinct elements in this RDD.</span>

<span class="sd">        &gt;&gt;&gt; sorted(sc.parallelize([1, 1, 2, 3]).distinct().collect())</span>
<span class="sd">        [1, 2, 3]</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">map</span><span class="p">(</span><span class="k">lambda</span> <span class="n">x</span><span class="p">:</span> <span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="bp">None</span><span class="p">))</span> \
                   <span class="o">.</span><span class="n">reduceByKey</span><span class="p">(</span><span class="k">lambda</span> <span class="n">x</span><span class="p">,</span> <span class="n">_</span><span class="p">:</span> <span class="n">x</span><span class="p">,</span> <span class="n">numPartitions</span><span class="p">)</span> \
                   <span class="o">.</span><span class="n">map</span><span class="p">(</span><span class="k">lambda</span> <span class="n">x</span><span class="p">:</span> <span class="n">x</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span>
</div>
<div class="viewcode-block" id="RDD.sample"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.sample">[docs]</a>    <span class="k">def</span> <span class="nf">sample</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">withReplacement</span><span class="p">,</span> <span class="n">fraction</span><span class="p">,</span> <span class="n">seed</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Return a sampled subset of this RDD.</span>

<span class="sd">        :param withReplacement: can elements be sampled multiple times (replaced when sampled out)</span>
<span class="sd">        :param fraction: expected size of the sample as a fraction of this RDD&#39;s size</span>
<span class="sd">            without replacement: probability that each element is chosen; fraction must be [0, 1]</span>
<span class="sd">            with replacement: expected number of times each element is chosen; fraction must be &gt;= 0</span>
<span class="sd">        :param seed: seed for the random number generator</span>

<span class="sd">        &gt;&gt;&gt; rdd = sc.parallelize(range(100), 4)</span>
<span class="sd">        &gt;&gt;&gt; 6 &lt;= rdd.sample(False, 0.1, 81).count() &lt;= 14</span>
<span class="sd">        True</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">assert</span> <span class="n">fraction</span> <span class="o">&gt;=</span> <span class="mf">0.0</span><span class="p">,</span> <span class="s">&quot;Negative fraction value: </span><span class="si">%s</span><span class="s">&quot;</span> <span class="o">%</span> <span class="n">fraction</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">mapPartitionsWithIndex</span><span class="p">(</span><span class="n">RDDSampler</span><span class="p">(</span><span class="n">withReplacement</span><span class="p">,</span> <span class="n">fraction</span><span class="p">,</span> <span class="n">seed</span><span class="p">)</span><span class="o">.</span><span class="n">func</span><span class="p">,</span> <span class="bp">True</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.randomSplit"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.randomSplit">[docs]</a>    <span class="k">def</span> <span class="nf">randomSplit</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">seed</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Randomly splits this RDD with the provided weights.</span>

<span class="sd">        :param weights: weights for splits, will be normalized if they don&#39;t sum to 1</span>
<span class="sd">        :param seed: random seed</span>
<span class="sd">        :return: split RDDs in a list</span>

<span class="sd">        &gt;&gt;&gt; rdd = sc.parallelize(range(500), 1)</span>
<span class="sd">        &gt;&gt;&gt; rdd1, rdd2 = rdd.randomSplit([2, 3], 17)</span>
<span class="sd">        &gt;&gt;&gt; len(rdd1.collect() + rdd2.collect())</span>
<span class="sd">        500</span>
<span class="sd">        &gt;&gt;&gt; 150 &lt; rdd1.count() &lt; 250</span>
<span class="sd">        True</span>
<span class="sd">        &gt;&gt;&gt; 250 &lt; rdd2.count() &lt; 350</span>
<span class="sd">        True</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="n">s</span> <span class="o">=</span> <span class="nb">float</span><span class="p">(</span><span class="nb">sum</span><span class="p">(</span><span class="n">weights</span><span class="p">))</span>
        <span class="n">cweights</span> <span class="o">=</span> <span class="p">[</span><span class="mf">0.0</span><span class="p">]</span>
        <span class="k">for</span> <span class="n">w</span> <span class="ow">in</span> <span class="n">weights</span><span class="p">:</span>
            <span class="n">cweights</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">cweights</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span> <span class="o">+</span> <span class="n">w</span> <span class="o">/</span> <span class="n">s</span><span class="p">)</span>
        <span class="k">if</span> <span class="n">seed</span> <span class="ow">is</span> <span class="bp">None</span><span class="p">:</span>
            <span class="n">seed</span> <span class="o">=</span> <span class="n">random</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">2</span> <span class="o">**</span> <span class="mi">32</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span>
        <span class="k">return</span> <span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">mapPartitionsWithIndex</span><span class="p">(</span><span class="n">RDDRangeSampler</span><span class="p">(</span><span class="n">lb</span><span class="p">,</span> <span class="n">ub</span><span class="p">,</span> <span class="n">seed</span><span class="p">)</span><span class="o">.</span><span class="n">func</span><span class="p">,</span> <span class="bp">True</span><span class="p">)</span>
                <span class="k">for</span> <span class="n">lb</span><span class="p">,</span> <span class="n">ub</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">cweights</span><span class="p">,</span> <span class="n">cweights</span><span class="p">[</span><span class="mi">1</span><span class="p">:])]</span>

    <span class="c"># this is ported from scala/spark/RDD.scala</span></div>
<div class="viewcode-block" id="RDD.takeSample"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.takeSample">[docs]</a>    <span class="k">def</span> <span class="nf">takeSample</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">withReplacement</span><span class="p">,</span> <span class="n">num</span><span class="p">,</span> <span class="n">seed</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Return a fixed-size sampled subset of this RDD.</span>

<span class="sd">        &gt;&gt;&gt; rdd = sc.parallelize(range(0, 10))</span>
<span class="sd">        &gt;&gt;&gt; len(rdd.takeSample(True, 20, 1))</span>
<span class="sd">        20</span>
<span class="sd">        &gt;&gt;&gt; len(rdd.takeSample(False, 5, 2))</span>
<span class="sd">        5</span>
<span class="sd">        &gt;&gt;&gt; len(rdd.takeSample(False, 15, 3))</span>
<span class="sd">        10</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="n">numStDev</span> <span class="o">=</span> <span class="mf">10.0</span>

        <span class="k">if</span> <span class="n">num</span> <span class="o">&lt;</span> <span class="mi">0</span><span class="p">:</span>
            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s">&quot;Sample size cannot be negative.&quot;</span><span class="p">)</span>
        <span class="k">elif</span> <span class="n">num</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
            <span class="k">return</span> <span class="p">[]</span>

        <span class="n">initialCount</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">count</span><span class="p">()</span>
        <span class="k">if</span> <span class="n">initialCount</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
            <span class="k">return</span> <span class="p">[]</span>

        <span class="n">rand</span> <span class="o">=</span> <span class="n">random</span><span class="o">.</span><span class="n">Random</span><span class="p">(</span><span class="n">seed</span><span class="p">)</span>

        <span class="k">if</span> <span class="p">(</span><span class="ow">not</span> <span class="n">withReplacement</span><span class="p">)</span> <span class="ow">and</span> <span class="n">num</span> <span class="o">&gt;=</span> <span class="n">initialCount</span><span class="p">:</span>
            <span class="c"># shuffle current RDD and return</span>
            <span class="n">samples</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">collect</span><span class="p">()</span>
            <span class="n">rand</span><span class="o">.</span><span class="n">shuffle</span><span class="p">(</span><span class="n">samples</span><span class="p">)</span>
            <span class="k">return</span> <span class="n">samples</span>

        <span class="n">maxSampleSize</span> <span class="o">=</span> <span class="n">sys</span><span class="o">.</span><span class="n">maxsize</span> <span class="o">-</span> <span class="nb">int</span><span class="p">(</span><span class="n">numStDev</span> <span class="o">*</span> <span class="n">sqrt</span><span class="p">(</span><span class="n">sys</span><span class="o">.</span><span class="n">maxsize</span><span class="p">))</span>
        <span class="k">if</span> <span class="n">num</span> <span class="o">&gt;</span> <span class="n">maxSampleSize</span><span class="p">:</span>
            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
                <span class="s">&quot;Sample size cannot be greater than </span><span class="si">%d</span><span class="s">.&quot;</span> <span class="o">%</span> <span class="n">maxSampleSize</span><span class="p">)</span>

        <span class="n">fraction</span> <span class="o">=</span> <span class="n">RDD</span><span class="o">.</span><span class="n">_computeFractionForSampleSize</span><span class="p">(</span>
            <span class="n">num</span><span class="p">,</span> <span class="n">initialCount</span><span class="p">,</span> <span class="n">withReplacement</span><span class="p">)</span>
        <span class="n">samples</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">sample</span><span class="p">(</span><span class="n">withReplacement</span><span class="p">,</span> <span class="n">fraction</span><span class="p">,</span> <span class="n">seed</span><span class="p">)</span><span class="o">.</span><span class="n">collect</span><span class="p">()</span>

        <span class="c"># If the first sample didn&#39;t turn out large enough, keep trying to take samples;</span>
        <span class="c"># this shouldn&#39;t happen often because we use a big multiplier for their initial size.</span>
        <span class="c"># See: scala/spark/RDD.scala</span>
        <span class="k">while</span> <span class="nb">len</span><span class="p">(</span><span class="n">samples</span><span class="p">)</span> <span class="o">&lt;</span> <span class="n">num</span><span class="p">:</span>
            <span class="c"># TODO: add log warning for when more than one iteration was run</span>
            <span class="n">seed</span> <span class="o">=</span> <span class="n">rand</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">sys</span><span class="o">.</span><span class="n">maxsize</span><span class="p">)</span>
            <span class="n">samples</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">sample</span><span class="p">(</span><span class="n">withReplacement</span><span class="p">,</span> <span class="n">fraction</span><span class="p">,</span> <span class="n">seed</span><span class="p">)</span><span class="o">.</span><span class="n">collect</span><span class="p">()</span>

        <span class="n">rand</span><span class="o">.</span><span class="n">shuffle</span><span class="p">(</span><span class="n">samples</span><span class="p">)</span>

        <span class="k">return</span> <span class="n">samples</span><span class="p">[</span><span class="mi">0</span><span class="p">:</span><span class="n">num</span><span class="p">]</span>
</div>
    <span class="nd">@staticmethod</span>
    <span class="k">def</span> <span class="nf">_computeFractionForSampleSize</span><span class="p">(</span><span class="n">sampleSizeLowerBound</span><span class="p">,</span> <span class="n">total</span><span class="p">,</span> <span class="n">withReplacement</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Returns a sampling rate that guarantees a sample of</span>
<span class="sd">        size &gt;= sampleSizeLowerBound 99.99% of the time.</span>

<span class="sd">        How the sampling rate is determined:</span>
<span class="sd">        Let p = num / total, where num is the sample size and total is the</span>
<span class="sd">        total number of data points in the RDD. We&#39;re trying to compute</span>
<span class="sd">        q &gt; p such that</span>
<span class="sd">          - when sampling with replacement, we&#39;re drawing each data point</span>
<span class="sd">            with prob_i ~ Pois(q), where we want to guarantee</span>
<span class="sd">            Pr[s &lt; num] &lt; 0.0001 for s = sum(prob_i for i from 0 to</span>
<span class="sd">            total), i.e. the failure rate of not having a sufficiently large</span>
<span class="sd">            sample &lt; 0.0001. Setting q = p + 5 * sqrt(p/total) is sufficient</span>
<span class="sd">            to guarantee 0.9999 success rate for num &gt; 12, but we need a</span>
<span class="sd">            slightly larger q (9 empirically determined).</span>
<span class="sd">          - when sampling without replacement, we&#39;re drawing each data point</span>
<span class="sd">            with prob_i ~ Binomial(total, fraction) and our choice of q</span>
<span class="sd">            guarantees 1-delta, or 0.9999 success rate, where success rate is</span>
<span class="sd">            defined the same as in sampling with replacement.</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="n">fraction</span> <span class="o">=</span> <span class="nb">float</span><span class="p">(</span><span class="n">sampleSizeLowerBound</span><span class="p">)</span> <span class="o">/</span> <span class="n">total</span>
        <span class="k">if</span> <span class="n">withReplacement</span><span class="p">:</span>
            <span class="n">numStDev</span> <span class="o">=</span> <span class="mi">5</span>
            <span class="k">if</span> <span class="p">(</span><span class="n">sampleSizeLowerBound</span> <span class="o">&lt;</span> <span class="mi">12</span><span class="p">):</span>
                <span class="n">numStDev</span> <span class="o">=</span> <span class="mi">9</span>
            <span class="k">return</span> <span class="n">fraction</span> <span class="o">+</span> <span class="n">numStDev</span> <span class="o">*</span> <span class="n">sqrt</span><span class="p">(</span><span class="n">fraction</span> <span class="o">/</span> <span class="n">total</span><span class="p">)</span>
        <span class="k">else</span><span class="p">:</span>
            <span class="n">delta</span> <span class="o">=</span> <span class="mf">0.00005</span>
            <span class="n">gamma</span> <span class="o">=</span> <span class="o">-</span> <span class="n">log</span><span class="p">(</span><span class="n">delta</span><span class="p">)</span> <span class="o">/</span> <span class="n">total</span>
            <span class="k">return</span> <span class="nb">min</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">fraction</span> <span class="o">+</span> <span class="n">gamma</span> <span class="o">+</span> <span class="n">sqrt</span><span class="p">(</span><span class="n">gamma</span> <span class="o">*</span> <span class="n">gamma</span> <span class="o">+</span> <span class="mi">2</span> <span class="o">*</span> <span class="n">gamma</span> <span class="o">*</span> <span class="n">fraction</span><span class="p">))</span>

<div class="viewcode-block" id="RDD.union"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.union">[docs]</a>    <span class="k">def</span> <span class="nf">union</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">other</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Return the union of this RDD and another one.</span>

<span class="sd">        &gt;&gt;&gt; rdd = sc.parallelize([1, 1, 2, 3])</span>
<span class="sd">        &gt;&gt;&gt; rdd.union(rdd).collect()</span>
<span class="sd">        [1, 1, 2, 3, 1, 1, 2, 3]</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_jrdd_deserializer</span> <span class="o">==</span> <span class="n">other</span><span class="o">.</span><span class="n">_jrdd_deserializer</span><span class="p">:</span>
            <span class="n">rdd</span> <span class="o">=</span> <span class="n">RDD</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_jrdd</span><span class="o">.</span><span class="n">union</span><span class="p">(</span><span class="n">other</span><span class="o">.</span><span class="n">_jrdd</span><span class="p">),</span> <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="p">,</span>
                      <span class="bp">self</span><span class="o">.</span><span class="n">_jrdd_deserializer</span><span class="p">)</span>
        <span class="k">else</span><span class="p">:</span>
            <span class="c"># These RDDs contain data in different serialized formats, so we</span>
            <span class="c"># must normalize them to the default serializer.</span>
            <span class="n">self_copy</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_reserialize</span><span class="p">()</span>
            <span class="n">other_copy</span> <span class="o">=</span> <span class="n">other</span><span class="o">.</span><span class="n">_reserialize</span><span class="p">()</span>
            <span class="n">rdd</span> <span class="o">=</span> <span class="n">RDD</span><span class="p">(</span><span class="n">self_copy</span><span class="o">.</span><span class="n">_jrdd</span><span class="o">.</span><span class="n">union</span><span class="p">(</span><span class="n">other_copy</span><span class="o">.</span><span class="n">_jrdd</span><span class="p">),</span> <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="p">,</span>
                      <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="o">.</span><span class="n">serializer</span><span class="p">)</span>
        <span class="k">if</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">partitioner</span> <span class="o">==</span> <span class="n">other</span><span class="o">.</span><span class="n">partitioner</span> <span class="ow">and</span>
                <span class="bp">self</span><span class="o">.</span><span class="n">getNumPartitions</span><span class="p">()</span> <span class="o">==</span> <span class="n">rdd</span><span class="o">.</span><span class="n">getNumPartitions</span><span class="p">()):</span>
            <span class="n">rdd</span><span class="o">.</span><span class="n">partitioner</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">partitioner</span>
        <span class="k">return</span> <span class="n">rdd</span>
</div>
<div class="viewcode-block" id="RDD.intersection"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.intersection">[docs]</a>    <span class="k">def</span> <span class="nf">intersection</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">other</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Return the intersection of this RDD and another one. The output will</span>
<span class="sd">        not contain any duplicate elements, even if the input RDDs did.</span>

<span class="sd">        Note that this method performs a shuffle internally.</span>

<span class="sd">        &gt;&gt;&gt; rdd1 = sc.parallelize([1, 10, 2, 3, 4, 5])</span>
<span class="sd">        &gt;&gt;&gt; rdd2 = sc.parallelize([1, 6, 2, 3, 7, 8])</span>
<span class="sd">        &gt;&gt;&gt; rdd1.intersection(rdd2).collect()</span>
<span class="sd">        [1, 2, 3]</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">map</span><span class="p">(</span><span class="k">lambda</span> <span class="n">v</span><span class="p">:</span> <span class="p">(</span><span class="n">v</span><span class="p">,</span> <span class="bp">None</span><span class="p">))</span> \
            <span class="o">.</span><span class="n">cogroup</span><span class="p">(</span><span class="n">other</span><span class="o">.</span><span class="n">map</span><span class="p">(</span><span class="k">lambda</span> <span class="n">v</span><span class="p">:</span> <span class="p">(</span><span class="n">v</span><span class="p">,</span> <span class="bp">None</span><span class="p">)))</span> \
            <span class="o">.</span><span class="n">filter</span><span class="p">(</span><span class="k">lambda</span> <span class="n">k_vs</span><span class="p">:</span> <span class="nb">all</span><span class="p">(</span><span class="n">k_vs</span><span class="p">[</span><span class="mi">1</span><span class="p">]))</span> \
            <span class="o">.</span><span class="n">keys</span><span class="p">()</span>
</div>
    <span class="k">def</span> <span class="nf">_reserialize</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">serializer</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
        <span class="n">serializer</span> <span class="o">=</span> <span class="n">serializer</span> <span class="ow">or</span> <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="o">.</span><span class="n">serializer</span>
        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_jrdd_deserializer</span> <span class="o">!=</span> <span class="n">serializer</span><span class="p">:</span>
            <span class="bp">self</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">map</span><span class="p">(</span><span class="k">lambda</span> <span class="n">x</span><span class="p">:</span> <span class="n">x</span><span class="p">,</span> <span class="n">preservesPartitioning</span><span class="o">=</span><span class="bp">True</span><span class="p">)</span>
            <span class="bp">self</span><span class="o">.</span><span class="n">_jrdd_deserializer</span> <span class="o">=</span> <span class="n">serializer</span>
        <span class="k">return</span> <span class="bp">self</span>

    <span class="k">def</span> <span class="nf">__add__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">other</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Return the union of this RDD and another one.</span>

<span class="sd">        &gt;&gt;&gt; rdd = sc.parallelize([1, 1, 2, 3])</span>
<span class="sd">        &gt;&gt;&gt; (rdd + rdd).collect()</span>
<span class="sd">        [1, 1, 2, 3, 1, 1, 2, 3]</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">other</span><span class="p">,</span> <span class="n">RDD</span><span class="p">):</span>
            <span class="k">raise</span> <span class="ne">TypeError</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">union</span><span class="p">(</span><span class="n">other</span><span class="p">)</span>

<div class="viewcode-block" id="RDD.repartitionAndSortWithinPartitions"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.repartitionAndSortWithinPartitions">[docs]</a>    <span class="k">def</span> <span class="nf">repartitionAndSortWithinPartitions</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">numPartitions</span><span class="o">=</span><span class="bp">None</span><span class="p">,</span> <span class="n">partitionFunc</span><span class="o">=</span><span class="n">portable_hash</span><span class="p">,</span>
                                           <span class="n">ascending</span><span class="o">=</span><span class="bp">True</span><span class="p">,</span> <span class="n">keyfunc</span><span class="o">=</span><span class="k">lambda</span> <span class="n">x</span><span class="p">:</span> <span class="n">x</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Repartition the RDD according to the given partitioner and, within each resulting partition,</span>
<span class="sd">        sort records by their keys.</span>

<span class="sd">        &gt;&gt;&gt; rdd = sc.parallelize([(0, 5), (3, 8), (2, 6), (0, 8), (3, 8), (1, 3)])</span>
<span class="sd">        &gt;&gt;&gt; rdd2 = rdd.repartitionAndSortWithinPartitions(2, lambda x: x % 2, 2)</span>
<span class="sd">        &gt;&gt;&gt; rdd2.glom().collect()</span>
<span class="sd">        [[(0, 5), (0, 8), (2, 6)], [(1, 3), (3, 8), (3, 8)]]</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">if</span> <span class="n">numPartitions</span> <span class="ow">is</span> <span class="bp">None</span><span class="p">:</span>
            <span class="n">numPartitions</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_defaultReducePartitions</span><span class="p">()</span>

        <span class="n">spill</span> <span class="o">=</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="o">.</span><span class="n">_conf</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s">&quot;spark.shuffle.spill&quot;</span><span class="p">,</span> <span class="s">&#39;True&#39;</span><span class="p">)</span><span class="o">.</span><span class="n">lower</span><span class="p">()</span> <span class="o">==</span> <span class="s">&quot;true&quot;</span><span class="p">)</span>
        <span class="n">memory</span> <span class="o">=</span> <span class="n">_parse_memory</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="o">.</span><span class="n">_conf</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s">&quot;spark.python.worker.memory&quot;</span><span class="p">,</span> <span class="s">&quot;512m&quot;</span><span class="p">))</span>
        <span class="n">serializer</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_jrdd_deserializer</span>

        <span class="k">def</span> <span class="nf">sortPartition</span><span class="p">(</span><span class="n">iterator</span><span class="p">):</span>
            <span class="n">sort</span> <span class="o">=</span> <span class="n">ExternalSorter</span><span class="p">(</span><span class="n">memory</span> <span class="o">*</span> <span class="mf">0.9</span><span class="p">,</span> <span class="n">serializer</span><span class="p">)</span><span class="o">.</span><span class="n">sorted</span> <span class="k">if</span> <span class="n">spill</span> <span class="k">else</span> <span class="nb">sorted</span>
            <span class="k">return</span> <span class="nb">iter</span><span class="p">(</span><span class="n">sort</span><span class="p">(</span><span class="n">iterator</span><span class="p">,</span> <span class="n">key</span><span class="o">=</span><span class="k">lambda</span> <span class="n">k_v</span><span class="p">:</span> <span class="n">keyfunc</span><span class="p">(</span><span class="n">k_v</span><span class="p">[</span><span class="mi">0</span><span class="p">]),</span> <span class="n">reverse</span><span class="o">=</span><span class="p">(</span><span class="ow">not</span> <span class="n">ascending</span><span class="p">)))</span>

        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">partitionBy</span><span class="p">(</span><span class="n">numPartitions</span><span class="p">,</span> <span class="n">partitionFunc</span><span class="p">)</span><span class="o">.</span><span class="n">mapPartitions</span><span class="p">(</span><span class="n">sortPartition</span><span class="p">,</span> <span class="bp">True</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.sortByKey"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.sortByKey">[docs]</a>    <span class="k">def</span> <span class="nf">sortByKey</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">ascending</span><span class="o">=</span><span class="bp">True</span><span class="p">,</span> <span class="n">numPartitions</span><span class="o">=</span><span class="bp">None</span><span class="p">,</span> <span class="n">keyfunc</span><span class="o">=</span><span class="k">lambda</span> <span class="n">x</span><span class="p">:</span> <span class="n">x</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Sorts this RDD, which is assumed to consist of (key, value) pairs.</span>
<span class="sd">        # noqa</span>

<span class="sd">        &gt;&gt;&gt; tmp = [(&#39;a&#39;, 1), (&#39;b&#39;, 2), (&#39;1&#39;, 3), (&#39;d&#39;, 4), (&#39;2&#39;, 5)]</span>
<span class="sd">        &gt;&gt;&gt; sc.parallelize(tmp).sortByKey().first()</span>
<span class="sd">        (&#39;1&#39;, 3)</span>
<span class="sd">        &gt;&gt;&gt; sc.parallelize(tmp).sortByKey(True, 1).collect()</span>
<span class="sd">        [(&#39;1&#39;, 3), (&#39;2&#39;, 5), (&#39;a&#39;, 1), (&#39;b&#39;, 2), (&#39;d&#39;, 4)]</span>
<span class="sd">        &gt;&gt;&gt; sc.parallelize(tmp).sortByKey(True, 2).collect()</span>
<span class="sd">        [(&#39;1&#39;, 3), (&#39;2&#39;, 5), (&#39;a&#39;, 1), (&#39;b&#39;, 2), (&#39;d&#39;, 4)]</span>
<span class="sd">        &gt;&gt;&gt; tmp2 = [(&#39;Mary&#39;, 1), (&#39;had&#39;, 2), (&#39;a&#39;, 3), (&#39;little&#39;, 4), (&#39;lamb&#39;, 5)]</span>
<span class="sd">        &gt;&gt;&gt; tmp2.extend([(&#39;whose&#39;, 6), (&#39;fleece&#39;, 7), (&#39;was&#39;, 8), (&#39;white&#39;, 9)])</span>
<span class="sd">        &gt;&gt;&gt; sc.parallelize(tmp2).sortByKey(True, 3, keyfunc=lambda k: k.lower()).collect()</span>
<span class="sd">        [(&#39;a&#39;, 3), (&#39;fleece&#39;, 7), (&#39;had&#39;, 2), (&#39;lamb&#39;, 5),...(&#39;white&#39;, 9), (&#39;whose&#39;, 6)]</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">if</span> <span class="n">numPartitions</span> <span class="ow">is</span> <span class="bp">None</span><span class="p">:</span>
            <span class="n">numPartitions</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_defaultReducePartitions</span><span class="p">()</span>

        <span class="n">spill</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_can_spill</span><span class="p">()</span>
        <span class="n">memory</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_memory_limit</span><span class="p">()</span>
        <span class="n">serializer</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_jrdd_deserializer</span>

        <span class="k">def</span> <span class="nf">sortPartition</span><span class="p">(</span><span class="n">iterator</span><span class="p">):</span>
            <span class="n">sort</span> <span class="o">=</span> <span class="n">ExternalSorter</span><span class="p">(</span><span class="n">memory</span> <span class="o">*</span> <span class="mf">0.9</span><span class="p">,</span> <span class="n">serializer</span><span class="p">)</span><span class="o">.</span><span class="n">sorted</span> <span class="k">if</span> <span class="n">spill</span> <span class="k">else</span> <span class="nb">sorted</span>
            <span class="k">return</span> <span class="nb">iter</span><span class="p">(</span><span class="n">sort</span><span class="p">(</span><span class="n">iterator</span><span class="p">,</span> <span class="n">key</span><span class="o">=</span><span class="k">lambda</span> <span class="n">kv</span><span class="p">:</span> <span class="n">keyfunc</span><span class="p">(</span><span class="n">kv</span><span class="p">[</span><span class="mi">0</span><span class="p">]),</span> <span class="n">reverse</span><span class="o">=</span><span class="p">(</span><span class="ow">not</span> <span class="n">ascending</span><span class="p">)))</span>

        <span class="k">if</span> <span class="n">numPartitions</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">getNumPartitions</span><span class="p">()</span> <span class="o">&gt;</span> <span class="mi">1</span><span class="p">:</span>
                <span class="bp">self</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">coalesce</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span>
            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">mapPartitions</span><span class="p">(</span><span class="n">sortPartition</span><span class="p">,</span> <span class="bp">True</span><span class="p">)</span>

        <span class="c"># first compute the boundary of each part via sampling: we want to partition</span>
        <span class="c"># the key-space into bins such that the bins have roughly the same</span>
        <span class="c"># number of (key, value) pairs falling into them</span>
        <span class="n">rddSize</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">count</span><span class="p">()</span>
        <span class="k">if</span> <span class="ow">not</span> <span class="n">rddSize</span><span class="p">:</span>
            <span class="k">return</span> <span class="bp">self</span>  <span class="c"># empty RDD</span>
        <span class="n">maxSampleSize</span> <span class="o">=</span> <span class="n">numPartitions</span> <span class="o">*</span> <span class="mf">20.0</span>  <span class="c"># constant from Spark&#39;s RangePartitioner</span>
        <span class="n">fraction</span> <span class="o">=</span> <span class="nb">min</span><span class="p">(</span><span class="n">maxSampleSize</span> <span class="o">/</span> <span class="nb">max</span><span class="p">(</span><span class="n">rddSize</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span> <span class="mf">1.0</span><span class="p">)</span>
        <span class="n">samples</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">sample</span><span class="p">(</span><span class="bp">False</span><span class="p">,</span> <span class="n">fraction</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">map</span><span class="p">(</span><span class="k">lambda</span> <span class="n">kv</span><span class="p">:</span> <span class="n">kv</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span><span class="o">.</span><span class="n">collect</span><span class="p">()</span>
        <span class="n">samples</span> <span class="o">=</span> <span class="nb">sorted</span><span class="p">(</span><span class="n">samples</span><span class="p">,</span> <span class="n">key</span><span class="o">=</span><span class="n">keyfunc</span><span class="p">)</span>

        <span class="c"># we have numPartitions many parts but one of the them has</span>
        <span class="c"># an implicit boundary</span>
        <span class="n">bounds</span> <span class="o">=</span> <span class="p">[</span><span class="n">samples</span><span class="p">[</span><span class="nb">int</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">samples</span><span class="p">)</span> <span class="o">*</span> <span class="p">(</span><span class="n">i</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)</span> <span class="o">/</span> <span class="n">numPartitions</span><span class="p">)]</span>
                  <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">numPartitions</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)]</span>

        <span class="k">def</span> <span class="nf">rangePartitioner</span><span class="p">(</span><span class="n">k</span><span class="p">):</span>
            <span class="n">p</span> <span class="o">=</span> <span class="n">bisect</span><span class="o">.</span><span class="n">bisect_left</span><span class="p">(</span><span class="n">bounds</span><span class="p">,</span> <span class="n">keyfunc</span><span class="p">(</span><span class="n">k</span><span class="p">))</span>
            <span class="k">if</span> <span class="n">ascending</span><span class="p">:</span>
                <span class="k">return</span> <span class="n">p</span>
            <span class="k">else</span><span class="p">:</span>
                <span class="k">return</span> <span class="n">numPartitions</span> <span class="o">-</span> <span class="mi">1</span> <span class="o">-</span> <span class="n">p</span>

        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">partitionBy</span><span class="p">(</span><span class="n">numPartitions</span><span class="p">,</span> <span class="n">rangePartitioner</span><span class="p">)</span><span class="o">.</span><span class="n">mapPartitions</span><span class="p">(</span><span class="n">sortPartition</span><span class="p">,</span> <span class="bp">True</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.sortBy"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.sortBy">[docs]</a>    <span class="k">def</span> <span class="nf">sortBy</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">keyfunc</span><span class="p">,</span> <span class="n">ascending</span><span class="o">=</span><span class="bp">True</span><span class="p">,</span> <span class="n">numPartitions</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Sorts this RDD by the given keyfunc</span>

<span class="sd">        &gt;&gt;&gt; tmp = [(&#39;a&#39;, 1), (&#39;b&#39;, 2), (&#39;1&#39;, 3), (&#39;d&#39;, 4), (&#39;2&#39;, 5)]</span>
<span class="sd">        &gt;&gt;&gt; sc.parallelize(tmp).sortBy(lambda x: x[0]).collect()</span>
<span class="sd">        [(&#39;1&#39;, 3), (&#39;2&#39;, 5), (&#39;a&#39;, 1), (&#39;b&#39;, 2), (&#39;d&#39;, 4)]</span>
<span class="sd">        &gt;&gt;&gt; sc.parallelize(tmp).sortBy(lambda x: x[1]).collect()</span>
<span class="sd">        [(&#39;a&#39;, 1), (&#39;b&#39;, 2), (&#39;1&#39;, 3), (&#39;d&#39;, 4), (&#39;2&#39;, 5)]</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">keyBy</span><span class="p">(</span><span class="n">keyfunc</span><span class="p">)</span><span class="o">.</span><span class="n">sortByKey</span><span class="p">(</span><span class="n">ascending</span><span class="p">,</span> <span class="n">numPartitions</span><span class="p">)</span><span class="o">.</span><span class="n">values</span><span class="p">()</span>
</div>
<div class="viewcode-block" id="RDD.glom"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.glom">[docs]</a>    <span class="k">def</span> <span class="nf">glom</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Return an RDD created by coalescing all elements within each partition</span>
<span class="sd">        into a list.</span>

<span class="sd">        &gt;&gt;&gt; rdd = sc.parallelize([1, 2, 3, 4], 2)</span>
<span class="sd">        &gt;&gt;&gt; sorted(rdd.glom().collect())</span>
<span class="sd">        [[1, 2], [3, 4]]</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">def</span> <span class="nf">func</span><span class="p">(</span><span class="n">iterator</span><span class="p">):</span>
            <span class="k">yield</span> <span class="nb">list</span><span class="p">(</span><span class="n">iterator</span><span class="p">)</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">mapPartitions</span><span class="p">(</span><span class="n">func</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.cartesian"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.cartesian">[docs]</a>    <span class="k">def</span> <span class="nf">cartesian</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">other</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Return the Cartesian product of this RDD and another one, that is, the</span>
<span class="sd">        RDD of all pairs of elements C{(a, b)} where C{a} is in C{self} and</span>
<span class="sd">        C{b} is in C{other}.</span>

<span class="sd">        &gt;&gt;&gt; rdd = sc.parallelize([1, 2])</span>
<span class="sd">        &gt;&gt;&gt; sorted(rdd.cartesian(rdd).collect())</span>
<span class="sd">        [(1, 1), (1, 2), (2, 1), (2, 2)]</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="c"># Due to batching, we can&#39;t use the Java cartesian method.</span>
        <span class="n">deserializer</span> <span class="o">=</span> <span class="n">CartesianDeserializer</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_jrdd_deserializer</span><span class="p">,</span>
                                             <span class="n">other</span><span class="o">.</span><span class="n">_jrdd_deserializer</span><span class="p">)</span>
        <span class="k">return</span> <span class="n">RDD</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_jrdd</span><span class="o">.</span><span class="n">cartesian</span><span class="p">(</span><span class="n">other</span><span class="o">.</span><span class="n">_jrdd</span><span class="p">),</span> <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="p">,</span> <span class="n">deserializer</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.groupBy"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.groupBy">[docs]</a>    <span class="k">def</span> <span class="nf">groupBy</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">f</span><span class="p">,</span> <span class="n">numPartitions</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Return an RDD of grouped items.</span>

<span class="sd">        &gt;&gt;&gt; rdd = sc.parallelize([1, 1, 2, 3, 5, 8])</span>
<span class="sd">        &gt;&gt;&gt; result = rdd.groupBy(lambda x: x % 2).collect()</span>
<span class="sd">        &gt;&gt;&gt; sorted([(x, sorted(y)) for (x, y) in result])</span>
<span class="sd">        [(0, [2, 8]), (1, [1, 1, 3, 5])]</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">map</span><span class="p">(</span><span class="k">lambda</span> <span class="n">x</span><span class="p">:</span> <span class="p">(</span><span class="n">f</span><span class="p">(</span><span class="n">x</span><span class="p">),</span> <span class="n">x</span><span class="p">))</span><span class="o">.</span><span class="n">groupByKey</span><span class="p">(</span><span class="n">numPartitions</span><span class="p">)</span>
</div>
    <span class="nd">@ignore_unicode_prefix</span>
<div class="viewcode-block" id="RDD.pipe"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.pipe">[docs]</a>    <span class="k">def</span> <span class="nf">pipe</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">command</span><span class="p">,</span> <span class="n">env</span><span class="o">=</span><span class="bp">None</span><span class="p">,</span> <span class="n">checkCode</span><span class="o">=</span><span class="bp">False</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Return an RDD created by piping elements to a forked external process.</span>

<span class="sd">        &gt;&gt;&gt; sc.parallelize([&#39;1&#39;, &#39;2&#39;, &#39;&#39;, &#39;3&#39;]).pipe(&#39;cat&#39;).collect()</span>
<span class="sd">        [u&#39;1&#39;, u&#39;2&#39;, u&#39;&#39;, u&#39;3&#39;]</span>

<span class="sd">        :param checkCode: whether or not to check the return value of the shell command.</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">if</span> <span class="n">env</span> <span class="ow">is</span> <span class="bp">None</span><span class="p">:</span>
            <span class="n">env</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">()</span>

        <span class="k">def</span> <span class="nf">func</span><span class="p">(</span><span class="n">iterator</span><span class="p">):</span>
            <span class="n">pipe</span> <span class="o">=</span> <span class="n">Popen</span><span class="p">(</span>
                <span class="n">shlex</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="n">command</span><span class="p">),</span> <span class="n">env</span><span class="o">=</span><span class="n">env</span><span class="p">,</span> <span class="n">stdin</span><span class="o">=</span><span class="n">PIPE</span><span class="p">,</span> <span class="n">stdout</span><span class="o">=</span><span class="n">PIPE</span><span class="p">)</span>

            <span class="k">def</span> <span class="nf">pipe_objs</span><span class="p">(</span><span class="n">out</span><span class="p">):</span>
                <span class="k">for</span> <span class="n">obj</span> <span class="ow">in</span> <span class="n">iterator</span><span class="p">:</span>
                    <span class="n">s</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">obj</span><span class="p">)</span><span class="o">.</span><span class="n">rstrip</span><span class="p">(</span><span class="s">&#39;</span><span class="se">\n</span><span class="s">&#39;</span><span class="p">)</span> <span class="o">+</span> <span class="s">&#39;</span><span class="se">\n</span><span class="s">&#39;</span>
                    <span class="n">out</span><span class="o">.</span><span class="n">write</span><span class="p">(</span><span class="n">s</span><span class="o">.</span><span class="n">encode</span><span class="p">(</span><span class="s">&#39;utf-8&#39;</span><span class="p">))</span>
                <span class="n">out</span><span class="o">.</span><span class="n">close</span><span class="p">()</span>
            <span class="n">Thread</span><span class="p">(</span><span class="n">target</span><span class="o">=</span><span class="n">pipe_objs</span><span class="p">,</span> <span class="n">args</span><span class="o">=</span><span class="p">[</span><span class="n">pipe</span><span class="o">.</span><span class="n">stdin</span><span class="p">])</span><span class="o">.</span><span class="n">start</span><span class="p">()</span>

            <span class="k">def</span> <span class="nf">check_return_code</span><span class="p">():</span>
                <span class="n">pipe</span><span class="o">.</span><span class="n">wait</span><span class="p">()</span>
                <span class="k">if</span> <span class="n">checkCode</span> <span class="ow">and</span> <span class="n">pipe</span><span class="o">.</span><span class="n">returncode</span><span class="p">:</span>
                    <span class="k">raise</span> <span class="ne">Exception</span><span class="p">(</span><span class="s">&quot;Pipe function `</span><span class="si">%s</span><span class="s">&#39; exited &quot;</span>
                                    <span class="s">&quot;with error code </span><span class="si">%d</span><span class="s">&quot;</span> <span class="o">%</span> <span class="p">(</span><span class="n">command</span><span class="p">,</span> <span class="n">pipe</span><span class="o">.</span><span class="n">returncode</span><span class="p">))</span>
                <span class="k">else</span><span class="p">:</span>
                    <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">0</span><span class="p">):</span>
                        <span class="k">yield</span> <span class="n">i</span>
            <span class="k">return</span> <span class="p">(</span><span class="n">x</span><span class="o">.</span><span class="n">rstrip</span><span class="p">(</span><span class="n">b</span><span class="s">&#39;</span><span class="se">\n</span><span class="s">&#39;</span><span class="p">)</span><span class="o">.</span><span class="n">decode</span><span class="p">(</span><span class="s">&#39;utf-8&#39;</span><span class="p">)</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span>
                    <span class="n">chain</span><span class="p">(</span><span class="nb">iter</span><span class="p">(</span><span class="n">pipe</span><span class="o">.</span><span class="n">stdout</span><span class="o">.</span><span class="n">readline</span><span class="p">,</span> <span class="n">b</span><span class="s">&#39;&#39;</span><span class="p">),</span> <span class="n">check_return_code</span><span class="p">()))</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">mapPartitions</span><span class="p">(</span><span class="n">func</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.foreach"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.foreach">[docs]</a>    <span class="k">def</span> <span class="nf">foreach</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">f</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Applies a function to all elements of this RDD.</span>

<span class="sd">        &gt;&gt;&gt; def f(x): print(x)</span>
<span class="sd">        &gt;&gt;&gt; sc.parallelize([1, 2, 3, 4, 5]).foreach(f)</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">def</span> <span class="nf">processPartition</span><span class="p">(</span><span class="n">iterator</span><span class="p">):</span>
            <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">iterator</span><span class="p">:</span>
                <span class="n">f</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
            <span class="k">return</span> <span class="nb">iter</span><span class="p">([])</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">mapPartitions</span><span class="p">(</span><span class="n">processPartition</span><span class="p">)</span><span class="o">.</span><span class="n">count</span><span class="p">()</span>  <span class="c"># Force evaluation</span>
</div>
<div class="viewcode-block" id="RDD.foreachPartition"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.foreachPartition">[docs]</a>    <span class="k">def</span> <span class="nf">foreachPartition</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">f</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Applies a function to each partition of this RDD.</span>

<span class="sd">        &gt;&gt;&gt; def f(iterator):</span>
<span class="sd">        ...      for x in iterator:</span>
<span class="sd">        ...           print(x)</span>
<span class="sd">        &gt;&gt;&gt; sc.parallelize([1, 2, 3, 4, 5]).foreachPartition(f)</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">def</span> <span class="nf">func</span><span class="p">(</span><span class="n">it</span><span class="p">):</span>
            <span class="n">r</span> <span class="o">=</span> <span class="n">f</span><span class="p">(</span><span class="n">it</span><span class="p">)</span>
            <span class="k">try</span><span class="p">:</span>
                <span class="k">return</span> <span class="nb">iter</span><span class="p">(</span><span class="n">r</span><span class="p">)</span>
            <span class="k">except</span> <span class="ne">TypeError</span><span class="p">:</span>
                <span class="k">return</span> <span class="nb">iter</span><span class="p">([])</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">mapPartitions</span><span class="p">(</span><span class="n">func</span><span class="p">)</span><span class="o">.</span><span class="n">count</span><span class="p">()</span>  <span class="c"># Force evaluation</span>
</div>
<div class="viewcode-block" id="RDD.collect"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.collect">[docs]</a>    <span class="k">def</span> <span class="nf">collect</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Return a list that contains all of the elements in this RDD.</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">with</span> <span class="n">SCCallSiteSync</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">context</span><span class="p">)</span> <span class="k">as</span> <span class="n">css</span><span class="p">:</span>
            <span class="n">port</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="o">.</span><span class="n">_jvm</span><span class="o">.</span><span class="n">PythonRDD</span><span class="o">.</span><span class="n">collectAndServe</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_jrdd</span><span class="o">.</span><span class="n">rdd</span><span class="p">())</span>
        <span class="k">return</span> <span class="nb">list</span><span class="p">(</span><span class="n">_load_from_socket</span><span class="p">(</span><span class="n">port</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_jrdd_deserializer</span><span class="p">))</span>
</div>
<div class="viewcode-block" id="RDD.reduce"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.reduce">[docs]</a>    <span class="k">def</span> <span class="nf">reduce</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">f</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Reduces the elements of this RDD using the specified commutative and</span>
<span class="sd">        associative binary operator. Currently reduces partitions locally.</span>

<span class="sd">        &gt;&gt;&gt; from operator import add</span>
<span class="sd">        &gt;&gt;&gt; sc.parallelize([1, 2, 3, 4, 5]).reduce(add)</span>
<span class="sd">        15</span>
<span class="sd">        &gt;&gt;&gt; sc.parallelize((2 for _ in range(10))).map(lambda x: 1).cache().reduce(add)</span>
<span class="sd">        10</span>
<span class="sd">        &gt;&gt;&gt; sc.parallelize([]).reduce(add)</span>
<span class="sd">        Traceback (most recent call last):</span>
<span class="sd">            ...</span>
<span class="sd">        ValueError: Can not reduce() empty RDD</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">def</span> <span class="nf">func</span><span class="p">(</span><span class="n">iterator</span><span class="p">):</span>
            <span class="n">iterator</span> <span class="o">=</span> <span class="nb">iter</span><span class="p">(</span><span class="n">iterator</span><span class="p">)</span>
            <span class="k">try</span><span class="p">:</span>
                <span class="n">initial</span> <span class="o">=</span> <span class="nb">next</span><span class="p">(</span><span class="n">iterator</span><span class="p">)</span>
            <span class="k">except</span> <span class="ne">StopIteration</span><span class="p">:</span>
                <span class="k">return</span>
            <span class="k">yield</span> <span class="nb">reduce</span><span class="p">(</span><span class="n">f</span><span class="p">,</span> <span class="n">iterator</span><span class="p">,</span> <span class="n">initial</span><span class="p">)</span>

        <span class="n">vals</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">mapPartitions</span><span class="p">(</span><span class="n">func</span><span class="p">)</span><span class="o">.</span><span class="n">collect</span><span class="p">()</span>
        <span class="k">if</span> <span class="n">vals</span><span class="p">:</span>
            <span class="k">return</span> <span class="nb">reduce</span><span class="p">(</span><span class="n">f</span><span class="p">,</span> <span class="n">vals</span><span class="p">)</span>
        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s">&quot;Can not reduce() empty RDD&quot;</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.treeReduce"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.treeReduce">[docs]</a>    <span class="k">def</span> <span class="nf">treeReduce</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">f</span><span class="p">,</span> <span class="n">depth</span><span class="o">=</span><span class="mi">2</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Reduces the elements of this RDD in a multi-level tree pattern.</span>

<span class="sd">        :param depth: suggested depth of the tree (default: 2)</span>

<span class="sd">        &gt;&gt;&gt; add = lambda x, y: x + y</span>
<span class="sd">        &gt;&gt;&gt; rdd = sc.parallelize([-5, -4, -3, -2, -1, 1, 2, 3, 4], 10)</span>
<span class="sd">        &gt;&gt;&gt; rdd.treeReduce(add)</span>
<span class="sd">        -5</span>
<span class="sd">        &gt;&gt;&gt; rdd.treeReduce(add, 1)</span>
<span class="sd">        -5</span>
<span class="sd">        &gt;&gt;&gt; rdd.treeReduce(add, 2)</span>
<span class="sd">        -5</span>
<span class="sd">        &gt;&gt;&gt; rdd.treeReduce(add, 5)</span>
<span class="sd">        -5</span>
<span class="sd">        &gt;&gt;&gt; rdd.treeReduce(add, 10)</span>
<span class="sd">        -5</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">if</span> <span class="n">depth</span> <span class="o">&lt;</span> <span class="mi">1</span><span class="p">:</span>
            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s">&quot;Depth cannot be smaller than 1 but got </span><span class="si">%d</span><span class="s">.&quot;</span> <span class="o">%</span> <span class="n">depth</span><span class="p">)</span>

        <span class="n">zeroValue</span> <span class="o">=</span> <span class="bp">None</span><span class="p">,</span> <span class="bp">True</span>  <span class="c"># Use the second entry to indicate whether this is a dummy value.</span>

        <span class="k">def</span> <span class="nf">op</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">y</span><span class="p">):</span>
            <span class="k">if</span> <span class="n">x</span><span class="p">[</span><span class="mi">1</span><span class="p">]:</span>
                <span class="k">return</span> <span class="n">y</span>
            <span class="k">elif</span> <span class="n">y</span><span class="p">[</span><span class="mi">1</span><span class="p">]:</span>
                <span class="k">return</span> <span class="n">x</span>
            <span class="k">else</span><span class="p">:</span>
                <span class="k">return</span> <span class="n">f</span><span class="p">(</span><span class="n">x</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">y</span><span class="p">[</span><span class="mi">0</span><span class="p">]),</span> <span class="bp">False</span>

        <span class="n">reduced</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">map</span><span class="p">(</span><span class="k">lambda</span> <span class="n">x</span><span class="p">:</span> <span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="bp">False</span><span class="p">))</span><span class="o">.</span><span class="n">treeAggregate</span><span class="p">(</span><span class="n">zeroValue</span><span class="p">,</span> <span class="n">op</span><span class="p">,</span> <span class="n">op</span><span class="p">,</span> <span class="n">depth</span><span class="p">)</span>
        <span class="k">if</span> <span class="n">reduced</span><span class="p">[</span><span class="mi">1</span><span class="p">]:</span>
            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s">&quot;Cannot reduce empty RDD.&quot;</span><span class="p">)</span>
        <span class="k">return</span> <span class="n">reduced</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
</div>
<div class="viewcode-block" id="RDD.fold"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.fold">[docs]</a>    <span class="k">def</span> <span class="nf">fold</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">zeroValue</span><span class="p">,</span> <span class="n">op</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Aggregate the elements of each partition, and then the results for all</span>
<span class="sd">        the partitions, using a given associative and commutative function and</span>
<span class="sd">        a neutral &quot;zero value.&quot;</span>

<span class="sd">        The function C{op(t1, t2)} is allowed to modify C{t1} and return it</span>
<span class="sd">        as its result value to avoid object allocation; however, it should not</span>
<span class="sd">        modify C{t2}.</span>

<span class="sd">        This behaves somewhat differently from fold operations implemented</span>
<span class="sd">        for non-distributed collections in functional languages like Scala.</span>
<span class="sd">        This fold operation may be applied to partitions individually, and then</span>
<span class="sd">        fold those results into the final result, rather than apply the fold</span>
<span class="sd">        to each element sequentially in some defined ordering. For functions</span>
<span class="sd">        that are not commutative, the result may differ from that of a fold</span>
<span class="sd">        applied to a non-distributed collection.</span>

<span class="sd">        &gt;&gt;&gt; from operator import add</span>
<span class="sd">        &gt;&gt;&gt; sc.parallelize([1, 2, 3, 4, 5]).fold(0, add)</span>
<span class="sd">        15</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">def</span> <span class="nf">func</span><span class="p">(</span><span class="n">iterator</span><span class="p">):</span>
            <span class="n">acc</span> <span class="o">=</span> <span class="n">zeroValue</span>
            <span class="k">for</span> <span class="n">obj</span> <span class="ow">in</span> <span class="n">iterator</span><span class="p">:</span>
                <span class="n">acc</span> <span class="o">=</span> <span class="n">op</span><span class="p">(</span><span class="n">obj</span><span class="p">,</span> <span class="n">acc</span><span class="p">)</span>
            <span class="k">yield</span> <span class="n">acc</span>
        <span class="c"># collecting result of mapPartitions here ensures that the copy of</span>
        <span class="c"># zeroValue provided to each partition is unique from the one provided</span>
        <span class="c"># to the final reduce call</span>
        <span class="n">vals</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">mapPartitions</span><span class="p">(</span><span class="n">func</span><span class="p">)</span><span class="o">.</span><span class="n">collect</span><span class="p">()</span>
        <span class="k">return</span> <span class="nb">reduce</span><span class="p">(</span><span class="n">op</span><span class="p">,</span> <span class="n">vals</span><span class="p">,</span> <span class="n">zeroValue</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.aggregate"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.aggregate">[docs]</a>    <span class="k">def</span> <span class="nf">aggregate</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">zeroValue</span><span class="p">,</span> <span class="n">seqOp</span><span class="p">,</span> <span class="n">combOp</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Aggregate the elements of each partition, and then the results for all</span>
<span class="sd">        the partitions, using a given combine functions and a neutral &quot;zero</span>
<span class="sd">        value.&quot;</span>

<span class="sd">        The functions C{op(t1, t2)} is allowed to modify C{t1} and return it</span>
<span class="sd">        as its result value to avoid object allocation; however, it should not</span>
<span class="sd">        modify C{t2}.</span>

<span class="sd">        The first function (seqOp) can return a different result type, U, than</span>
<span class="sd">        the type of this RDD. Thus, we need one operation for merging a T into</span>
<span class="sd">        an U and one operation for merging two U</span>

<span class="sd">        &gt;&gt;&gt; seqOp = (lambda x, y: (x[0] + y, x[1] + 1))</span>
<span class="sd">        &gt;&gt;&gt; combOp = (lambda x, y: (x[0] + y[0], x[1] + y[1]))</span>
<span class="sd">        &gt;&gt;&gt; sc.parallelize([1, 2, 3, 4]).aggregate((0, 0), seqOp, combOp)</span>
<span class="sd">        (10, 4)</span>
<span class="sd">        &gt;&gt;&gt; sc.parallelize([]).aggregate((0, 0), seqOp, combOp)</span>
<span class="sd">        (0, 0)</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">def</span> <span class="nf">func</span><span class="p">(</span><span class="n">iterator</span><span class="p">):</span>
            <span class="n">acc</span> <span class="o">=</span> <span class="n">zeroValue</span>
            <span class="k">for</span> <span class="n">obj</span> <span class="ow">in</span> <span class="n">iterator</span><span class="p">:</span>
                <span class="n">acc</span> <span class="o">=</span> <span class="n">seqOp</span><span class="p">(</span><span class="n">acc</span><span class="p">,</span> <span class="n">obj</span><span class="p">)</span>
            <span class="k">yield</span> <span class="n">acc</span>
        <span class="c"># collecting result of mapPartitions here ensures that the copy of</span>
        <span class="c"># zeroValue provided to each partition is unique from the one provided</span>
        <span class="c"># to the final reduce call</span>
        <span class="n">vals</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">mapPartitions</span><span class="p">(</span><span class="n">func</span><span class="p">)</span><span class="o">.</span><span class="n">collect</span><span class="p">()</span>
        <span class="k">return</span> <span class="nb">reduce</span><span class="p">(</span><span class="n">combOp</span><span class="p">,</span> <span class="n">vals</span><span class="p">,</span> <span class="n">zeroValue</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.treeAggregate"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.treeAggregate">[docs]</a>    <span class="k">def</span> <span class="nf">treeAggregate</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">zeroValue</span><span class="p">,</span> <span class="n">seqOp</span><span class="p">,</span> <span class="n">combOp</span><span class="p">,</span> <span class="n">depth</span><span class="o">=</span><span class="mi">2</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Aggregates the elements of this RDD in a multi-level tree</span>
<span class="sd">        pattern.</span>

<span class="sd">        :param depth: suggested depth of the tree (default: 2)</span>

<span class="sd">        &gt;&gt;&gt; add = lambda x, y: x + y</span>
<span class="sd">        &gt;&gt;&gt; rdd = sc.parallelize([-5, -4, -3, -2, -1, 1, 2, 3, 4], 10)</span>
<span class="sd">        &gt;&gt;&gt; rdd.treeAggregate(0, add, add)</span>
<span class="sd">        -5</span>
<span class="sd">        &gt;&gt;&gt; rdd.treeAggregate(0, add, add, 1)</span>
<span class="sd">        -5</span>
<span class="sd">        &gt;&gt;&gt; rdd.treeAggregate(0, add, add, 2)</span>
<span class="sd">        -5</span>
<span class="sd">        &gt;&gt;&gt; rdd.treeAggregate(0, add, add, 5)</span>
<span class="sd">        -5</span>
<span class="sd">        &gt;&gt;&gt; rdd.treeAggregate(0, add, add, 10)</span>
<span class="sd">        -5</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">if</span> <span class="n">depth</span> <span class="o">&lt;</span> <span class="mi">1</span><span class="p">:</span>
            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s">&quot;Depth cannot be smaller than 1 but got </span><span class="si">%d</span><span class="s">.&quot;</span> <span class="o">%</span> <span class="n">depth</span><span class="p">)</span>

        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">getNumPartitions</span><span class="p">()</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
            <span class="k">return</span> <span class="n">zeroValue</span>

        <span class="k">def</span> <span class="nf">aggregatePartition</span><span class="p">(</span><span class="n">iterator</span><span class="p">):</span>
            <span class="n">acc</span> <span class="o">=</span> <span class="n">zeroValue</span>
            <span class="k">for</span> <span class="n">obj</span> <span class="ow">in</span> <span class="n">iterator</span><span class="p">:</span>
                <span class="n">acc</span> <span class="o">=</span> <span class="n">seqOp</span><span class="p">(</span><span class="n">acc</span><span class="p">,</span> <span class="n">obj</span><span class="p">)</span>
            <span class="k">yield</span> <span class="n">acc</span>

        <span class="n">partiallyAggregated</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">mapPartitions</span><span class="p">(</span><span class="n">aggregatePartition</span><span class="p">)</span>
        <span class="n">numPartitions</span> <span class="o">=</span> <span class="n">partiallyAggregated</span><span class="o">.</span><span class="n">getNumPartitions</span><span class="p">()</span>
        <span class="n">scale</span> <span class="o">=</span> <span class="nb">max</span><span class="p">(</span><span class="nb">int</span><span class="p">(</span><span class="n">ceil</span><span class="p">(</span><span class="nb">pow</span><span class="p">(</span><span class="n">numPartitions</span><span class="p">,</span> <span class="mf">1.0</span> <span class="o">/</span> <span class="n">depth</span><span class="p">))),</span> <span class="mi">2</span><span class="p">)</span>
        <span class="c"># If creating an extra level doesn&#39;t help reduce the wall-clock time, we stop the tree</span>
        <span class="c"># aggregation.</span>
        <span class="k">while</span> <span class="n">numPartitions</span> <span class="o">&gt;</span> <span class="n">scale</span> <span class="o">+</span> <span class="n">numPartitions</span> <span class="o">/</span> <span class="n">scale</span><span class="p">:</span>
            <span class="n">numPartitions</span> <span class="o">/=</span> <span class="n">scale</span>
            <span class="n">curNumPartitions</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">numPartitions</span><span class="p">)</span>

            <span class="k">def</span> <span class="nf">mapPartition</span><span class="p">(</span><span class="n">i</span><span class="p">,</span> <span class="n">iterator</span><span class="p">):</span>
                <span class="k">for</span> <span class="n">obj</span> <span class="ow">in</span> <span class="n">iterator</span><span class="p">:</span>
                    <span class="k">yield</span> <span class="p">(</span><span class="n">i</span> <span class="o">%</span> <span class="n">curNumPartitions</span><span class="p">,</span> <span class="n">obj</span><span class="p">)</span>

            <span class="n">partiallyAggregated</span> <span class="o">=</span> <span class="n">partiallyAggregated</span> \
                <span class="o">.</span><span class="n">mapPartitionsWithIndex</span><span class="p">(</span><span class="n">mapPartition</span><span class="p">)</span> \
                <span class="o">.</span><span class="n">reduceByKey</span><span class="p">(</span><span class="n">combOp</span><span class="p">,</span> <span class="n">curNumPartitions</span><span class="p">)</span> \
                <span class="o">.</span><span class="n">values</span><span class="p">()</span>

        <span class="k">return</span> <span class="n">partiallyAggregated</span><span class="o">.</span><span class="n">reduce</span><span class="p">(</span><span class="n">combOp</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.max"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.max">[docs]</a>    <span class="k">def</span> <span class="nf">max</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">key</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Find the maximum item in this RDD.</span>

<span class="sd">        :param key: A function used to generate key for comparing</span>

<span class="sd">        &gt;&gt;&gt; rdd = sc.parallelize([1.0, 5.0, 43.0, 10.0])</span>
<span class="sd">        &gt;&gt;&gt; rdd.max()</span>
<span class="sd">        43.0</span>
<span class="sd">        &gt;&gt;&gt; rdd.max(key=str)</span>
<span class="sd">        5.0</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">if</span> <span class="n">key</span> <span class="ow">is</span> <span class="bp">None</span><span class="p">:</span>
            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">reduce</span><span class="p">(</span><span class="nb">max</span><span class="p">)</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">reduce</span><span class="p">(</span><span class="k">lambda</span> <span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">:</span> <span class="nb">max</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">,</span> <span class="n">key</span><span class="o">=</span><span class="n">key</span><span class="p">))</span>
</div>
<div class="viewcode-block" id="RDD.min"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.min">[docs]</a>    <span class="k">def</span> <span class="nf">min</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">key</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Find the minimum item in this RDD.</span>

<span class="sd">        :param key: A function used to generate key for comparing</span>

<span class="sd">        &gt;&gt;&gt; rdd = sc.parallelize([2.0, 5.0, 43.0, 10.0])</span>
<span class="sd">        &gt;&gt;&gt; rdd.min()</span>
<span class="sd">        2.0</span>
<span class="sd">        &gt;&gt;&gt; rdd.min(key=str)</span>
<span class="sd">        10.0</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">if</span> <span class="n">key</span> <span class="ow">is</span> <span class="bp">None</span><span class="p">:</span>
            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">reduce</span><span class="p">(</span><span class="nb">min</span><span class="p">)</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">reduce</span><span class="p">(</span><span class="k">lambda</span> <span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">:</span> <span class="nb">min</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">,</span> <span class="n">key</span><span class="o">=</span><span class="n">key</span><span class="p">))</span>
</div>
<div class="viewcode-block" id="RDD.sum"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.sum">[docs]</a>    <span class="k">def</span> <span class="nf">sum</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Add up the elements in this RDD.</span>

<span class="sd">        &gt;&gt;&gt; sc.parallelize([1.0, 2.0, 3.0]).sum()</span>
<span class="sd">        6.0</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">mapPartitions</span><span class="p">(</span><span class="k">lambda</span> <span class="n">x</span><span class="p">:</span> <span class="p">[</span><span class="nb">sum</span><span class="p">(</span><span class="n">x</span><span class="p">)])</span><span class="o">.</span><span class="n">fold</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">operator</span><span class="o">.</span><span class="n">add</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.count"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.count">[docs]</a>    <span class="k">def</span> <span class="nf">count</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Return the number of elements in this RDD.</span>

<span class="sd">        &gt;&gt;&gt; sc.parallelize([2, 3, 4]).count()</span>
<span class="sd">        3</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">mapPartitions</span><span class="p">(</span><span class="k">lambda</span> <span class="n">i</span><span class="p">:</span> <span class="p">[</span><span class="nb">sum</span><span class="p">(</span><span class="mi">1</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="n">i</span><span class="p">)])</span><span class="o">.</span><span class="n">sum</span><span class="p">()</span>
</div>
<div class="viewcode-block" id="RDD.stats"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.stats">[docs]</a>    <span class="k">def</span> <span class="nf">stats</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Return a L{StatCounter} object that captures the mean, variance</span>
<span class="sd">        and count of the RDD&#39;s elements in one operation.</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">def</span> <span class="nf">redFunc</span><span class="p">(</span><span class="n">left_counter</span><span class="p">,</span> <span class="n">right_counter</span><span class="p">):</span>
            <span class="k">return</span> <span class="n">left_counter</span><span class="o">.</span><span class="n">mergeStats</span><span class="p">(</span><span class="n">right_counter</span><span class="p">)</span>

        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">mapPartitions</span><span class="p">(</span><span class="k">lambda</span> <span class="n">i</span><span class="p">:</span> <span class="p">[</span><span class="n">StatCounter</span><span class="p">(</span><span class="n">i</span><span class="p">)])</span><span class="o">.</span><span class="n">reduce</span><span class="p">(</span><span class="n">redFunc</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.histogram"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.histogram">[docs]</a>    <span class="k">def</span> <span class="nf">histogram</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">buckets</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Compute a histogram using the provided buckets. The buckets</span>
<span class="sd">        are all open to the right except for the last which is closed.</span>
<span class="sd">        e.g. [1,10,20,50] means the buckets are [1,10) [10,20) [20,50],</span>
<span class="sd">        which means 1&lt;=x&lt;10, 10&lt;=x&lt;20, 20&lt;=x&lt;=50. And on the input of 1</span>
<span class="sd">        and 50 we would have a histogram of 1,0,1.</span>

<span class="sd">        If your histogram is evenly spaced (e.g. [0, 10, 20, 30]),</span>
<span class="sd">        this can be switched from an O(log n) inseration to O(1) per</span>
<span class="sd">        element(where n = # buckets).</span>

<span class="sd">        Buckets must be sorted and not contain any duplicates, must be</span>
<span class="sd">        at least two elements.</span>

<span class="sd">        If `buckets` is a number, it will generates buckets which are</span>
<span class="sd">        evenly spaced between the minimum and maximum of the RDD. For</span>
<span class="sd">        example, if the min value is 0 and the max is 100, given buckets</span>
<span class="sd">        as 2, the resulting buckets will be [0,50) [50,100]. buckets must</span>
<span class="sd">        be at least 1 If the RDD contains infinity, NaN throws an exception</span>
<span class="sd">        If the elements in RDD do not vary (max == min) always returns</span>
<span class="sd">        a single bucket.</span>

<span class="sd">        It will return an tuple of buckets and histogram.</span>

<span class="sd">        &gt;&gt;&gt; rdd = sc.parallelize(range(51))</span>
<span class="sd">        &gt;&gt;&gt; rdd.histogram(2)</span>
<span class="sd">        ([0, 25, 50], [25, 26])</span>
<span class="sd">        &gt;&gt;&gt; rdd.histogram([0, 5, 25, 50])</span>
<span class="sd">        ([0, 5, 25, 50], [5, 20, 26])</span>
<span class="sd">        &gt;&gt;&gt; rdd.histogram([0, 15, 30, 45, 60])  # evenly spaced buckets</span>
<span class="sd">        ([0, 15, 30, 45, 60], [15, 15, 15, 6])</span>
<span class="sd">        &gt;&gt;&gt; rdd = sc.parallelize([&quot;ab&quot;, &quot;ac&quot;, &quot;b&quot;, &quot;bd&quot;, &quot;ef&quot;])</span>
<span class="sd">        &gt;&gt;&gt; rdd.histogram((&quot;a&quot;, &quot;b&quot;, &quot;c&quot;))</span>
<span class="sd">        ((&#39;a&#39;, &#39;b&#39;, &#39;c&#39;), [2, 2])</span>
<span class="sd">        &quot;&quot;&quot;</span>

        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">buckets</span><span class="p">,</span> <span class="nb">int</span><span class="p">):</span>
            <span class="k">if</span> <span class="n">buckets</span> <span class="o">&lt;</span> <span class="mi">1</span><span class="p">:</span>
                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s">&quot;number of buckets must be &gt;= 1&quot;</span><span class="p">)</span>

            <span class="c"># filter out non-comparable elements</span>
            <span class="k">def</span> <span class="nf">comparable</span><span class="p">(</span><span class="n">x</span><span class="p">):</span>
                <span class="k">if</span> <span class="n">x</span> <span class="ow">is</span> <span class="bp">None</span><span class="p">:</span>
                    <span class="k">return</span> <span class="bp">False</span>
                <span class="k">if</span> <span class="nb">type</span><span class="p">(</span><span class="n">x</span><span class="p">)</span> <span class="ow">is</span> <span class="nb">float</span> <span class="ow">and</span> <span class="n">isnan</span><span class="p">(</span><span class="n">x</span><span class="p">):</span>
                    <span class="k">return</span> <span class="bp">False</span>
                <span class="k">return</span> <span class="bp">True</span>

            <span class="n">filtered</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">filter</span><span class="p">(</span><span class="n">comparable</span><span class="p">)</span>

            <span class="c"># faster than stats()</span>
            <span class="k">def</span> <span class="nf">minmax</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">):</span>
                <span class="k">return</span> <span class="nb">min</span><span class="p">(</span><span class="n">a</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">b</span><span class="p">[</span><span class="mi">0</span><span class="p">]),</span> <span class="nb">max</span><span class="p">(</span><span class="n">a</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">b</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span>
            <span class="k">try</span><span class="p">:</span>
                <span class="n">minv</span><span class="p">,</span> <span class="n">maxv</span> <span class="o">=</span> <span class="n">filtered</span><span class="o">.</span><span class="n">map</span><span class="p">(</span><span class="k">lambda</span> <span class="n">x</span><span class="p">:</span> <span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">x</span><span class="p">))</span><span class="o">.</span><span class="n">reduce</span><span class="p">(</span><span class="n">minmax</span><span class="p">)</span>
            <span class="k">except</span> <span class="ne">TypeError</span> <span class="k">as</span> <span class="n">e</span><span class="p">:</span>
                <span class="k">if</span> <span class="s">&quot; empty &quot;</span> <span class="ow">in</span> <span class="nb">str</span><span class="p">(</span><span class="n">e</span><span class="p">):</span>
                    <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s">&quot;can not generate buckets from empty RDD&quot;</span><span class="p">)</span>
                <span class="k">raise</span>

            <span class="k">if</span> <span class="n">minv</span> <span class="o">==</span> <span class="n">maxv</span> <span class="ow">or</span> <span class="n">buckets</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
                <span class="k">return</span> <span class="p">[</span><span class="n">minv</span><span class="p">,</span> <span class="n">maxv</span><span class="p">],</span> <span class="p">[</span><span class="n">filtered</span><span class="o">.</span><span class="n">count</span><span class="p">()]</span>

            <span class="k">try</span><span class="p">:</span>
                <span class="n">inc</span> <span class="o">=</span> <span class="p">(</span><span class="n">maxv</span> <span class="o">-</span> <span class="n">minv</span><span class="p">)</span> <span class="o">/</span> <span class="n">buckets</span>
            <span class="k">except</span> <span class="ne">TypeError</span><span class="p">:</span>
                <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s">&quot;Can not generate buckets with non-number in RDD&quot;</span><span class="p">)</span>

            <span class="k">if</span> <span class="n">isinf</span><span class="p">(</span><span class="n">inc</span><span class="p">):</span>
                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s">&quot;Can not generate buckets with infinite value&quot;</span><span class="p">)</span>

            <span class="c"># keep them as integer if possible</span>
            <span class="n">inc</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">inc</span><span class="p">)</span>
            <span class="k">if</span> <span class="n">inc</span> <span class="o">*</span> <span class="n">buckets</span> <span class="o">!=</span> <span class="n">maxv</span> <span class="o">-</span> <span class="n">minv</span><span class="p">:</span>
                <span class="n">inc</span> <span class="o">=</span> <span class="p">(</span><span class="n">maxv</span> <span class="o">-</span> <span class="n">minv</span><span class="p">)</span> <span class="o">*</span> <span class="mf">1.0</span> <span class="o">/</span> <span class="n">buckets</span>

            <span class="n">buckets</span> <span class="o">=</span> <span class="p">[</span><span class="n">i</span> <span class="o">*</span> <span class="n">inc</span> <span class="o">+</span> <span class="n">minv</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">buckets</span><span class="p">)]</span>
            <span class="n">buckets</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">maxv</span><span class="p">)</span>  <span class="c"># fix accumulated error</span>
            <span class="n">even</span> <span class="o">=</span> <span class="bp">True</span>

        <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">buckets</span><span class="p">,</span> <span class="p">(</span><span class="nb">list</span><span class="p">,</span> <span class="nb">tuple</span><span class="p">)):</span>
            <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">buckets</span><span class="p">)</span> <span class="o">&lt;</span> <span class="mi">2</span><span class="p">:</span>
                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s">&quot;buckets should have more than one value&quot;</span><span class="p">)</span>

            <span class="k">if</span> <span class="nb">any</span><span class="p">(</span><span class="n">i</span> <span class="ow">is</span> <span class="bp">None</span> <span class="ow">or</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">i</span><span class="p">,</span> <span class="nb">float</span><span class="p">)</span> <span class="ow">and</span> <span class="n">isnan</span><span class="p">(</span><span class="n">i</span><span class="p">)</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="n">buckets</span><span class="p">):</span>
                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s">&quot;can not have None or NaN in buckets&quot;</span><span class="p">)</span>

            <span class="k">if</span> <span class="nb">sorted</span><span class="p">(</span><span class="n">buckets</span><span class="p">)</span> <span class="o">!=</span> <span class="nb">list</span><span class="p">(</span><span class="n">buckets</span><span class="p">):</span>
                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s">&quot;buckets should be sorted&quot;</span><span class="p">)</span>

            <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="nb">set</span><span class="p">(</span><span class="n">buckets</span><span class="p">))</span> <span class="o">!=</span> <span class="nb">len</span><span class="p">(</span><span class="n">buckets</span><span class="p">):</span>
                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s">&quot;buckets should not contain duplicated values&quot;</span><span class="p">)</span>

            <span class="n">minv</span> <span class="o">=</span> <span class="n">buckets</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
            <span class="n">maxv</span> <span class="o">=</span> <span class="n">buckets</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span>
            <span class="n">even</span> <span class="o">=</span> <span class="bp">False</span>
            <span class="n">inc</span> <span class="o">=</span> <span class="bp">None</span>
            <span class="k">try</span><span class="p">:</span>
                <span class="n">steps</span> <span class="o">=</span> <span class="p">[</span><span class="n">buckets</span><span class="p">[</span><span class="n">i</span> <span class="o">+</span> <span class="mi">1</span><span class="p">]</span> <span class="o">-</span> <span class="n">buckets</span><span class="p">[</span><span class="n">i</span><span class="p">]</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">buckets</span><span class="p">)</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)]</span>
            <span class="k">except</span> <span class="ne">TypeError</span><span class="p">:</span>
                <span class="k">pass</span>  <span class="c"># objects in buckets do not support &#39;-&#39;</span>
            <span class="k">else</span><span class="p">:</span>
                <span class="k">if</span> <span class="nb">max</span><span class="p">(</span><span class="n">steps</span><span class="p">)</span> <span class="o">-</span> <span class="nb">min</span><span class="p">(</span><span class="n">steps</span><span class="p">)</span> <span class="o">&lt;</span> <span class="mf">1e-10</span><span class="p">:</span>  <span class="c"># handle precision errors</span>
                    <span class="n">even</span> <span class="o">=</span> <span class="bp">True</span>
                    <span class="n">inc</span> <span class="o">=</span> <span class="p">(</span><span class="n">maxv</span> <span class="o">-</span> <span class="n">minv</span><span class="p">)</span> <span class="o">/</span> <span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">buckets</span><span class="p">)</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span>

        <span class="k">else</span><span class="p">:</span>
            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s">&quot;buckets should be a list or tuple or number(int or long)&quot;</span><span class="p">)</span>

        <span class="k">def</span> <span class="nf">histogram</span><span class="p">(</span><span class="n">iterator</span><span class="p">):</span>
            <span class="n">counters</span> <span class="o">=</span> <span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">*</span> <span class="nb">len</span><span class="p">(</span><span class="n">buckets</span><span class="p">)</span>
            <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="n">iterator</span><span class="p">:</span>
                <span class="k">if</span> <span class="n">i</span> <span class="ow">is</span> <span class="bp">None</span> <span class="ow">or</span> <span class="p">(</span><span class="nb">type</span><span class="p">(</span><span class="n">i</span><span class="p">)</span> <span class="ow">is</span> <span class="nb">float</span> <span class="ow">and</span> <span class="n">isnan</span><span class="p">(</span><span class="n">i</span><span class="p">))</span> <span class="ow">or</span> <span class="n">i</span> <span class="o">&gt;</span> <span class="n">maxv</span> <span class="ow">or</span> <span class="n">i</span> <span class="o">&lt;</span> <span class="n">minv</span><span class="p">:</span>
                    <span class="k">continue</span>
                <span class="n">t</span> <span class="o">=</span> <span class="p">(</span><span class="nb">int</span><span class="p">((</span><span class="n">i</span> <span class="o">-</span> <span class="n">minv</span><span class="p">)</span> <span class="o">/</span> <span class="n">inc</span><span class="p">)</span> <span class="k">if</span> <span class="n">even</span>
                     <span class="k">else</span> <span class="n">bisect</span><span class="o">.</span><span class="n">bisect_right</span><span class="p">(</span><span class="n">buckets</span><span class="p">,</span> <span class="n">i</span><span class="p">)</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span>
                <span class="n">counters</span><span class="p">[</span><span class="n">t</span><span class="p">]</span> <span class="o">+=</span> <span class="mi">1</span>
            <span class="c"># add last two together</span>
            <span class="n">last</span> <span class="o">=</span> <span class="n">counters</span><span class="o">.</span><span class="n">pop</span><span class="p">()</span>
            <span class="n">counters</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span> <span class="o">+=</span> <span class="n">last</span>
            <span class="k">return</span> <span class="p">[</span><span class="n">counters</span><span class="p">]</span>

        <span class="k">def</span> <span class="nf">mergeCounters</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">):</span>
            <span class="k">return</span> <span class="p">[</span><span class="n">i</span> <span class="o">+</span> <span class="n">j</span> <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">j</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">)]</span>

        <span class="k">return</span> <span class="n">buckets</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">mapPartitions</span><span class="p">(</span><span class="n">histogram</span><span class="p">)</span><span class="o">.</span><span class="n">reduce</span><span class="p">(</span><span class="n">mergeCounters</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.mean"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.mean">[docs]</a>    <span class="k">def</span> <span class="nf">mean</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Compute the mean of this RDD&#39;s elements.</span>

<span class="sd">        &gt;&gt;&gt; sc.parallelize([1, 2, 3]).mean()</span>
<span class="sd">        2.0</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">stats</span><span class="p">()</span><span class="o">.</span><span class="n">mean</span><span class="p">()</span>
</div>
<div class="viewcode-block" id="RDD.variance"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.variance">[docs]</a>    <span class="k">def</span> <span class="nf">variance</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Compute the variance of this RDD&#39;s elements.</span>

<span class="sd">        &gt;&gt;&gt; sc.parallelize([1, 2, 3]).variance()</span>
<span class="sd">        0.666...</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">stats</span><span class="p">()</span><span class="o">.</span><span class="n">variance</span><span class="p">()</span>
</div>
<div class="viewcode-block" id="RDD.stdev"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.stdev">[docs]</a>    <span class="k">def</span> <span class="nf">stdev</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Compute the standard deviation of this RDD&#39;s elements.</span>

<span class="sd">        &gt;&gt;&gt; sc.parallelize([1, 2, 3]).stdev()</span>
<span class="sd">        0.816...</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">stats</span><span class="p">()</span><span class="o">.</span><span class="n">stdev</span><span class="p">()</span>
</div>
<div class="viewcode-block" id="RDD.sampleStdev"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.sampleStdev">[docs]</a>    <span class="k">def</span> <span class="nf">sampleStdev</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Compute the sample standard deviation of this RDD&#39;s elements (which</span>
<span class="sd">        corrects for bias in estimating the standard deviation by dividing by</span>
<span class="sd">        N-1 instead of N).</span>

<span class="sd">        &gt;&gt;&gt; sc.parallelize([1, 2, 3]).sampleStdev()</span>
<span class="sd">        1.0</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">stats</span><span class="p">()</span><span class="o">.</span><span class="n">sampleStdev</span><span class="p">()</span>
</div>
<div class="viewcode-block" id="RDD.sampleVariance"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.sampleVariance">[docs]</a>    <span class="k">def</span> <span class="nf">sampleVariance</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Compute the sample variance of this RDD&#39;s elements (which corrects</span>
<span class="sd">        for bias in estimating the variance by dividing by N-1 instead of N).</span>

<span class="sd">        &gt;&gt;&gt; sc.parallelize([1, 2, 3]).sampleVariance()</span>
<span class="sd">        1.0</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">stats</span><span class="p">()</span><span class="o">.</span><span class="n">sampleVariance</span><span class="p">()</span>
</div>
<div class="viewcode-block" id="RDD.countByValue"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.countByValue">[docs]</a>    <span class="k">def</span> <span class="nf">countByValue</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Return the count of each unique value in this RDD as a dictionary of</span>
<span class="sd">        (value, count) pairs.</span>

<span class="sd">        &gt;&gt;&gt; sorted(sc.parallelize([1, 2, 1, 2, 2], 2).countByValue().items())</span>
<span class="sd">        [(1, 2), (2, 3)]</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">def</span> <span class="nf">countPartition</span><span class="p">(</span><span class="n">iterator</span><span class="p">):</span>
            <span class="n">counts</span> <span class="o">=</span> <span class="n">defaultdict</span><span class="p">(</span><span class="nb">int</span><span class="p">)</span>
            <span class="k">for</span> <span class="n">obj</span> <span class="ow">in</span> <span class="n">iterator</span><span class="p">:</span>
                <span class="n">counts</span><span class="p">[</span><span class="n">obj</span><span class="p">]</span> <span class="o">+=</span> <span class="mi">1</span>
            <span class="k">yield</span> <span class="n">counts</span>

        <span class="k">def</span> <span class="nf">mergeMaps</span><span class="p">(</span><span class="n">m1</span><span class="p">,</span> <span class="n">m2</span><span class="p">):</span>
            <span class="k">for</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span> <span class="ow">in</span> <span class="n">m2</span><span class="o">.</span><span class="n">items</span><span class="p">():</span>
                <span class="n">m1</span><span class="p">[</span><span class="n">k</span><span class="p">]</span> <span class="o">+=</span> <span class="n">v</span>
            <span class="k">return</span> <span class="n">m1</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">mapPartitions</span><span class="p">(</span><span class="n">countPartition</span><span class="p">)</span><span class="o">.</span><span class="n">reduce</span><span class="p">(</span><span class="n">mergeMaps</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.top"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.top">[docs]</a>    <span class="k">def</span> <span class="nf">top</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">num</span><span class="p">,</span> <span class="n">key</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Get the top N elements from a RDD.</span>

<span class="sd">        Note: It returns the list sorted in descending order.</span>

<span class="sd">        &gt;&gt;&gt; sc.parallelize([10, 4, 2, 12, 3]).top(1)</span>
<span class="sd">        [12]</span>
<span class="sd">        &gt;&gt;&gt; sc.parallelize([2, 3, 4, 5, 6], 2).top(2)</span>
<span class="sd">        [6, 5]</span>
<span class="sd">        &gt;&gt;&gt; sc.parallelize([10, 4, 2, 12, 3]).top(3, key=str)</span>
<span class="sd">        [4, 3, 2]</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">def</span> <span class="nf">topIterator</span><span class="p">(</span><span class="n">iterator</span><span class="p">):</span>
            <span class="k">yield</span> <span class="n">heapq</span><span class="o">.</span><span class="n">nlargest</span><span class="p">(</span><span class="n">num</span><span class="p">,</span> <span class="n">iterator</span><span class="p">,</span> <span class="n">key</span><span class="o">=</span><span class="n">key</span><span class="p">)</span>

        <span class="k">def</span> <span class="nf">merge</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">):</span>
            <span class="k">return</span> <span class="n">heapq</span><span class="o">.</span><span class="n">nlargest</span><span class="p">(</span><span class="n">num</span><span class="p">,</span> <span class="n">a</span> <span class="o">+</span> <span class="n">b</span><span class="p">,</span> <span class="n">key</span><span class="o">=</span><span class="n">key</span><span class="p">)</span>

        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">mapPartitions</span><span class="p">(</span><span class="n">topIterator</span><span class="p">)</span><span class="o">.</span><span class="n">reduce</span><span class="p">(</span><span class="n">merge</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.takeOrdered"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.takeOrdered">[docs]</a>    <span class="k">def</span> <span class="nf">takeOrdered</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">num</span><span class="p">,</span> <span class="n">key</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Get the N elements from a RDD ordered in ascending order or as</span>
<span class="sd">        specified by the optional key function.</span>

<span class="sd">        &gt;&gt;&gt; sc.parallelize([10, 1, 2, 9, 3, 4, 5, 6, 7]).takeOrdered(6)</span>
<span class="sd">        [1, 2, 3, 4, 5, 6]</span>
<span class="sd">        &gt;&gt;&gt; sc.parallelize([10, 1, 2, 9, 3, 4, 5, 6, 7], 2).takeOrdered(6, key=lambda x: -x)</span>
<span class="sd">        [10, 9, 7, 6, 5, 4]</span>
<span class="sd">        &quot;&quot;&quot;</span>

        <span class="k">def</span> <span class="nf">merge</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">):</span>
            <span class="k">return</span> <span class="n">heapq</span><span class="o">.</span><span class="n">nsmallest</span><span class="p">(</span><span class="n">num</span><span class="p">,</span> <span class="n">a</span> <span class="o">+</span> <span class="n">b</span><span class="p">,</span> <span class="n">key</span><span class="p">)</span>

        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">mapPartitions</span><span class="p">(</span><span class="k">lambda</span> <span class="n">it</span><span class="p">:</span> <span class="p">[</span><span class="n">heapq</span><span class="o">.</span><span class="n">nsmallest</span><span class="p">(</span><span class="n">num</span><span class="p">,</span> <span class="n">it</span><span class="p">,</span> <span class="n">key</span><span class="p">)])</span><span class="o">.</span><span class="n">reduce</span><span class="p">(</span><span class="n">merge</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.take"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.take">[docs]</a>    <span class="k">def</span> <span class="nf">take</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">num</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Take the first num elements of the RDD.</span>

<span class="sd">        It works by first scanning one partition, and use the results from</span>
<span class="sd">        that partition to estimate the number of additional partitions needed</span>
<span class="sd">        to satisfy the limit.</span>

<span class="sd">        Translated from the Scala implementation in RDD#take().</span>

<span class="sd">        &gt;&gt;&gt; sc.parallelize([2, 3, 4, 5, 6]).cache().take(2)</span>
<span class="sd">        [2, 3]</span>
<span class="sd">        &gt;&gt;&gt; sc.parallelize([2, 3, 4, 5, 6]).take(10)</span>
<span class="sd">        [2, 3, 4, 5, 6]</span>
<span class="sd">        &gt;&gt;&gt; sc.parallelize(range(100), 100).filter(lambda x: x &gt; 90).take(3)</span>
<span class="sd">        [91, 92, 93]</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="n">items</span> <span class="o">=</span> <span class="p">[]</span>
        <span class="n">totalParts</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">getNumPartitions</span><span class="p">()</span>
        <span class="n">partsScanned</span> <span class="o">=</span> <span class="mi">0</span>

        <span class="k">while</span> <span class="nb">len</span><span class="p">(</span><span class="n">items</span><span class="p">)</span> <span class="o">&lt;</span> <span class="n">num</span> <span class="ow">and</span> <span class="n">partsScanned</span> <span class="o">&lt;</span> <span class="n">totalParts</span><span class="p">:</span>
            <span class="c"># The number of partitions to try in this iteration.</span>
            <span class="c"># It is ok for this number to be greater than totalParts because</span>
            <span class="c"># we actually cap it at totalParts in runJob.</span>
            <span class="n">numPartsToTry</span> <span class="o">=</span> <span class="mi">1</span>
            <span class="k">if</span> <span class="n">partsScanned</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
                <span class="c"># If we didn&#39;t find any rows after the previous iteration,</span>
                <span class="c"># quadruple and retry.  Otherwise, interpolate the number of</span>
                <span class="c"># partitions we need to try, but overestimate it by 50%.</span>
                <span class="c"># We also cap the estimation in the end.</span>
                <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">items</span><span class="p">)</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
                    <span class="n">numPartsToTry</span> <span class="o">=</span> <span class="n">partsScanned</span> <span class="o">*</span> <span class="mi">4</span>
                <span class="k">else</span><span class="p">:</span>
                    <span class="c"># the first paramter of max is &gt;=1 whenever partsScanned &gt;= 2</span>
                    <span class="n">numPartsToTry</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="mf">1.5</span> <span class="o">*</span> <span class="n">num</span> <span class="o">*</span> <span class="n">partsScanned</span> <span class="o">/</span> <span class="nb">len</span><span class="p">(</span><span class="n">items</span><span class="p">))</span> <span class="o">-</span> <span class="n">partsScanned</span>
                    <span class="n">numPartsToTry</span> <span class="o">=</span> <span class="nb">min</span><span class="p">(</span><span class="nb">max</span><span class="p">(</span><span class="n">numPartsToTry</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span> <span class="n">partsScanned</span> <span class="o">*</span> <span class="mi">4</span><span class="p">)</span>

            <span class="n">left</span> <span class="o">=</span> <span class="n">num</span> <span class="o">-</span> <span class="nb">len</span><span class="p">(</span><span class="n">items</span><span class="p">)</span>

            <span class="k">def</span> <span class="nf">takeUpToNumLeft</span><span class="p">(</span><span class="n">iterator</span><span class="p">):</span>
                <span class="n">iterator</span> <span class="o">=</span> <span class="nb">iter</span><span class="p">(</span><span class="n">iterator</span><span class="p">)</span>
                <span class="n">taken</span> <span class="o">=</span> <span class="mi">0</span>
                <span class="k">while</span> <span class="n">taken</span> <span class="o">&lt;</span> <span class="n">left</span><span class="p">:</span>
                    <span class="k">yield</span> <span class="nb">next</span><span class="p">(</span><span class="n">iterator</span><span class="p">)</span>
                    <span class="n">taken</span> <span class="o">+=</span> <span class="mi">1</span>

            <span class="n">p</span> <span class="o">=</span> <span class="nb">range</span><span class="p">(</span><span class="n">partsScanned</span><span class="p">,</span> <span class="nb">min</span><span class="p">(</span><span class="n">partsScanned</span> <span class="o">+</span> <span class="n">numPartsToTry</span><span class="p">,</span> <span class="n">totalParts</span><span class="p">))</span>
            <span class="n">res</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">context</span><span class="o">.</span><span class="n">runJob</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">takeUpToNumLeft</span><span class="p">,</span> <span class="n">p</span><span class="p">)</span>

            <span class="n">items</span> <span class="o">+=</span> <span class="n">res</span>
            <span class="n">partsScanned</span> <span class="o">+=</span> <span class="n">numPartsToTry</span>

        <span class="k">return</span> <span class="n">items</span><span class="p">[:</span><span class="n">num</span><span class="p">]</span>
</div>
<div class="viewcode-block" id="RDD.first"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.first">[docs]</a>    <span class="k">def</span> <span class="nf">first</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Return the first element in this RDD.</span>

<span class="sd">        &gt;&gt;&gt; sc.parallelize([2, 3, 4]).first()</span>
<span class="sd">        2</span>
<span class="sd">        &gt;&gt;&gt; sc.parallelize([]).first()</span>
<span class="sd">        Traceback (most recent call last):</span>
<span class="sd">            ...</span>
<span class="sd">        ValueError: RDD is empty</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="n">rs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">take</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span>
        <span class="k">if</span> <span class="n">rs</span><span class="p">:</span>
            <span class="k">return</span> <span class="n">rs</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s">&quot;RDD is empty&quot;</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.isEmpty"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.isEmpty">[docs]</a>    <span class="k">def</span> <span class="nf">isEmpty</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Returns true if and only if the RDD contains no elements at all. Note that an RDD</span>
<span class="sd">        may be empty even when it has at least 1 partition.</span>

<span class="sd">        &gt;&gt;&gt; sc.parallelize([]).isEmpty()</span>
<span class="sd">        True</span>
<span class="sd">        &gt;&gt;&gt; sc.parallelize([1]).isEmpty()</span>
<span class="sd">        False</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">getNumPartitions</span><span class="p">()</span> <span class="o">==</span> <span class="mi">0</span> <span class="ow">or</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">take</span><span class="p">(</span><span class="mi">1</span><span class="p">))</span> <span class="o">==</span> <span class="mi">0</span>
</div>
<div class="viewcode-block" id="RDD.saveAsNewAPIHadoopDataset"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.saveAsNewAPIHadoopDataset">[docs]</a>    <span class="k">def</span> <span class="nf">saveAsNewAPIHadoopDataset</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">conf</span><span class="p">,</span> <span class="n">keyConverter</span><span class="o">=</span><span class="bp">None</span><span class="p">,</span> <span class="n">valueConverter</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Output a Python RDD of key-value pairs (of form C{RDD[(K, V)]}) to any Hadoop file</span>
<span class="sd">        system, using the new Hadoop OutputFormat API (mapreduce package). Keys/values are</span>
<span class="sd">        converted for output using either user specified converters or, by default,</span>
<span class="sd">        L{org.apache.spark.api.python.JavaToWritableConverter}.</span>

<span class="sd">        :param conf: Hadoop job configuration, passed in as a dict</span>
<span class="sd">        :param keyConverter: (None by default)</span>
<span class="sd">        :param valueConverter: (None by default)</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="n">jconf</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="o">.</span><span class="n">_dictToJavaMap</span><span class="p">(</span><span class="n">conf</span><span class="p">)</span>
        <span class="n">pickledRDD</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_pickled</span><span class="p">()</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="o">.</span><span class="n">_jvm</span><span class="o">.</span><span class="n">PythonRDD</span><span class="o">.</span><span class="n">saveAsHadoopDataset</span><span class="p">(</span><span class="n">pickledRDD</span><span class="o">.</span><span class="n">_jrdd</span><span class="p">,</span> <span class="bp">True</span><span class="p">,</span> <span class="n">jconf</span><span class="p">,</span>
                                                    <span class="n">keyConverter</span><span class="p">,</span> <span class="n">valueConverter</span><span class="p">,</span> <span class="bp">True</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.saveAsNewAPIHadoopFile"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.saveAsNewAPIHadoopFile">[docs]</a>    <span class="k">def</span> <span class="nf">saveAsNewAPIHadoopFile</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">path</span><span class="p">,</span> <span class="n">outputFormatClass</span><span class="p">,</span> <span class="n">keyClass</span><span class="o">=</span><span class="bp">None</span><span class="p">,</span> <span class="n">valueClass</span><span class="o">=</span><span class="bp">None</span><span class="p">,</span>
                               <span class="n">keyConverter</span><span class="o">=</span><span class="bp">None</span><span class="p">,</span> <span class="n">valueConverter</span><span class="o">=</span><span class="bp">None</span><span class="p">,</span> <span class="n">conf</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Output a Python RDD of key-value pairs (of form C{RDD[(K, V)]}) to any Hadoop file</span>
<span class="sd">        system, using the new Hadoop OutputFormat API (mapreduce package). Key and value types</span>
<span class="sd">        will be inferred if not specified. Keys and values are converted for output using either</span>
<span class="sd">        user specified converters or L{org.apache.spark.api.python.JavaToWritableConverter}. The</span>
<span class="sd">        C{conf} is applied on top of the base Hadoop conf associated with the SparkContext</span>
<span class="sd">        of this RDD to create a merged Hadoop MapReduce job configuration for saving the data.</span>

<span class="sd">        :param path: path to Hadoop file</span>
<span class="sd">        :param outputFormatClass: fully qualified classname of Hadoop OutputFormat</span>
<span class="sd">               (e.g. &quot;org.apache.hadoop.mapreduce.lib.output.SequenceFileOutputFormat&quot;)</span>
<span class="sd">        :param keyClass: fully qualified classname of key Writable class</span>
<span class="sd">               (e.g. &quot;org.apache.hadoop.io.IntWritable&quot;, None by default)</span>
<span class="sd">        :param valueClass: fully qualified classname of value Writable class</span>
<span class="sd">               (e.g. &quot;org.apache.hadoop.io.Text&quot;, None by default)</span>
<span class="sd">        :param keyConverter: (None by default)</span>
<span class="sd">        :param valueConverter: (None by default)</span>
<span class="sd">        :param conf: Hadoop job configuration, passed in as a dict (None by default)</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="n">jconf</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="o">.</span><span class="n">_dictToJavaMap</span><span class="p">(</span><span class="n">conf</span><span class="p">)</span>
        <span class="n">pickledRDD</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_pickled</span><span class="p">()</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="o">.</span><span class="n">_jvm</span><span class="o">.</span><span class="n">PythonRDD</span><span class="o">.</span><span class="n">saveAsNewAPIHadoopFile</span><span class="p">(</span><span class="n">pickledRDD</span><span class="o">.</span><span class="n">_jrdd</span><span class="p">,</span> <span class="bp">True</span><span class="p">,</span> <span class="n">path</span><span class="p">,</span>
                                                       <span class="n">outputFormatClass</span><span class="p">,</span>
                                                       <span class="n">keyClass</span><span class="p">,</span> <span class="n">valueClass</span><span class="p">,</span>
                                                       <span class="n">keyConverter</span><span class="p">,</span> <span class="n">valueConverter</span><span class="p">,</span> <span class="n">jconf</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.saveAsHadoopDataset"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.saveAsHadoopDataset">[docs]</a>    <span class="k">def</span> <span class="nf">saveAsHadoopDataset</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">conf</span><span class="p">,</span> <span class="n">keyConverter</span><span class="o">=</span><span class="bp">None</span><span class="p">,</span> <span class="n">valueConverter</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Output a Python RDD of key-value pairs (of form C{RDD[(K, V)]}) to any Hadoop file</span>
<span class="sd">        system, using the old Hadoop OutputFormat API (mapred package). Keys/values are</span>
<span class="sd">        converted for output using either user specified converters or, by default,</span>
<span class="sd">        L{org.apache.spark.api.python.JavaToWritableConverter}.</span>

<span class="sd">        :param conf: Hadoop job configuration, passed in as a dict</span>
<span class="sd">        :param keyConverter: (None by default)</span>
<span class="sd">        :param valueConverter: (None by default)</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="n">jconf</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="o">.</span><span class="n">_dictToJavaMap</span><span class="p">(</span><span class="n">conf</span><span class="p">)</span>
        <span class="n">pickledRDD</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_pickled</span><span class="p">()</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="o">.</span><span class="n">_jvm</span><span class="o">.</span><span class="n">PythonRDD</span><span class="o">.</span><span class="n">saveAsHadoopDataset</span><span class="p">(</span><span class="n">pickledRDD</span><span class="o">.</span><span class="n">_jrdd</span><span class="p">,</span> <span class="bp">True</span><span class="p">,</span> <span class="n">jconf</span><span class="p">,</span>
                                                    <span class="n">keyConverter</span><span class="p">,</span> <span class="n">valueConverter</span><span class="p">,</span> <span class="bp">False</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.saveAsHadoopFile"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.saveAsHadoopFile">[docs]</a>    <span class="k">def</span> <span class="nf">saveAsHadoopFile</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">path</span><span class="p">,</span> <span class="n">outputFormatClass</span><span class="p">,</span> <span class="n">keyClass</span><span class="o">=</span><span class="bp">None</span><span class="p">,</span> <span class="n">valueClass</span><span class="o">=</span><span class="bp">None</span><span class="p">,</span>
                         <span class="n">keyConverter</span><span class="o">=</span><span class="bp">None</span><span class="p">,</span> <span class="n">valueConverter</span><span class="o">=</span><span class="bp">None</span><span class="p">,</span> <span class="n">conf</span><span class="o">=</span><span class="bp">None</span><span class="p">,</span>
                         <span class="n">compressionCodecClass</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Output a Python RDD of key-value pairs (of form C{RDD[(K, V)]}) to any Hadoop file</span>
<span class="sd">        system, using the old Hadoop OutputFormat API (mapred package). Key and value types</span>
<span class="sd">        will be inferred if not specified. Keys and values are converted for output using either</span>
<span class="sd">        user specified converters or L{org.apache.spark.api.python.JavaToWritableConverter}. The</span>
<span class="sd">        C{conf} is applied on top of the base Hadoop conf associated with the SparkContext</span>
<span class="sd">        of this RDD to create a merged Hadoop MapReduce job configuration for saving the data.</span>

<span class="sd">        :param path: path to Hadoop file</span>
<span class="sd">        :param outputFormatClass: fully qualified classname of Hadoop OutputFormat</span>
<span class="sd">               (e.g. &quot;org.apache.hadoop.mapred.SequenceFileOutputFormat&quot;)</span>
<span class="sd">        :param keyClass: fully qualified classname of key Writable class</span>
<span class="sd">               (e.g. &quot;org.apache.hadoop.io.IntWritable&quot;, None by default)</span>
<span class="sd">        :param valueClass: fully qualified classname of value Writable class</span>
<span class="sd">               (e.g. &quot;org.apache.hadoop.io.Text&quot;, None by default)</span>
<span class="sd">        :param keyConverter: (None by default)</span>
<span class="sd">        :param valueConverter: (None by default)</span>
<span class="sd">        :param conf: (None by default)</span>
<span class="sd">        :param compressionCodecClass: (None by default)</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="n">jconf</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="o">.</span><span class="n">_dictToJavaMap</span><span class="p">(</span><span class="n">conf</span><span class="p">)</span>
        <span class="n">pickledRDD</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_pickled</span><span class="p">()</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="o">.</span><span class="n">_jvm</span><span class="o">.</span><span class="n">PythonRDD</span><span class="o">.</span><span class="n">saveAsHadoopFile</span><span class="p">(</span><span class="n">pickledRDD</span><span class="o">.</span><span class="n">_jrdd</span><span class="p">,</span> <span class="bp">True</span><span class="p">,</span> <span class="n">path</span><span class="p">,</span>
                                                 <span class="n">outputFormatClass</span><span class="p">,</span>
                                                 <span class="n">keyClass</span><span class="p">,</span> <span class="n">valueClass</span><span class="p">,</span>
                                                 <span class="n">keyConverter</span><span class="p">,</span> <span class="n">valueConverter</span><span class="p">,</span>
                                                 <span class="n">jconf</span><span class="p">,</span> <span class="n">compressionCodecClass</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.saveAsSequenceFile"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.saveAsSequenceFile">[docs]</a>    <span class="k">def</span> <span class="nf">saveAsSequenceFile</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">path</span><span class="p">,</span> <span class="n">compressionCodecClass</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Output a Python RDD of key-value pairs (of form C{RDD[(K, V)]}) to any Hadoop file</span>
<span class="sd">        system, using the L{org.apache.hadoop.io.Writable} types that we convert from the</span>
<span class="sd">        RDD&#39;s key and value types. The mechanism is as follows:</span>

<span class="sd">            1. Pyrolite is used to convert pickled Python RDD into RDD of Java objects.</span>
<span class="sd">            2. Keys and values of this Java RDD are converted to Writables and written out.</span>

<span class="sd">        :param path: path to sequence file</span>
<span class="sd">        :param compressionCodecClass: (None by default)</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="n">pickledRDD</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_pickled</span><span class="p">()</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="o">.</span><span class="n">_jvm</span><span class="o">.</span><span class="n">PythonRDD</span><span class="o">.</span><span class="n">saveAsSequenceFile</span><span class="p">(</span><span class="n">pickledRDD</span><span class="o">.</span><span class="n">_jrdd</span><span class="p">,</span> <span class="bp">True</span><span class="p">,</span>
                                                   <span class="n">path</span><span class="p">,</span> <span class="n">compressionCodecClass</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.saveAsPickleFile"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.saveAsPickleFile">[docs]</a>    <span class="k">def</span> <span class="nf">saveAsPickleFile</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">path</span><span class="p">,</span> <span class="n">batchSize</span><span class="o">=</span><span class="mi">10</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Save this RDD as a SequenceFile of serialized objects. The serializer</span>
<span class="sd">        used is L{pyspark.serializers.PickleSerializer}, default batch size</span>
<span class="sd">        is 10.</span>

<span class="sd">        &gt;&gt;&gt; tmpFile = NamedTemporaryFile(delete=True)</span>
<span class="sd">        &gt;&gt;&gt; tmpFile.close()</span>
<span class="sd">        &gt;&gt;&gt; sc.parallelize([1, 2, &#39;spark&#39;, &#39;rdd&#39;]).saveAsPickleFile(tmpFile.name, 3)</span>
<span class="sd">        &gt;&gt;&gt; sorted(sc.pickleFile(tmpFile.name, 5).map(str).collect())</span>
<span class="sd">        [&#39;1&#39;, &#39;2&#39;, &#39;rdd&#39;, &#39;spark&#39;]</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">if</span> <span class="n">batchSize</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
            <span class="n">ser</span> <span class="o">=</span> <span class="n">AutoBatchedSerializer</span><span class="p">(</span><span class="n">PickleSerializer</span><span class="p">())</span>
        <span class="k">else</span><span class="p">:</span>
            <span class="n">ser</span> <span class="o">=</span> <span class="n">BatchedSerializer</span><span class="p">(</span><span class="n">PickleSerializer</span><span class="p">(),</span> <span class="n">batchSize</span><span class="p">)</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">_reserialize</span><span class="p">(</span><span class="n">ser</span><span class="p">)</span><span class="o">.</span><span class="n">_jrdd</span><span class="o">.</span><span class="n">saveAsObjectFile</span><span class="p">(</span><span class="n">path</span><span class="p">)</span>
</div>
    <span class="nd">@ignore_unicode_prefix</span>
<div class="viewcode-block" id="RDD.saveAsTextFile"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.saveAsTextFile">[docs]</a>    <span class="k">def</span> <span class="nf">saveAsTextFile</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">path</span><span class="p">,</span> <span class="n">compressionCodecClass</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Save this RDD as a text file, using string representations of elements.</span>

<span class="sd">        @param path: path to text file</span>
<span class="sd">        @param compressionCodecClass: (None by default) string i.e.</span>
<span class="sd">            &quot;org.apache.hadoop.io.compress.GzipCodec&quot;</span>

<span class="sd">        &gt;&gt;&gt; tempFile = NamedTemporaryFile(delete=True)</span>
<span class="sd">        &gt;&gt;&gt; tempFile.close()</span>
<span class="sd">        &gt;&gt;&gt; sc.parallelize(range(10)).saveAsTextFile(tempFile.name)</span>
<span class="sd">        &gt;&gt;&gt; from fileinput import input</span>
<span class="sd">        &gt;&gt;&gt; from glob import glob</span>
<span class="sd">        &gt;&gt;&gt; &#39;&#39;.join(sorted(input(glob(tempFile.name + &quot;/part-0000*&quot;))))</span>
<span class="sd">        &#39;0\\n1\\n2\\n3\\n4\\n5\\n6\\n7\\n8\\n9\\n&#39;</span>

<span class="sd">        Empty lines are tolerated when saving to text files.</span>

<span class="sd">        &gt;&gt;&gt; tempFile2 = NamedTemporaryFile(delete=True)</span>
<span class="sd">        &gt;&gt;&gt; tempFile2.close()</span>
<span class="sd">        &gt;&gt;&gt; sc.parallelize([&#39;&#39;, &#39;foo&#39;, &#39;&#39;, &#39;bar&#39;, &#39;&#39;]).saveAsTextFile(tempFile2.name)</span>
<span class="sd">        &gt;&gt;&gt; &#39;&#39;.join(sorted(input(glob(tempFile2.name + &quot;/part-0000*&quot;))))</span>
<span class="sd">        &#39;\\n\\n\\nbar\\nfoo\\n&#39;</span>

<span class="sd">        Using compressionCodecClass</span>

<span class="sd">        &gt;&gt;&gt; tempFile3 = NamedTemporaryFile(delete=True)</span>
<span class="sd">        &gt;&gt;&gt; tempFile3.close()</span>
<span class="sd">        &gt;&gt;&gt; codec = &quot;org.apache.hadoop.io.compress.GzipCodec&quot;</span>
<span class="sd">        &gt;&gt;&gt; sc.parallelize([&#39;foo&#39;, &#39;bar&#39;]).saveAsTextFile(tempFile3.name, codec)</span>
<span class="sd">        &gt;&gt;&gt; from fileinput import input, hook_compressed</span>
<span class="sd">        &gt;&gt;&gt; result = sorted(input(glob(tempFile3.name + &quot;/part*.gz&quot;), openhook=hook_compressed))</span>
<span class="sd">        &gt;&gt;&gt; b&#39;&#39;.join(result).decode(&#39;utf-8&#39;)</span>
<span class="sd">        u&#39;bar\\nfoo\\n&#39;</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">def</span> <span class="nf">func</span><span class="p">(</span><span class="n">split</span><span class="p">,</span> <span class="n">iterator</span><span class="p">):</span>
            <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">iterator</span><span class="p">:</span>
                <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="p">(</span><span class="nb">unicode</span><span class="p">,</span> <span class="nb">bytes</span><span class="p">)):</span>
                    <span class="n">x</span> <span class="o">=</span> <span class="nb">unicode</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
                <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="nb">unicode</span><span class="p">):</span>
                    <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">encode</span><span class="p">(</span><span class="s">&quot;utf-8&quot;</span><span class="p">)</span>
                <span class="k">yield</span> <span class="n">x</span>
        <span class="n">keyed</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">mapPartitionsWithIndex</span><span class="p">(</span><span class="n">func</span><span class="p">)</span>
        <span class="n">keyed</span><span class="o">.</span><span class="n">_bypass_serializer</span> <span class="o">=</span> <span class="bp">True</span>
        <span class="k">if</span> <span class="n">compressionCodecClass</span><span class="p">:</span>
            <span class="n">compressionCodec</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="o">.</span><span class="n">_jvm</span><span class="o">.</span><span class="n">java</span><span class="o">.</span><span class="n">lang</span><span class="o">.</span><span class="n">Class</span><span class="o">.</span><span class="n">forName</span><span class="p">(</span><span class="n">compressionCodecClass</span><span class="p">)</span>
            <span class="n">keyed</span><span class="o">.</span><span class="n">_jrdd</span><span class="o">.</span><span class="n">map</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="o">.</span><span class="n">_jvm</span><span class="o">.</span><span class="n">BytesToString</span><span class="p">())</span><span class="o">.</span><span class="n">saveAsTextFile</span><span class="p">(</span><span class="n">path</span><span class="p">,</span> <span class="n">compressionCodec</span><span class="p">)</span>
        <span class="k">else</span><span class="p">:</span>
            <span class="n">keyed</span><span class="o">.</span><span class="n">_jrdd</span><span class="o">.</span><span class="n">map</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="o">.</span><span class="n">_jvm</span><span class="o">.</span><span class="n">BytesToString</span><span class="p">())</span><span class="o">.</span><span class="n">saveAsTextFile</span><span class="p">(</span><span class="n">path</span><span class="p">)</span>

    <span class="c"># Pair functions</span>
</div>
<div class="viewcode-block" id="RDD.collectAsMap"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.collectAsMap">[docs]</a>    <span class="k">def</span> <span class="nf">collectAsMap</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Return the key-value pairs in this RDD to the master as a dictionary.</span>

<span class="sd">        &gt;&gt;&gt; m = sc.parallelize([(1, 2), (3, 4)]).collectAsMap()</span>
<span class="sd">        &gt;&gt;&gt; m[1]</span>
<span class="sd">        2</span>
<span class="sd">        &gt;&gt;&gt; m[3]</span>
<span class="sd">        4</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">return</span> <span class="nb">dict</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">collect</span><span class="p">())</span>
</div>
<div class="viewcode-block" id="RDD.keys"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.keys">[docs]</a>    <span class="k">def</span> <span class="nf">keys</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Return an RDD with the keys of each tuple.</span>

<span class="sd">        &gt;&gt;&gt; m = sc.parallelize([(1, 2), (3, 4)]).keys()</span>
<span class="sd">        &gt;&gt;&gt; m.collect()</span>
<span class="sd">        [1, 3]</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">map</span><span class="p">(</span><span class="k">lambda</span> <span class="n">x</span><span class="p">:</span> <span class="n">x</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span>
</div>
<div class="viewcode-block" id="RDD.values"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.values">[docs]</a>    <span class="k">def</span> <span class="nf">values</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Return an RDD with the values of each tuple.</span>

<span class="sd">        &gt;&gt;&gt; m = sc.parallelize([(1, 2), (3, 4)]).values()</span>
<span class="sd">        &gt;&gt;&gt; m.collect()</span>
<span class="sd">        [2, 4]</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">map</span><span class="p">(</span><span class="k">lambda</span> <span class="n">x</span><span class="p">:</span> <span class="n">x</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span>
</div>
<div class="viewcode-block" id="RDD.reduceByKey"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.reduceByKey">[docs]</a>    <span class="k">def</span> <span class="nf">reduceByKey</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">func</span><span class="p">,</span> <span class="n">numPartitions</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Merge the values for each key using an associative reduce function.</span>

<span class="sd">        This will also perform the merging locally on each mapper before</span>
<span class="sd">        sending results to a reducer, similarly to a &quot;combiner&quot; in MapReduce.</span>

<span class="sd">        Output will be hash-partitioned with C{numPartitions} partitions, or</span>
<span class="sd">        the default parallelism level if C{numPartitions} is not specified.</span>

<span class="sd">        &gt;&gt;&gt; from operator import add</span>
<span class="sd">        &gt;&gt;&gt; rdd = sc.parallelize([(&quot;a&quot;, 1), (&quot;b&quot;, 1), (&quot;a&quot;, 1)])</span>
<span class="sd">        &gt;&gt;&gt; sorted(rdd.reduceByKey(add).collect())</span>
<span class="sd">        [(&#39;a&#39;, 2), (&#39;b&#39;, 1)]</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">combineByKey</span><span class="p">(</span><span class="k">lambda</span> <span class="n">x</span><span class="p">:</span> <span class="n">x</span><span class="p">,</span> <span class="n">func</span><span class="p">,</span> <span class="n">func</span><span class="p">,</span> <span class="n">numPartitions</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.reduceByKeyLocally"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.reduceByKeyLocally">[docs]</a>    <span class="k">def</span> <span class="nf">reduceByKeyLocally</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">func</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Merge the values for each key using an associative reduce function, but</span>
<span class="sd">        return the results immediately to the master as a dictionary.</span>

<span class="sd">        This will also perform the merging locally on each mapper before</span>
<span class="sd">        sending results to a reducer, similarly to a &quot;combiner&quot; in MapReduce.</span>

<span class="sd">        &gt;&gt;&gt; from operator import add</span>
<span class="sd">        &gt;&gt;&gt; rdd = sc.parallelize([(&quot;a&quot;, 1), (&quot;b&quot;, 1), (&quot;a&quot;, 1)])</span>
<span class="sd">        &gt;&gt;&gt; sorted(rdd.reduceByKeyLocally(add).items())</span>
<span class="sd">        [(&#39;a&#39;, 2), (&#39;b&#39;, 1)]</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">def</span> <span class="nf">reducePartition</span><span class="p">(</span><span class="n">iterator</span><span class="p">):</span>
            <span class="n">m</span> <span class="o">=</span> <span class="p">{}</span>
            <span class="k">for</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span> <span class="ow">in</span> <span class="n">iterator</span><span class="p">:</span>
                <span class="n">m</span><span class="p">[</span><span class="n">k</span><span class="p">]</span> <span class="o">=</span> <span class="n">func</span><span class="p">(</span><span class="n">m</span><span class="p">[</span><span class="n">k</span><span class="p">],</span> <span class="n">v</span><span class="p">)</span> <span class="k">if</span> <span class="n">k</span> <span class="ow">in</span> <span class="n">m</span> <span class="k">else</span> <span class="n">v</span>
            <span class="k">yield</span> <span class="n">m</span>

        <span class="k">def</span> <span class="nf">mergeMaps</span><span class="p">(</span><span class="n">m1</span><span class="p">,</span> <span class="n">m2</span><span class="p">):</span>
            <span class="k">for</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span> <span class="ow">in</span> <span class="n">m2</span><span class="o">.</span><span class="n">items</span><span class="p">():</span>
                <span class="n">m1</span><span class="p">[</span><span class="n">k</span><span class="p">]</span> <span class="o">=</span> <span class="n">func</span><span class="p">(</span><span class="n">m1</span><span class="p">[</span><span class="n">k</span><span class="p">],</span> <span class="n">v</span><span class="p">)</span> <span class="k">if</span> <span class="n">k</span> <span class="ow">in</span> <span class="n">m1</span> <span class="k">else</span> <span class="n">v</span>
            <span class="k">return</span> <span class="n">m1</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">mapPartitions</span><span class="p">(</span><span class="n">reducePartition</span><span class="p">)</span><span class="o">.</span><span class="n">reduce</span><span class="p">(</span><span class="n">mergeMaps</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.countByKey"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.countByKey">[docs]</a>    <span class="k">def</span> <span class="nf">countByKey</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Count the number of elements for each key, and return the result to the</span>
<span class="sd">        master as a dictionary.</span>

<span class="sd">        &gt;&gt;&gt; rdd = sc.parallelize([(&quot;a&quot;, 1), (&quot;b&quot;, 1), (&quot;a&quot;, 1)])</span>
<span class="sd">        &gt;&gt;&gt; sorted(rdd.countByKey().items())</span>
<span class="sd">        [(&#39;a&#39;, 2), (&#39;b&#39;, 1)]</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">map</span><span class="p">(</span><span class="k">lambda</span> <span class="n">x</span><span class="p">:</span> <span class="n">x</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span><span class="o">.</span><span class="n">countByValue</span><span class="p">()</span>
</div>
<div class="viewcode-block" id="RDD.join"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.join">[docs]</a>    <span class="k">def</span> <span class="nf">join</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">other</span><span class="p">,</span> <span class="n">numPartitions</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Return an RDD containing all pairs of elements with matching keys in</span>
<span class="sd">        C{self} and C{other}.</span>

<span class="sd">        Each pair of elements will be returned as a (k, (v1, v2)) tuple, where</span>
<span class="sd">        (k, v1) is in C{self} and (k, v2) is in C{other}.</span>

<span class="sd">        Performs a hash join across the cluster.</span>

<span class="sd">        &gt;&gt;&gt; x = sc.parallelize([(&quot;a&quot;, 1), (&quot;b&quot;, 4)])</span>
<span class="sd">        &gt;&gt;&gt; y = sc.parallelize([(&quot;a&quot;, 2), (&quot;a&quot;, 3)])</span>
<span class="sd">        &gt;&gt;&gt; sorted(x.join(y).collect())</span>
<span class="sd">        [(&#39;a&#39;, (1, 2)), (&#39;a&#39;, (1, 3))]</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">return</span> <span class="n">python_join</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">other</span><span class="p">,</span> <span class="n">numPartitions</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.leftOuterJoin"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.leftOuterJoin">[docs]</a>    <span class="k">def</span> <span class="nf">leftOuterJoin</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">other</span><span class="p">,</span> <span class="n">numPartitions</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Perform a left outer join of C{self} and C{other}.</span>

<span class="sd">        For each element (k, v) in C{self}, the resulting RDD will either</span>
<span class="sd">        contain all pairs (k, (v, w)) for w in C{other}, or the pair</span>
<span class="sd">        (k, (v, None)) if no elements in C{other} have key k.</span>

<span class="sd">        Hash-partitions the resulting RDD into the given number of partitions.</span>

<span class="sd">        &gt;&gt;&gt; x = sc.parallelize([(&quot;a&quot;, 1), (&quot;b&quot;, 4)])</span>
<span class="sd">        &gt;&gt;&gt; y = sc.parallelize([(&quot;a&quot;, 2)])</span>
<span class="sd">        &gt;&gt;&gt; sorted(x.leftOuterJoin(y).collect())</span>
<span class="sd">        [(&#39;a&#39;, (1, 2)), (&#39;b&#39;, (4, None))]</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">return</span> <span class="n">python_left_outer_join</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">other</span><span class="p">,</span> <span class="n">numPartitions</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.rightOuterJoin"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.rightOuterJoin">[docs]</a>    <span class="k">def</span> <span class="nf">rightOuterJoin</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">other</span><span class="p">,</span> <span class="n">numPartitions</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Perform a right outer join of C{self} and C{other}.</span>

<span class="sd">        For each element (k, w) in C{other}, the resulting RDD will either</span>
<span class="sd">        contain all pairs (k, (v, w)) for v in this, or the pair (k, (None, w))</span>
<span class="sd">        if no elements in C{self} have key k.</span>

<span class="sd">        Hash-partitions the resulting RDD into the given number of partitions.</span>

<span class="sd">        &gt;&gt;&gt; x = sc.parallelize([(&quot;a&quot;, 1), (&quot;b&quot;, 4)])</span>
<span class="sd">        &gt;&gt;&gt; y = sc.parallelize([(&quot;a&quot;, 2)])</span>
<span class="sd">        &gt;&gt;&gt; sorted(y.rightOuterJoin(x).collect())</span>
<span class="sd">        [(&#39;a&#39;, (2, 1)), (&#39;b&#39;, (None, 4))]</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">return</span> <span class="n">python_right_outer_join</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">other</span><span class="p">,</span> <span class="n">numPartitions</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.fullOuterJoin"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.fullOuterJoin">[docs]</a>    <span class="k">def</span> <span class="nf">fullOuterJoin</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">other</span><span class="p">,</span> <span class="n">numPartitions</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Perform a right outer join of C{self} and C{other}.</span>

<span class="sd">        For each element (k, v) in C{self}, the resulting RDD will either</span>
<span class="sd">        contain all pairs (k, (v, w)) for w in C{other}, or the pair</span>
<span class="sd">        (k, (v, None)) if no elements in C{other} have key k.</span>

<span class="sd">        Similarly, for each element (k, w) in C{other}, the resulting RDD will</span>
<span class="sd">        either contain all pairs (k, (v, w)) for v in C{self}, or the pair</span>
<span class="sd">        (k, (None, w)) if no elements in C{self} have key k.</span>

<span class="sd">        Hash-partitions the resulting RDD into the given number of partitions.</span>

<span class="sd">        &gt;&gt;&gt; x = sc.parallelize([(&quot;a&quot;, 1), (&quot;b&quot;, 4)])</span>
<span class="sd">        &gt;&gt;&gt; y = sc.parallelize([(&quot;a&quot;, 2), (&quot;c&quot;, 8)])</span>
<span class="sd">        &gt;&gt;&gt; sorted(x.fullOuterJoin(y).collect())</span>
<span class="sd">        [(&#39;a&#39;, (1, 2)), (&#39;b&#39;, (4, None)), (&#39;c&#39;, (None, 8))]</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">return</span> <span class="n">python_full_outer_join</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">other</span><span class="p">,</span> <span class="n">numPartitions</span><span class="p">)</span>

    <span class="c"># TODO: add option to control map-side combining</span>
    <span class="c"># portable_hash is used as default, because builtin hash of None is different</span>
    <span class="c"># cross machines.</span></div>
<div class="viewcode-block" id="RDD.partitionBy"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.partitionBy">[docs]</a>    <span class="k">def</span> <span class="nf">partitionBy</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">numPartitions</span><span class="p">,</span> <span class="n">partitionFunc</span><span class="o">=</span><span class="n">portable_hash</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Return a copy of the RDD partitioned using the specified partitioner.</span>

<span class="sd">        &gt;&gt;&gt; pairs = sc.parallelize([1, 2, 3, 4, 2, 4, 1]).map(lambda x: (x, x))</span>
<span class="sd">        &gt;&gt;&gt; sets = pairs.partitionBy(2).glom().collect()</span>
<span class="sd">        &gt;&gt;&gt; len(set(sets[0]).intersection(set(sets[1])))</span>
<span class="sd">        0</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">if</span> <span class="n">numPartitions</span> <span class="ow">is</span> <span class="bp">None</span><span class="p">:</span>
            <span class="n">numPartitions</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_defaultReducePartitions</span><span class="p">()</span>
        <span class="n">partitioner</span> <span class="o">=</span> <span class="n">Partitioner</span><span class="p">(</span><span class="n">numPartitions</span><span class="p">,</span> <span class="n">partitionFunc</span><span class="p">)</span>
        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">partitioner</span> <span class="o">==</span> <span class="n">partitioner</span><span class="p">:</span>
            <span class="k">return</span> <span class="bp">self</span>

        <span class="c"># Transferring O(n) objects to Java is too expensive.</span>
        <span class="c"># Instead, we&#39;ll form the hash buckets in Python,</span>
        <span class="c"># transferring O(numPartitions) objects to Java.</span>
        <span class="c"># Each object is a (splitNumber, [objects]) pair.</span>
        <span class="c"># In order to avoid too huge objects, the objects are</span>
        <span class="c"># grouped into chunks.</span>
        <span class="n">outputSerializer</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="o">.</span><span class="n">_unbatched_serializer</span>

        <span class="n">limit</span> <span class="o">=</span> <span class="p">(</span><span class="n">_parse_memory</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="o">.</span><span class="n">_conf</span><span class="o">.</span><span class="n">get</span><span class="p">(</span>
            <span class="s">&quot;spark.python.worker.memory&quot;</span><span class="p">,</span> <span class="s">&quot;512m&quot;</span><span class="p">))</span> <span class="o">/</span> <span class="mi">2</span><span class="p">)</span>

        <span class="k">def</span> <span class="nf">add_shuffle_key</span><span class="p">(</span><span class="n">split</span><span class="p">,</span> <span class="n">iterator</span><span class="p">):</span>

            <span class="n">buckets</span> <span class="o">=</span> <span class="n">defaultdict</span><span class="p">(</span><span class="nb">list</span><span class="p">)</span>
            <span class="n">c</span><span class="p">,</span> <span class="n">batch</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span> <span class="nb">min</span><span class="p">(</span><span class="mi">10</span> <span class="o">*</span> <span class="n">numPartitions</span><span class="p">,</span> <span class="mi">1000</span><span class="p">)</span>

            <span class="k">for</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span> <span class="ow">in</span> <span class="n">iterator</span><span class="p">:</span>
                <span class="n">buckets</span><span class="p">[</span><span class="n">partitionFunc</span><span class="p">(</span><span class="n">k</span><span class="p">)</span> <span class="o">%</span> <span class="n">numPartitions</span><span class="p">]</span><span class="o">.</span><span class="n">append</span><span class="p">((</span><span class="n">k</span><span class="p">,</span> <span class="n">v</span><span class="p">))</span>
                <span class="n">c</span> <span class="o">+=</span> <span class="mi">1</span>

                <span class="c"># check used memory and avg size of chunk of objects</span>
                <span class="k">if</span> <span class="p">(</span><span class="n">c</span> <span class="o">%</span> <span class="mi">1000</span> <span class="o">==</span> <span class="mi">0</span> <span class="ow">and</span> <span class="n">get_used_memory</span><span class="p">()</span> <span class="o">&gt;</span> <span class="n">limit</span>
                        <span class="ow">or</span> <span class="n">c</span> <span class="o">&gt;</span> <span class="n">batch</span><span class="p">):</span>
                    <span class="n">n</span><span class="p">,</span> <span class="n">size</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">buckets</span><span class="p">),</span> <span class="mi">0</span>
                    <span class="k">for</span> <span class="n">split</span> <span class="ow">in</span> <span class="nb">list</span><span class="p">(</span><span class="n">buckets</span><span class="o">.</span><span class="n">keys</span><span class="p">()):</span>
                        <span class="k">yield</span> <span class="n">pack_long</span><span class="p">(</span><span class="n">split</span><span class="p">)</span>
                        <span class="n">d</span> <span class="o">=</span> <span class="n">outputSerializer</span><span class="o">.</span><span class="n">dumps</span><span class="p">(</span><span class="n">buckets</span><span class="p">[</span><span class="n">split</span><span class="p">])</span>
                        <span class="k">del</span> <span class="n">buckets</span><span class="p">[</span><span class="n">split</span><span class="p">]</span>
                        <span class="k">yield</span> <span class="n">d</span>
                        <span class="n">size</span> <span class="o">+=</span> <span class="nb">len</span><span class="p">(</span><span class="n">d</span><span class="p">)</span>

                    <span class="n">avg</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">size</span> <span class="o">/</span> <span class="n">n</span><span class="p">)</span> <span class="o">&gt;&gt;</span> <span class="mi">20</span>
                    <span class="c"># let 1M &lt; avg &lt; 10M</span>
                    <span class="k">if</span> <span class="n">avg</span> <span class="o">&lt;</span> <span class="mi">1</span><span class="p">:</span>
                        <span class="n">batch</span> <span class="o">*=</span> <span class="mf">1.5</span>
                    <span class="k">elif</span> <span class="n">avg</span> <span class="o">&gt;</span> <span class="mi">10</span><span class="p">:</span>
                        <span class="n">batch</span> <span class="o">=</span> <span class="nb">max</span><span class="p">(</span><span class="nb">int</span><span class="p">(</span><span class="n">batch</span> <span class="o">/</span> <span class="mf">1.5</span><span class="p">),</span> <span class="mi">1</span><span class="p">)</span>
                    <span class="n">c</span> <span class="o">=</span> <span class="mi">0</span>

            <span class="k">for</span> <span class="n">split</span><span class="p">,</span> <span class="n">items</span> <span class="ow">in</span> <span class="n">buckets</span><span class="o">.</span><span class="n">items</span><span class="p">():</span>
                <span class="k">yield</span> <span class="n">pack_long</span><span class="p">(</span><span class="n">split</span><span class="p">)</span>
                <span class="k">yield</span> <span class="n">outputSerializer</span><span class="o">.</span><span class="n">dumps</span><span class="p">(</span><span class="n">items</span><span class="p">)</span>

        <span class="n">keyed</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">mapPartitionsWithIndex</span><span class="p">(</span><span class="n">add_shuffle_key</span><span class="p">,</span> <span class="n">preservesPartitioning</span><span class="o">=</span><span class="bp">True</span><span class="p">)</span>
        <span class="n">keyed</span><span class="o">.</span><span class="n">_bypass_serializer</span> <span class="o">=</span> <span class="bp">True</span>
        <span class="k">with</span> <span class="n">SCCallSiteSync</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">context</span><span class="p">)</span> <span class="k">as</span> <span class="n">css</span><span class="p">:</span>
            <span class="n">pairRDD</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="o">.</span><span class="n">_jvm</span><span class="o">.</span><span class="n">PairwiseRDD</span><span class="p">(</span>
                <span class="n">keyed</span><span class="o">.</span><span class="n">_jrdd</span><span class="o">.</span><span class="n">rdd</span><span class="p">())</span><span class="o">.</span><span class="n">asJavaPairRDD</span><span class="p">()</span>
            <span class="n">jpartitioner</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="o">.</span><span class="n">_jvm</span><span class="o">.</span><span class="n">PythonPartitioner</span><span class="p">(</span><span class="n">numPartitions</span><span class="p">,</span>
                                                           <span class="nb">id</span><span class="p">(</span><span class="n">partitionFunc</span><span class="p">))</span>
        <span class="n">jrdd</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="o">.</span><span class="n">_jvm</span><span class="o">.</span><span class="n">PythonRDD</span><span class="o">.</span><span class="n">valueOfPair</span><span class="p">(</span><span class="n">pairRDD</span><span class="o">.</span><span class="n">partitionBy</span><span class="p">(</span><span class="n">jpartitioner</span><span class="p">))</span>
        <span class="n">rdd</span> <span class="o">=</span> <span class="n">RDD</span><span class="p">(</span><span class="n">jrdd</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="p">,</span> <span class="n">BatchedSerializer</span><span class="p">(</span><span class="n">outputSerializer</span><span class="p">))</span>
        <span class="n">rdd</span><span class="o">.</span><span class="n">partitioner</span> <span class="o">=</span> <span class="n">partitioner</span>
        <span class="k">return</span> <span class="n">rdd</span>

    <span class="c"># TODO: add control over map-side aggregation</span></div>
<div class="viewcode-block" id="RDD.combineByKey"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.combineByKey">[docs]</a>    <span class="k">def</span> <span class="nf">combineByKey</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">createCombiner</span><span class="p">,</span> <span class="n">mergeValue</span><span class="p">,</span> <span class="n">mergeCombiners</span><span class="p">,</span>
                     <span class="n">numPartitions</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Generic function to combine the elements for each key using a custom</span>
<span class="sd">        set of aggregation functions.</span>

<span class="sd">        Turns an RDD[(K, V)] into a result of type RDD[(K, C)], for a &quot;combined</span>
<span class="sd">        type&quot; C.  Note that V and C can be different -- for example, one might</span>
<span class="sd">        group an RDD of type (Int, Int) into an RDD of type (Int, List[Int]).</span>

<span class="sd">        Users provide three functions:</span>

<span class="sd">            - C{createCombiner}, which turns a V into a C (e.g., creates</span>
<span class="sd">              a one-element list)</span>
<span class="sd">            - C{mergeValue}, to merge a V into a C (e.g., adds it to the end of</span>
<span class="sd">              a list)</span>
<span class="sd">            - C{mergeCombiners}, to combine two C&#39;s into a single one.</span>

<span class="sd">        In addition, users can control the partitioning of the output RDD.</span>

<span class="sd">        &gt;&gt;&gt; x = sc.parallelize([(&quot;a&quot;, 1), (&quot;b&quot;, 1), (&quot;a&quot;, 1)])</span>
<span class="sd">        &gt;&gt;&gt; def f(x): return x</span>
<span class="sd">        &gt;&gt;&gt; def add(a, b): return a + str(b)</span>
<span class="sd">        &gt;&gt;&gt; sorted(x.combineByKey(str, add, add).collect())</span>
<span class="sd">        [(&#39;a&#39;, &#39;11&#39;), (&#39;b&#39;, &#39;1&#39;)]</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">if</span> <span class="n">numPartitions</span> <span class="ow">is</span> <span class="bp">None</span><span class="p">:</span>
            <span class="n">numPartitions</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_defaultReducePartitions</span><span class="p">()</span>

        <span class="n">serializer</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="o">.</span><span class="n">serializer</span>
        <span class="n">spill</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_can_spill</span><span class="p">()</span>
        <span class="n">memory</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_memory_limit</span><span class="p">()</span>
        <span class="n">agg</span> <span class="o">=</span> <span class="n">Aggregator</span><span class="p">(</span><span class="n">createCombiner</span><span class="p">,</span> <span class="n">mergeValue</span><span class="p">,</span> <span class="n">mergeCombiners</span><span class="p">)</span>

        <span class="k">def</span> <span class="nf">combineLocally</span><span class="p">(</span><span class="n">iterator</span><span class="p">):</span>
            <span class="n">merger</span> <span class="o">=</span> <span class="n">ExternalMerger</span><span class="p">(</span><span class="n">agg</span><span class="p">,</span> <span class="n">memory</span> <span class="o">*</span> <span class="mf">0.9</span><span class="p">,</span> <span class="n">serializer</span><span class="p">)</span> \
                <span class="k">if</span> <span class="n">spill</span> <span class="k">else</span> <span class="n">InMemoryMerger</span><span class="p">(</span><span class="n">agg</span><span class="p">)</span>
            <span class="n">merger</span><span class="o">.</span><span class="n">mergeValues</span><span class="p">(</span><span class="n">iterator</span><span class="p">)</span>
            <span class="k">return</span> <span class="n">merger</span><span class="o">.</span><span class="n">items</span><span class="p">()</span>

        <span class="n">locally_combined</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">mapPartitions</span><span class="p">(</span><span class="n">combineLocally</span><span class="p">,</span> <span class="n">preservesPartitioning</span><span class="o">=</span><span class="bp">True</span><span class="p">)</span>
        <span class="n">shuffled</span> <span class="o">=</span> <span class="n">locally_combined</span><span class="o">.</span><span class="n">partitionBy</span><span class="p">(</span><span class="n">numPartitions</span><span class="p">)</span>

        <span class="k">def</span> <span class="nf">_mergeCombiners</span><span class="p">(</span><span class="n">iterator</span><span class="p">):</span>
            <span class="n">merger</span> <span class="o">=</span> <span class="n">ExternalMerger</span><span class="p">(</span><span class="n">agg</span><span class="p">,</span> <span class="n">memory</span><span class="p">,</span> <span class="n">serializer</span><span class="p">)</span> \
                <span class="k">if</span> <span class="n">spill</span> <span class="k">else</span> <span class="n">InMemoryMerger</span><span class="p">(</span><span class="n">agg</span><span class="p">)</span>
            <span class="n">merger</span><span class="o">.</span><span class="n">mergeCombiners</span><span class="p">(</span><span class="n">iterator</span><span class="p">)</span>
            <span class="k">return</span> <span class="n">merger</span><span class="o">.</span><span class="n">items</span><span class="p">()</span>

        <span class="k">return</span> <span class="n">shuffled</span><span class="o">.</span><span class="n">mapPartitions</span><span class="p">(</span><span class="n">_mergeCombiners</span><span class="p">,</span> <span class="n">preservesPartitioning</span><span class="o">=</span><span class="bp">True</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.aggregateByKey"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.aggregateByKey">[docs]</a>    <span class="k">def</span> <span class="nf">aggregateByKey</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">zeroValue</span><span class="p">,</span> <span class="n">seqFunc</span><span class="p">,</span> <span class="n">combFunc</span><span class="p">,</span> <span class="n">numPartitions</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Aggregate the values of each key, using given combine functions and a neutral</span>
<span class="sd">        &quot;zero value&quot;. This function can return a different result type, U, than the type</span>
<span class="sd">        of the values in this RDD, V. Thus, we need one operation for merging a V into</span>
<span class="sd">        a U and one operation for merging two U&#39;s, The former operation is used for merging</span>
<span class="sd">        values within a partition, and the latter is used for merging values between</span>
<span class="sd">        partitions. To avoid memory allocation, both of these functions are</span>
<span class="sd">        allowed to modify and return their first argument instead of creating a new U.</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">def</span> <span class="nf">createZero</span><span class="p">():</span>
            <span class="k">return</span> <span class="n">copy</span><span class="o">.</span><span class="n">deepcopy</span><span class="p">(</span><span class="n">zeroValue</span><span class="p">)</span>

        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">combineByKey</span><span class="p">(</span>
            <span class="k">lambda</span> <span class="n">v</span><span class="p">:</span> <span class="n">seqFunc</span><span class="p">(</span><span class="n">createZero</span><span class="p">(),</span> <span class="n">v</span><span class="p">),</span> <span class="n">seqFunc</span><span class="p">,</span> <span class="n">combFunc</span><span class="p">,</span> <span class="n">numPartitions</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.foldByKey"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.foldByKey">[docs]</a>    <span class="k">def</span> <span class="nf">foldByKey</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">zeroValue</span><span class="p">,</span> <span class="n">func</span><span class="p">,</span> <span class="n">numPartitions</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Merge the values for each key using an associative function &quot;func&quot;</span>
<span class="sd">        and a neutral &quot;zeroValue&quot; which may be added to the result an</span>
<span class="sd">        arbitrary number of times, and must not change the result</span>
<span class="sd">        (e.g., 0 for addition, or 1 for multiplication.).</span>

<span class="sd">        &gt;&gt;&gt; rdd = sc.parallelize([(&quot;a&quot;, 1), (&quot;b&quot;, 1), (&quot;a&quot;, 1)])</span>
<span class="sd">        &gt;&gt;&gt; from operator import add</span>
<span class="sd">        &gt;&gt;&gt; sorted(rdd.foldByKey(0, add).collect())</span>
<span class="sd">        [(&#39;a&#39;, 2), (&#39;b&#39;, 1)]</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">def</span> <span class="nf">createZero</span><span class="p">():</span>
            <span class="k">return</span> <span class="n">copy</span><span class="o">.</span><span class="n">deepcopy</span><span class="p">(</span><span class="n">zeroValue</span><span class="p">)</span>

        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">combineByKey</span><span class="p">(</span><span class="k">lambda</span> <span class="n">v</span><span class="p">:</span> <span class="n">func</span><span class="p">(</span><span class="n">createZero</span><span class="p">(),</span> <span class="n">v</span><span class="p">),</span> <span class="n">func</span><span class="p">,</span> <span class="n">func</span><span class="p">,</span> <span class="n">numPartitions</span><span class="p">)</span>
</div>
    <span class="k">def</span> <span class="nf">_can_spill</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="o">.</span><span class="n">_conf</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s">&quot;spark.shuffle.spill&quot;</span><span class="p">,</span> <span class="s">&quot;True&quot;</span><span class="p">)</span><span class="o">.</span><span class="n">lower</span><span class="p">()</span> <span class="o">==</span> <span class="s">&quot;true&quot;</span>

    <span class="k">def</span> <span class="nf">_memory_limit</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="k">return</span> <span class="n">_parse_memory</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="o">.</span><span class="n">_conf</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s">&quot;spark.python.worker.memory&quot;</span><span class="p">,</span> <span class="s">&quot;512m&quot;</span><span class="p">))</span>

    <span class="c"># TODO: support variant with custom partitioner</span>
<div class="viewcode-block" id="RDD.groupByKey"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.groupByKey">[docs]</a>    <span class="k">def</span> <span class="nf">groupByKey</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">numPartitions</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Group the values for each key in the RDD into a single sequence.</span>
<span class="sd">        Hash-partitions the resulting RDD with numPartitions partitions.</span>

<span class="sd">        Note: If you are grouping in order to perform an aggregation (such as a</span>
<span class="sd">        sum or average) over each key, using reduceByKey or aggregateByKey will</span>
<span class="sd">        provide much better performance.</span>

<span class="sd">        &gt;&gt;&gt; rdd = sc.parallelize([(&quot;a&quot;, 1), (&quot;b&quot;, 1), (&quot;a&quot;, 1)])</span>
<span class="sd">        &gt;&gt;&gt; sorted(rdd.groupByKey().mapValues(len).collect())</span>
<span class="sd">        [(&#39;a&#39;, 2), (&#39;b&#39;, 1)]</span>
<span class="sd">        &gt;&gt;&gt; sorted(rdd.groupByKey().mapValues(list).collect())</span>
<span class="sd">        [(&#39;a&#39;, [1, 1]), (&#39;b&#39;, [1])]</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">def</span> <span class="nf">createCombiner</span><span class="p">(</span><span class="n">x</span><span class="p">):</span>
            <span class="k">return</span> <span class="p">[</span><span class="n">x</span><span class="p">]</span>

        <span class="k">def</span> <span class="nf">mergeValue</span><span class="p">(</span><span class="n">xs</span><span class="p">,</span> <span class="n">x</span><span class="p">):</span>
            <span class="n">xs</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
            <span class="k">return</span> <span class="n">xs</span>

        <span class="k">def</span> <span class="nf">mergeCombiners</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">):</span>
            <span class="n">a</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span><span class="n">b</span><span class="p">)</span>
            <span class="k">return</span> <span class="n">a</span>

        <span class="n">spill</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_can_spill</span><span class="p">()</span>
        <span class="n">memory</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_memory_limit</span><span class="p">()</span>
        <span class="n">serializer</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_jrdd_deserializer</span>
        <span class="n">agg</span> <span class="o">=</span> <span class="n">Aggregator</span><span class="p">(</span><span class="n">createCombiner</span><span class="p">,</span> <span class="n">mergeValue</span><span class="p">,</span> <span class="n">mergeCombiners</span><span class="p">)</span>

        <span class="k">def</span> <span class="nf">combine</span><span class="p">(</span><span class="n">iterator</span><span class="p">):</span>
            <span class="n">merger</span> <span class="o">=</span> <span class="n">ExternalMerger</span><span class="p">(</span><span class="n">agg</span><span class="p">,</span> <span class="n">memory</span> <span class="o">*</span> <span class="mf">0.9</span><span class="p">,</span> <span class="n">serializer</span><span class="p">)</span> \
                <span class="k">if</span> <span class="n">spill</span> <span class="k">else</span> <span class="n">InMemoryMerger</span><span class="p">(</span><span class="n">agg</span><span class="p">)</span>
            <span class="n">merger</span><span class="o">.</span><span class="n">mergeValues</span><span class="p">(</span><span class="n">iterator</span><span class="p">)</span>
            <span class="k">return</span> <span class="n">merger</span><span class="o">.</span><span class="n">items</span><span class="p">()</span>

        <span class="n">locally_combined</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">mapPartitions</span><span class="p">(</span><span class="n">combine</span><span class="p">,</span> <span class="n">preservesPartitioning</span><span class="o">=</span><span class="bp">True</span><span class="p">)</span>
        <span class="n">shuffled</span> <span class="o">=</span> <span class="n">locally_combined</span><span class="o">.</span><span class="n">partitionBy</span><span class="p">(</span><span class="n">numPartitions</span><span class="p">)</span>

        <span class="k">def</span> <span class="nf">groupByKey</span><span class="p">(</span><span class="n">it</span><span class="p">):</span>
            <span class="n">merger</span> <span class="o">=</span> <span class="n">ExternalGroupBy</span><span class="p">(</span><span class="n">agg</span><span class="p">,</span> <span class="n">memory</span><span class="p">,</span> <span class="n">serializer</span><span class="p">)</span>\
                <span class="k">if</span> <span class="n">spill</span> <span class="k">else</span> <span class="n">InMemoryMerger</span><span class="p">(</span><span class="n">agg</span><span class="p">)</span>
            <span class="n">merger</span><span class="o">.</span><span class="n">mergeCombiners</span><span class="p">(</span><span class="n">it</span><span class="p">)</span>
            <span class="k">return</span> <span class="n">merger</span><span class="o">.</span><span class="n">items</span><span class="p">()</span>

        <span class="k">return</span> <span class="n">shuffled</span><span class="o">.</span><span class="n">mapPartitions</span><span class="p">(</span><span class="n">groupByKey</span><span class="p">,</span> <span class="bp">True</span><span class="p">)</span><span class="o">.</span><span class="n">mapValues</span><span class="p">(</span><span class="n">ResultIterable</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.flatMapValues"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.flatMapValues">[docs]</a>    <span class="k">def</span> <span class="nf">flatMapValues</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">f</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Pass each value in the key-value pair RDD through a flatMap function</span>
<span class="sd">        without changing the keys; this also retains the original RDD&#39;s</span>
<span class="sd">        partitioning.</span>

<span class="sd">        &gt;&gt;&gt; x = sc.parallelize([(&quot;a&quot;, [&quot;x&quot;, &quot;y&quot;, &quot;z&quot;]), (&quot;b&quot;, [&quot;p&quot;, &quot;r&quot;])])</span>
<span class="sd">        &gt;&gt;&gt; def f(x): return x</span>
<span class="sd">        &gt;&gt;&gt; x.flatMapValues(f).collect()</span>
<span class="sd">        [(&#39;a&#39;, &#39;x&#39;), (&#39;a&#39;, &#39;y&#39;), (&#39;a&#39;, &#39;z&#39;), (&#39;b&#39;, &#39;p&#39;), (&#39;b&#39;, &#39;r&#39;)]</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="n">flat_map_fn</span> <span class="o">=</span> <span class="k">lambda</span> <span class="n">kv</span><span class="p">:</span> <span class="p">((</span><span class="n">kv</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">x</span><span class="p">)</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">f</span><span class="p">(</span><span class="n">kv</span><span class="p">[</span><span class="mi">1</span><span class="p">]))</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">flatMap</span><span class="p">(</span><span class="n">flat_map_fn</span><span class="p">,</span> <span class="n">preservesPartitioning</span><span class="o">=</span><span class="bp">True</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.mapValues"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.mapValues">[docs]</a>    <span class="k">def</span> <span class="nf">mapValues</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">f</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Pass each value in the key-value pair RDD through a map function</span>
<span class="sd">        without changing the keys; this also retains the original RDD&#39;s</span>
<span class="sd">        partitioning.</span>

<span class="sd">        &gt;&gt;&gt; x = sc.parallelize([(&quot;a&quot;, [&quot;apple&quot;, &quot;banana&quot;, &quot;lemon&quot;]), (&quot;b&quot;, [&quot;grapes&quot;])])</span>
<span class="sd">        &gt;&gt;&gt; def f(x): return len(x)</span>
<span class="sd">        &gt;&gt;&gt; x.mapValues(f).collect()</span>
<span class="sd">        [(&#39;a&#39;, 3), (&#39;b&#39;, 1)]</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="n">map_values_fn</span> <span class="o">=</span> <span class="k">lambda</span> <span class="n">kv</span><span class="p">:</span> <span class="p">(</span><span class="n">kv</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">f</span><span class="p">(</span><span class="n">kv</span><span class="p">[</span><span class="mi">1</span><span class="p">]))</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">map</span><span class="p">(</span><span class="n">map_values_fn</span><span class="p">,</span> <span class="n">preservesPartitioning</span><span class="o">=</span><span class="bp">True</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.groupWith"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.groupWith">[docs]</a>    <span class="k">def</span> <span class="nf">groupWith</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">other</span><span class="p">,</span> <span class="o">*</span><span class="n">others</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Alias for cogroup but with support for multiple RDDs.</span>

<span class="sd">        &gt;&gt;&gt; w = sc.parallelize([(&quot;a&quot;, 5), (&quot;b&quot;, 6)])</span>
<span class="sd">        &gt;&gt;&gt; x = sc.parallelize([(&quot;a&quot;, 1), (&quot;b&quot;, 4)])</span>
<span class="sd">        &gt;&gt;&gt; y = sc.parallelize([(&quot;a&quot;, 2)])</span>
<span class="sd">        &gt;&gt;&gt; z = sc.parallelize([(&quot;b&quot;, 42)])</span>
<span class="sd">        &gt;&gt;&gt; [(x, tuple(map(list, y))) for x, y in sorted(list(w.groupWith(x, y, z).collect()))]</span>
<span class="sd">        [(&#39;a&#39;, ([5], [1], [2], [])), (&#39;b&#39;, ([6], [4], [], [42]))]</span>

<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">return</span> <span class="n">python_cogroup</span><span class="p">((</span><span class="bp">self</span><span class="p">,</span> <span class="n">other</span><span class="p">)</span> <span class="o">+</span> <span class="n">others</span><span class="p">,</span> <span class="n">numPartitions</span><span class="o">=</span><span class="bp">None</span><span class="p">)</span>

    <span class="c"># TODO: add variant with custom parittioner</span></div>
<div class="viewcode-block" id="RDD.cogroup"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.cogroup">[docs]</a>    <span class="k">def</span> <span class="nf">cogroup</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">other</span><span class="p">,</span> <span class="n">numPartitions</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        For each key k in C{self} or C{other}, return a resulting RDD that</span>
<span class="sd">        contains a tuple with the list of values for that key in C{self} as</span>
<span class="sd">        well as C{other}.</span>

<span class="sd">        &gt;&gt;&gt; x = sc.parallelize([(&quot;a&quot;, 1), (&quot;b&quot;, 4)])</span>
<span class="sd">        &gt;&gt;&gt; y = sc.parallelize([(&quot;a&quot;, 2)])</span>
<span class="sd">        &gt;&gt;&gt; [(x, tuple(map(list, y))) for x, y in sorted(list(x.cogroup(y).collect()))]</span>
<span class="sd">        [(&#39;a&#39;, ([1], [2])), (&#39;b&#39;, ([4], []))]</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">return</span> <span class="n">python_cogroup</span><span class="p">((</span><span class="bp">self</span><span class="p">,</span> <span class="n">other</span><span class="p">),</span> <span class="n">numPartitions</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.sampleByKey"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.sampleByKey">[docs]</a>    <span class="k">def</span> <span class="nf">sampleByKey</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">withReplacement</span><span class="p">,</span> <span class="n">fractions</span><span class="p">,</span> <span class="n">seed</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Return a subset of this RDD sampled by key (via stratified sampling).</span>
<span class="sd">        Create a sample of this RDD using variable sampling rates for</span>
<span class="sd">        different keys as specified by fractions, a key to sampling rate map.</span>

<span class="sd">        &gt;&gt;&gt; fractions = {&quot;a&quot;: 0.2, &quot;b&quot;: 0.1}</span>
<span class="sd">        &gt;&gt;&gt; rdd = sc.parallelize(fractions.keys()).cartesian(sc.parallelize(range(0, 1000)))</span>
<span class="sd">        &gt;&gt;&gt; sample = dict(rdd.sampleByKey(False, fractions, 2).groupByKey().collect())</span>
<span class="sd">        &gt;&gt;&gt; 100 &lt; len(sample[&quot;a&quot;]) &lt; 300 and 50 &lt; len(sample[&quot;b&quot;]) &lt; 150</span>
<span class="sd">        True</span>
<span class="sd">        &gt;&gt;&gt; max(sample[&quot;a&quot;]) &lt;= 999 and min(sample[&quot;a&quot;]) &gt;= 0</span>
<span class="sd">        True</span>
<span class="sd">        &gt;&gt;&gt; max(sample[&quot;b&quot;]) &lt;= 999 and min(sample[&quot;b&quot;]) &gt;= 0</span>
<span class="sd">        True</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">for</span> <span class="n">fraction</span> <span class="ow">in</span> <span class="n">fractions</span><span class="o">.</span><span class="n">values</span><span class="p">():</span>
            <span class="k">assert</span> <span class="n">fraction</span> <span class="o">&gt;=</span> <span class="mf">0.0</span><span class="p">,</span> <span class="s">&quot;Negative fraction value: </span><span class="si">%s</span><span class="s">&quot;</span> <span class="o">%</span> <span class="n">fraction</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">mapPartitionsWithIndex</span><span class="p">(</span>
            <span class="n">RDDStratifiedSampler</span><span class="p">(</span><span class="n">withReplacement</span><span class="p">,</span> <span class="n">fractions</span><span class="p">,</span> <span class="n">seed</span><span class="p">)</span><span class="o">.</span><span class="n">func</span><span class="p">,</span> <span class="bp">True</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.subtractByKey"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.subtractByKey">[docs]</a>    <span class="k">def</span> <span class="nf">subtractByKey</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">other</span><span class="p">,</span> <span class="n">numPartitions</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Return each (key, value) pair in C{self} that has no pair with matching</span>
<span class="sd">        key in C{other}.</span>

<span class="sd">        &gt;&gt;&gt; x = sc.parallelize([(&quot;a&quot;, 1), (&quot;b&quot;, 4), (&quot;b&quot;, 5), (&quot;a&quot;, 2)])</span>
<span class="sd">        &gt;&gt;&gt; y = sc.parallelize([(&quot;a&quot;, 3), (&quot;c&quot;, None)])</span>
<span class="sd">        &gt;&gt;&gt; sorted(x.subtractByKey(y).collect())</span>
<span class="sd">        [(&#39;b&#39;, 4), (&#39;b&#39;, 5)]</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">def</span> <span class="nf">filter_func</span><span class="p">(</span><span class="n">pair</span><span class="p">):</span>
            <span class="n">key</span><span class="p">,</span> <span class="p">(</span><span class="n">val1</span><span class="p">,</span> <span class="n">val2</span><span class="p">)</span> <span class="o">=</span> <span class="n">pair</span>
            <span class="k">return</span> <span class="n">val1</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">val2</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">cogroup</span><span class="p">(</span><span class="n">other</span><span class="p">,</span> <span class="n">numPartitions</span><span class="p">)</span><span class="o">.</span><span class="n">filter</span><span class="p">(</span><span class="n">filter_func</span><span class="p">)</span><span class="o">.</span><span class="n">flatMapValues</span><span class="p">(</span><span class="k">lambda</span> <span class="n">x</span><span class="p">:</span> <span class="n">x</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span>
</div>
<div class="viewcode-block" id="RDD.subtract"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.subtract">[docs]</a>    <span class="k">def</span> <span class="nf">subtract</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">other</span><span class="p">,</span> <span class="n">numPartitions</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Return each value in C{self} that is not contained in C{other}.</span>

<span class="sd">        &gt;&gt;&gt; x = sc.parallelize([(&quot;a&quot;, 1), (&quot;b&quot;, 4), (&quot;b&quot;, 5), (&quot;a&quot;, 3)])</span>
<span class="sd">        &gt;&gt;&gt; y = sc.parallelize([(&quot;a&quot;, 3), (&quot;c&quot;, None)])</span>
<span class="sd">        &gt;&gt;&gt; sorted(x.subtract(y).collect())</span>
<span class="sd">        [(&#39;a&#39;, 1), (&#39;b&#39;, 4), (&#39;b&#39;, 5)]</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="c"># note: here &#39;True&#39; is just a placeholder</span>
        <span class="n">rdd</span> <span class="o">=</span> <span class="n">other</span><span class="o">.</span><span class="n">map</span><span class="p">(</span><span class="k">lambda</span> <span class="n">x</span><span class="p">:</span> <span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="bp">True</span><span class="p">))</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">map</span><span class="p">(</span><span class="k">lambda</span> <span class="n">x</span><span class="p">:</span> <span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="bp">True</span><span class="p">))</span><span class="o">.</span><span class="n">subtractByKey</span><span class="p">(</span><span class="n">rdd</span><span class="p">,</span> <span class="n">numPartitions</span><span class="p">)</span><span class="o">.</span><span class="n">keys</span><span class="p">()</span>
</div>
<div class="viewcode-block" id="RDD.keyBy"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.keyBy">[docs]</a>    <span class="k">def</span> <span class="nf">keyBy</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">f</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Creates tuples of the elements in this RDD by applying C{f}.</span>

<span class="sd">        &gt;&gt;&gt; x = sc.parallelize(range(0,3)).keyBy(lambda x: x*x)</span>
<span class="sd">        &gt;&gt;&gt; y = sc.parallelize(zip(range(0,5), range(0,5)))</span>
<span class="sd">        &gt;&gt;&gt; [(x, list(map(list, y))) for x, y in sorted(x.cogroup(y).collect())]</span>
<span class="sd">        [(0, [[0], [0]]), (1, [[1], [1]]), (2, [[], [2]]), (3, [[], [3]]), (4, [[2], [4]])]</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">map</span><span class="p">(</span><span class="k">lambda</span> <span class="n">x</span><span class="p">:</span> <span class="p">(</span><span class="n">f</span><span class="p">(</span><span class="n">x</span><span class="p">),</span> <span class="n">x</span><span class="p">))</span>
</div>
<div class="viewcode-block" id="RDD.repartition"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.repartition">[docs]</a>    <span class="k">def</span> <span class="nf">repartition</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">numPartitions</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">         Return a new RDD that has exactly numPartitions partitions.</span>

<span class="sd">         Can increase or decrease the level of parallelism in this RDD.</span>
<span class="sd">         Internally, this uses a shuffle to redistribute data.</span>
<span class="sd">         If you are decreasing the number of partitions in this RDD, consider</span>
<span class="sd">         using `coalesce`, which can avoid performing a shuffle.</span>

<span class="sd">         &gt;&gt;&gt; rdd = sc.parallelize([1,2,3,4,5,6,7], 4)</span>
<span class="sd">         &gt;&gt;&gt; sorted(rdd.glom().collect())</span>
<span class="sd">         [[1], [2, 3], [4, 5], [6, 7]]</span>
<span class="sd">         &gt;&gt;&gt; len(rdd.repartition(2).glom().collect())</span>
<span class="sd">         2</span>
<span class="sd">         &gt;&gt;&gt; len(rdd.repartition(10).glom().collect())</span>
<span class="sd">         10</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="n">jrdd</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_jrdd</span><span class="o">.</span><span class="n">repartition</span><span class="p">(</span><span class="n">numPartitions</span><span class="p">)</span>
        <span class="k">return</span> <span class="n">RDD</span><span class="p">(</span><span class="n">jrdd</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_jrdd_deserializer</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.coalesce"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.coalesce">[docs]</a>    <span class="k">def</span> <span class="nf">coalesce</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">numPartitions</span><span class="p">,</span> <span class="n">shuffle</span><span class="o">=</span><span class="bp">False</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Return a new RDD that is reduced into `numPartitions` partitions.</span>

<span class="sd">        &gt;&gt;&gt; sc.parallelize([1, 2, 3, 4, 5], 3).glom().collect()</span>
<span class="sd">        [[1], [2, 3], [4, 5]]</span>
<span class="sd">        &gt;&gt;&gt; sc.parallelize([1, 2, 3, 4, 5], 3).coalesce(1).glom().collect()</span>
<span class="sd">        [[1, 2, 3, 4, 5]]</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="n">jrdd</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_jrdd</span><span class="o">.</span><span class="n">coalesce</span><span class="p">(</span><span class="n">numPartitions</span><span class="p">)</span>
        <span class="k">return</span> <span class="n">RDD</span><span class="p">(</span><span class="n">jrdd</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_jrdd_deserializer</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.zip"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.zip">[docs]</a>    <span class="k">def</span> <span class="nf">zip</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">other</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Zips this RDD with another one, returning key-value pairs with the</span>
<span class="sd">        first element in each RDD second element in each RDD, etc. Assumes</span>
<span class="sd">        that the two RDDs have the same number of partitions and the same</span>
<span class="sd">        number of elements in each partition (e.g. one was made through</span>
<span class="sd">        a map on the other).</span>

<span class="sd">        &gt;&gt;&gt; x = sc.parallelize(range(0,5))</span>
<span class="sd">        &gt;&gt;&gt; y = sc.parallelize(range(1000, 1005))</span>
<span class="sd">        &gt;&gt;&gt; x.zip(y).collect()</span>
<span class="sd">        [(0, 1000), (1, 1001), (2, 1002), (3, 1003), (4, 1004)]</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">def</span> <span class="nf">get_batch_size</span><span class="p">(</span><span class="n">ser</span><span class="p">):</span>
            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">ser</span><span class="p">,</span> <span class="n">BatchedSerializer</span><span class="p">):</span>
                <span class="k">return</span> <span class="n">ser</span><span class="o">.</span><span class="n">batchSize</span>
            <span class="k">return</span> <span class="mi">1</span>  <span class="c"># not batched</span>

        <span class="k">def</span> <span class="nf">batch_as</span><span class="p">(</span><span class="n">rdd</span><span class="p">,</span> <span class="n">batchSize</span><span class="p">):</span>
            <span class="k">return</span> <span class="n">rdd</span><span class="o">.</span><span class="n">_reserialize</span><span class="p">(</span><span class="n">BatchedSerializer</span><span class="p">(</span><span class="n">PickleSerializer</span><span class="p">(),</span> <span class="n">batchSize</span><span class="p">))</span>

        <span class="n">my_batch</span> <span class="o">=</span> <span class="n">get_batch_size</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_jrdd_deserializer</span><span class="p">)</span>
        <span class="n">other_batch</span> <span class="o">=</span> <span class="n">get_batch_size</span><span class="p">(</span><span class="n">other</span><span class="o">.</span><span class="n">_jrdd_deserializer</span><span class="p">)</span>
        <span class="k">if</span> <span class="n">my_batch</span> <span class="o">!=</span> <span class="n">other_batch</span> <span class="ow">or</span> <span class="ow">not</span> <span class="n">my_batch</span><span class="p">:</span>
            <span class="c"># use the smallest batchSize for both of them</span>
            <span class="n">batchSize</span> <span class="o">=</span> <span class="nb">min</span><span class="p">(</span><span class="n">my_batch</span><span class="p">,</span> <span class="n">other_batch</span><span class="p">)</span>
            <span class="k">if</span> <span class="n">batchSize</span> <span class="o">&lt;=</span> <span class="mi">0</span><span class="p">:</span>
                <span class="c"># auto batched or unlimited</span>
                <span class="n">batchSize</span> <span class="o">=</span> <span class="mi">100</span>
            <span class="n">other</span> <span class="o">=</span> <span class="n">batch_as</span><span class="p">(</span><span class="n">other</span><span class="p">,</span> <span class="n">batchSize</span><span class="p">)</span>
            <span class="bp">self</span> <span class="o">=</span> <span class="n">batch_as</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">batchSize</span><span class="p">)</span>

        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">getNumPartitions</span><span class="p">()</span> <span class="o">!=</span> <span class="n">other</span><span class="o">.</span><span class="n">getNumPartitions</span><span class="p">():</span>
            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s">&quot;Can only zip with RDD which has the same number of partitions&quot;</span><span class="p">)</span>

        <span class="c"># There will be an Exception in JVM if there are different number</span>
        <span class="c"># of items in each partitions.</span>
        <span class="n">pairRDD</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_jrdd</span><span class="o">.</span><span class="n">zip</span><span class="p">(</span><span class="n">other</span><span class="o">.</span><span class="n">_jrdd</span><span class="p">)</span>
        <span class="n">deserializer</span> <span class="o">=</span> <span class="n">PairDeserializer</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_jrdd_deserializer</span><span class="p">,</span>
                                        <span class="n">other</span><span class="o">.</span><span class="n">_jrdd_deserializer</span><span class="p">)</span>
        <span class="k">return</span> <span class="n">RDD</span><span class="p">(</span><span class="n">pairRDD</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="p">,</span> <span class="n">deserializer</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.zipWithIndex"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.zipWithIndex">[docs]</a>    <span class="k">def</span> <span class="nf">zipWithIndex</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Zips this RDD with its element indices.</span>

<span class="sd">        The ordering is first based on the partition index and then the</span>
<span class="sd">        ordering of items within each partition. So the first item in</span>
<span class="sd">        the first partition gets index 0, and the last item in the last</span>
<span class="sd">        partition receives the largest index.</span>

<span class="sd">        This method needs to trigger a spark job when this RDD contains</span>
<span class="sd">        more than one partitions.</span>

<span class="sd">        &gt;&gt;&gt; sc.parallelize([&quot;a&quot;, &quot;b&quot;, &quot;c&quot;, &quot;d&quot;], 3).zipWithIndex().collect()</span>
<span class="sd">        [(&#39;a&#39;, 0), (&#39;b&#39;, 1), (&#39;c&#39;, 2), (&#39;d&#39;, 3)]</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="n">starts</span> <span class="o">=</span> <span class="p">[</span><span class="mi">0</span><span class="p">]</span>
        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">getNumPartitions</span><span class="p">()</span> <span class="o">&gt;</span> <span class="mi">1</span><span class="p">:</span>
            <span class="n">nums</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">mapPartitions</span><span class="p">(</span><span class="k">lambda</span> <span class="n">it</span><span class="p">:</span> <span class="p">[</span><span class="nb">sum</span><span class="p">(</span><span class="mi">1</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="n">it</span><span class="p">)])</span><span class="o">.</span><span class="n">collect</span><span class="p">()</span>
            <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">nums</span><span class="p">)</span> <span class="o">-</span> <span class="mi">1</span><span class="p">):</span>
                <span class="n">starts</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">starts</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span> <span class="o">+</span> <span class="n">nums</span><span class="p">[</span><span class="n">i</span><span class="p">])</span>

        <span class="k">def</span> <span class="nf">func</span><span class="p">(</span><span class="n">k</span><span class="p">,</span> <span class="n">it</span><span class="p">):</span>
            <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">v</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">it</span><span class="p">,</span> <span class="n">starts</span><span class="p">[</span><span class="n">k</span><span class="p">]):</span>
                <span class="k">yield</span> <span class="n">v</span><span class="p">,</span> <span class="n">i</span>

        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">mapPartitionsWithIndex</span><span class="p">(</span><span class="n">func</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.zipWithUniqueId"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.zipWithUniqueId">[docs]</a>    <span class="k">def</span> <span class="nf">zipWithUniqueId</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Zips this RDD with generated unique Long ids.</span>

<span class="sd">        Items in the kth partition will get ids k, n+k, 2*n+k, ..., where</span>
<span class="sd">        n is the number of partitions. So there may exist gaps, but this</span>
<span class="sd">        method won&#39;t trigger a spark job, which is different from</span>
<span class="sd">        L{zipWithIndex}</span>

<span class="sd">        &gt;&gt;&gt; sc.parallelize([&quot;a&quot;, &quot;b&quot;, &quot;c&quot;, &quot;d&quot;, &quot;e&quot;], 3).zipWithUniqueId().collect()</span>
<span class="sd">        [(&#39;a&#39;, 0), (&#39;b&#39;, 1), (&#39;c&#39;, 4), (&#39;d&#39;, 2), (&#39;e&#39;, 5)]</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="n">n</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">getNumPartitions</span><span class="p">()</span>

        <span class="k">def</span> <span class="nf">func</span><span class="p">(</span><span class="n">k</span><span class="p">,</span> <span class="n">it</span><span class="p">):</span>
            <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">v</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">it</span><span class="p">):</span>
                <span class="k">yield</span> <span class="n">v</span><span class="p">,</span> <span class="n">i</span> <span class="o">*</span> <span class="n">n</span> <span class="o">+</span> <span class="n">k</span>

        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">mapPartitionsWithIndex</span><span class="p">(</span><span class="n">func</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.name"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.name">[docs]</a>    <span class="k">def</span> <span class="nf">name</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Return the name of this RDD.</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="n">n</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_jrdd</span><span class="o">.</span><span class="n">name</span><span class="p">()</span>
        <span class="k">if</span> <span class="n">n</span><span class="p">:</span>
            <span class="k">return</span> <span class="n">n</span>
</div>
    <span class="nd">@ignore_unicode_prefix</span>
<div class="viewcode-block" id="RDD.setName"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.setName">[docs]</a>    <span class="k">def</span> <span class="nf">setName</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">name</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Assign a name to this RDD.</span>

<span class="sd">        &gt;&gt;&gt; rdd1 = sc.parallelize([1, 2])</span>
<span class="sd">        &gt;&gt;&gt; rdd1.setName(&#39;RDD1&#39;).name()</span>
<span class="sd">        u&#39;RDD1&#39;</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">_jrdd</span><span class="o">.</span><span class="n">setName</span><span class="p">(</span><span class="n">name</span><span class="p">)</span>
        <span class="k">return</span> <span class="bp">self</span>
</div>
<div class="viewcode-block" id="RDD.toDebugString"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.toDebugString">[docs]</a>    <span class="k">def</span> <span class="nf">toDebugString</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        A description of this RDD and its recursive dependencies for debugging.</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="n">debug_string</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_jrdd</span><span class="o">.</span><span class="n">toDebugString</span><span class="p">()</span>
        <span class="k">if</span> <span class="n">debug_string</span><span class="p">:</span>
            <span class="k">return</span> <span class="n">debug_string</span><span class="o">.</span><span class="n">encode</span><span class="p">(</span><span class="s">&#39;utf-8&#39;</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.getStorageLevel"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.getStorageLevel">[docs]</a>    <span class="k">def</span> <span class="nf">getStorageLevel</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Get the RDD&#39;s current storage level.</span>

<span class="sd">        &gt;&gt;&gt; rdd1 = sc.parallelize([1,2])</span>
<span class="sd">        &gt;&gt;&gt; rdd1.getStorageLevel()</span>
<span class="sd">        StorageLevel(False, False, False, False, 1)</span>
<span class="sd">        &gt;&gt;&gt; print(rdd1.getStorageLevel())</span>
<span class="sd">        Serialized 1x Replicated</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="n">java_storage_level</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_jrdd</span><span class="o">.</span><span class="n">getStorageLevel</span><span class="p">()</span>
        <span class="n">storage_level</span> <span class="o">=</span> <span class="n">StorageLevel</span><span class="p">(</span><span class="n">java_storage_level</span><span class="o">.</span><span class="n">useDisk</span><span class="p">(),</span>
                                     <span class="n">java_storage_level</span><span class="o">.</span><span class="n">useMemory</span><span class="p">(),</span>
                                     <span class="n">java_storage_level</span><span class="o">.</span><span class="n">useOffHeap</span><span class="p">(),</span>
                                     <span class="n">java_storage_level</span><span class="o">.</span><span class="n">deserialized</span><span class="p">(),</span>
                                     <span class="n">java_storage_level</span><span class="o">.</span><span class="n">replication</span><span class="p">())</span>
        <span class="k">return</span> <span class="n">storage_level</span>
</div>
    <span class="k">def</span> <span class="nf">_defaultReducePartitions</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Returns the default number of partitions to use during reduce tasks (e.g., groupBy).</span>
<span class="sd">        If spark.default.parallelism is set, then we&#39;ll use the value from SparkContext</span>
<span class="sd">        defaultParallelism, otherwise we&#39;ll use the number of partitions in this RDD.</span>

<span class="sd">        This mirrors the behavior of the Scala Partitioner#defaultPartitioner, intended to reduce</span>
<span class="sd">        the likelihood of OOMs. Once PySpark adopts Partitioner-based APIs, this behavior will</span>
<span class="sd">        be inherent.</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="o">.</span><span class="n">_conf</span><span class="o">.</span><span class="n">contains</span><span class="p">(</span><span class="s">&quot;spark.default.parallelism&quot;</span><span class="p">):</span>
            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="o">.</span><span class="n">defaultParallelism</span>
        <span class="k">else</span><span class="p">:</span>
            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">getNumPartitions</span><span class="p">()</span>

<div class="viewcode-block" id="RDD.lookup"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.lookup">[docs]</a>    <span class="k">def</span> <span class="nf">lookup</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">key</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Return the list of values in the RDD for key `key`. This operation</span>
<span class="sd">        is done efficiently if the RDD has a known partitioner by only</span>
<span class="sd">        searching the partition that the key maps to.</span>

<span class="sd">        &gt;&gt;&gt; l = range(1000)</span>
<span class="sd">        &gt;&gt;&gt; rdd = sc.parallelize(zip(l, l), 10)</span>
<span class="sd">        &gt;&gt;&gt; rdd.lookup(42)  # slow</span>
<span class="sd">        [42]</span>
<span class="sd">        &gt;&gt;&gt; sorted = rdd.sortByKey()</span>
<span class="sd">        &gt;&gt;&gt; sorted.lookup(42)  # fast</span>
<span class="sd">        [42]</span>
<span class="sd">        &gt;&gt;&gt; sorted.lookup(1024)</span>
<span class="sd">        []</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="n">values</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">filter</span><span class="p">(</span><span class="k">lambda</span> <span class="n">kv</span><span class="p">:</span> <span class="n">kv</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">==</span> <span class="n">key</span><span class="p">)</span><span class="o">.</span><span class="n">values</span><span class="p">()</span>

        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">partitioner</span> <span class="ow">is</span> <span class="ow">not</span> <span class="bp">None</span><span class="p">:</span>
            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="o">.</span><span class="n">runJob</span><span class="p">(</span><span class="n">values</span><span class="p">,</span> <span class="k">lambda</span> <span class="n">x</span><span class="p">:</span> <span class="n">x</span><span class="p">,</span> <span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">partitioner</span><span class="p">(</span><span class="n">key</span><span class="p">)])</span>

        <span class="k">return</span> <span class="n">values</span><span class="o">.</span><span class="n">collect</span><span class="p">()</span>
</div>
    <span class="k">def</span> <span class="nf">_to_java_object_rdd</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot; Return an JavaRDD of Object by unpickling</span>

<span class="sd">        It will convert each Python object into Java object by Pyrolite, whenever the</span>
<span class="sd">        RDD is serialized in batch or not.</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="n">rdd</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_pickled</span><span class="p">()</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="o">.</span><span class="n">_jvm</span><span class="o">.</span><span class="n">SerDeUtil</span><span class="o">.</span><span class="n">pythonToJava</span><span class="p">(</span><span class="n">rdd</span><span class="o">.</span><span class="n">_jrdd</span><span class="p">,</span> <span class="bp">True</span><span class="p">)</span>

<div class="viewcode-block" id="RDD.countApprox"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.countApprox">[docs]</a>    <span class="k">def</span> <span class="nf">countApprox</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">timeout</span><span class="p">,</span> <span class="n">confidence</span><span class="o">=</span><span class="mf">0.95</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        .. note:: Experimental</span>

<span class="sd">        Approximate version of count() that returns a potentially incomplete</span>
<span class="sd">        result within a timeout, even if not all tasks have finished.</span>

<span class="sd">        &gt;&gt;&gt; rdd = sc.parallelize(range(1000), 10)</span>
<span class="sd">        &gt;&gt;&gt; rdd.countApprox(1000, 1.0)</span>
<span class="sd">        1000</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="n">drdd</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">mapPartitions</span><span class="p">(</span><span class="k">lambda</span> <span class="n">it</span><span class="p">:</span> <span class="p">[</span><span class="nb">float</span><span class="p">(</span><span class="nb">sum</span><span class="p">(</span><span class="mi">1</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="n">it</span><span class="p">))])</span>
        <span class="k">return</span> <span class="nb">int</span><span class="p">(</span><span class="n">drdd</span><span class="o">.</span><span class="n">sumApprox</span><span class="p">(</span><span class="n">timeout</span><span class="p">,</span> <span class="n">confidence</span><span class="p">))</span>
</div>
<div class="viewcode-block" id="RDD.sumApprox"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.sumApprox">[docs]</a>    <span class="k">def</span> <span class="nf">sumApprox</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">timeout</span><span class="p">,</span> <span class="n">confidence</span><span class="o">=</span><span class="mf">0.95</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        .. note:: Experimental</span>

<span class="sd">        Approximate operation to return the sum within a timeout</span>
<span class="sd">        or meet the confidence.</span>

<span class="sd">        &gt;&gt;&gt; rdd = sc.parallelize(range(1000), 10)</span>
<span class="sd">        &gt;&gt;&gt; r = sum(range(1000))</span>
<span class="sd">        &gt;&gt;&gt; abs(rdd.sumApprox(1000) - r) / r &lt; 0.05</span>
<span class="sd">        True</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="n">jrdd</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">mapPartitions</span><span class="p">(</span><span class="k">lambda</span> <span class="n">it</span><span class="p">:</span> <span class="p">[</span><span class="nb">float</span><span class="p">(</span><span class="nb">sum</span><span class="p">(</span><span class="n">it</span><span class="p">))])</span><span class="o">.</span><span class="n">_to_java_object_rdd</span><span class="p">()</span>
        <span class="n">jdrdd</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="o">.</span><span class="n">_jvm</span><span class="o">.</span><span class="n">JavaDoubleRDD</span><span class="o">.</span><span class="n">fromRDD</span><span class="p">(</span><span class="n">jrdd</span><span class="o">.</span><span class="n">rdd</span><span class="p">())</span>
        <span class="n">r</span> <span class="o">=</span> <span class="n">jdrdd</span><span class="o">.</span><span class="n">sumApprox</span><span class="p">(</span><span class="n">timeout</span><span class="p">,</span> <span class="n">confidence</span><span class="p">)</span><span class="o">.</span><span class="n">getFinalValue</span><span class="p">()</span>
        <span class="k">return</span> <span class="n">BoundedFloat</span><span class="p">(</span><span class="n">r</span><span class="o">.</span><span class="n">mean</span><span class="p">(),</span> <span class="n">r</span><span class="o">.</span><span class="n">confidence</span><span class="p">(),</span> <span class="n">r</span><span class="o">.</span><span class="n">low</span><span class="p">(),</span> <span class="n">r</span><span class="o">.</span><span class="n">high</span><span class="p">())</span>
</div>
<div class="viewcode-block" id="RDD.meanApprox"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.meanApprox">[docs]</a>    <span class="k">def</span> <span class="nf">meanApprox</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">timeout</span><span class="p">,</span> <span class="n">confidence</span><span class="o">=</span><span class="mf">0.95</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        .. note:: Experimental</span>

<span class="sd">        Approximate operation to return the mean within a timeout</span>
<span class="sd">        or meet the confidence.</span>

<span class="sd">        &gt;&gt;&gt; rdd = sc.parallelize(range(1000), 10)</span>
<span class="sd">        &gt;&gt;&gt; r = sum(range(1000)) / 1000.0</span>
<span class="sd">        &gt;&gt;&gt; abs(rdd.meanApprox(1000) - r) / r &lt; 0.05</span>
<span class="sd">        True</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="n">jrdd</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">map</span><span class="p">(</span><span class="nb">float</span><span class="p">)</span><span class="o">.</span><span class="n">_to_java_object_rdd</span><span class="p">()</span>
        <span class="n">jdrdd</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="o">.</span><span class="n">_jvm</span><span class="o">.</span><span class="n">JavaDoubleRDD</span><span class="o">.</span><span class="n">fromRDD</span><span class="p">(</span><span class="n">jrdd</span><span class="o">.</span><span class="n">rdd</span><span class="p">())</span>
        <span class="n">r</span> <span class="o">=</span> <span class="n">jdrdd</span><span class="o">.</span><span class="n">meanApprox</span><span class="p">(</span><span class="n">timeout</span><span class="p">,</span> <span class="n">confidence</span><span class="p">)</span><span class="o">.</span><span class="n">getFinalValue</span><span class="p">()</span>
        <span class="k">return</span> <span class="n">BoundedFloat</span><span class="p">(</span><span class="n">r</span><span class="o">.</span><span class="n">mean</span><span class="p">(),</span> <span class="n">r</span><span class="o">.</span><span class="n">confidence</span><span class="p">(),</span> <span class="n">r</span><span class="o">.</span><span class="n">low</span><span class="p">(),</span> <span class="n">r</span><span class="o">.</span><span class="n">high</span><span class="p">())</span>
</div>
<div class="viewcode-block" id="RDD.countApproxDistinct"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.countApproxDistinct">[docs]</a>    <span class="k">def</span> <span class="nf">countApproxDistinct</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">relativeSD</span><span class="o">=</span><span class="mf">0.05</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        .. note:: Experimental</span>

<span class="sd">        Return approximate number of distinct elements in the RDD.</span>

<span class="sd">        The algorithm used is based on streamlib&#39;s implementation of</span>
<span class="sd">        &quot;HyperLogLog in Practice: Algorithmic Engineering of a State</span>
<span class="sd">        of The Art Cardinality Estimation Algorithm&quot;, available</span>
<span class="sd">        &lt;a href=&quot;http://dx.doi.org/10.1145/2452376.2452456&quot;&gt;here&lt;/a&gt;.</span>

<span class="sd">        :param relativeSD: Relative accuracy. Smaller values create</span>
<span class="sd">                           counters that require more space.</span>
<span class="sd">                           It must be greater than 0.000017.</span>

<span class="sd">        &gt;&gt;&gt; n = sc.parallelize(range(1000)).map(str).countApproxDistinct()</span>
<span class="sd">        &gt;&gt;&gt; 900 &lt; n &lt; 1100</span>
<span class="sd">        True</span>
<span class="sd">        &gt;&gt;&gt; n = sc.parallelize([i % 20 for i in range(1000)]).countApproxDistinct()</span>
<span class="sd">        &gt;&gt;&gt; 16 &lt; n &lt; 24</span>
<span class="sd">        True</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">if</span> <span class="n">relativeSD</span> <span class="o">&lt;</span> <span class="mf">0.000017</span><span class="p">:</span>
            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s">&quot;relativeSD should be greater than 0.000017&quot;</span><span class="p">)</span>
        <span class="c"># the hash space in Java is 2^32</span>
        <span class="n">hashRDD</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">map</span><span class="p">(</span><span class="k">lambda</span> <span class="n">x</span><span class="p">:</span> <span class="n">portable_hash</span><span class="p">(</span><span class="n">x</span><span class="p">)</span> <span class="o">&amp;</span> <span class="mh">0xFFFFFFFF</span><span class="p">)</span>
        <span class="k">return</span> <span class="n">hashRDD</span><span class="o">.</span><span class="n">_to_java_object_rdd</span><span class="p">()</span><span class="o">.</span><span class="n">countApproxDistinct</span><span class="p">(</span><span class="n">relativeSD</span><span class="p">)</span>
</div>
<div class="viewcode-block" id="RDD.toLocalIterator"><a class="viewcode-back" href="../../pyspark.html#pyspark.RDD.toLocalIterator">[docs]</a>    <span class="k">def</span> <span class="nf">toLocalIterator</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        Return an iterator that contains all of the elements in this RDD.</span>
<span class="sd">        The iterator will consume as much memory as the largest partition in this RDD.</span>
<span class="sd">        &gt;&gt;&gt; rdd = sc.parallelize(range(10))</span>
<span class="sd">        &gt;&gt;&gt; [x for x in rdd.toLocalIterator()]</span>
<span class="sd">        [0, 1, 2, 3, 4, 5, 6, 7, 8, 9]</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="k">for</span> <span class="n">partition</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">getNumPartitions</span><span class="p">()):</span>
            <span class="n">rows</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">context</span><span class="o">.</span><span class="n">runJob</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="k">lambda</span> <span class="n">x</span><span class="p">:</span> <span class="n">x</span><span class="p">,</span> <span class="p">[</span><span class="n">partition</span><span class="p">])</span>
            <span class="k">for</span> <span class="n">row</span> <span class="ow">in</span> <span class="n">rows</span><span class="p">:</span>
                <span class="k">yield</span> <span class="n">row</span>

</div></div>
<span class="k">def</span> <span class="nf">_prepare_for_python_RDD</span><span class="p">(</span><span class="n">sc</span><span class="p">,</span> <span class="n">command</span><span class="p">,</span> <span class="n">obj</span><span class="o">=</span><span class="bp">None</span><span class="p">):</span>
    <span class="c"># the serialized command will be compressed by broadcast</span>
    <span class="n">ser</span> <span class="o">=</span> <span class="n">CloudPickleSerializer</span><span class="p">()</span>
    <span class="n">pickled_command</span> <span class="o">=</span> <span class="n">ser</span><span class="o">.</span><span class="n">dumps</span><span class="p">(</span><span class="n">command</span><span class="p">)</span>
    <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">pickled_command</span><span class="p">)</span> <span class="o">&gt;</span> <span class="p">(</span><span class="mi">1</span> <span class="o">&lt;&lt;</span> <span class="mi">20</span><span class="p">):</span>  <span class="c"># 1M</span>
        <span class="c"># The broadcast will have same life cycle as created PythonRDD</span>
        <span class="n">broadcast</span> <span class="o">=</span> <span class="n">sc</span><span class="o">.</span><span class="n">broadcast</span><span class="p">(</span><span class="n">pickled_command</span><span class="p">)</span>
        <span class="n">pickled_command</span> <span class="o">=</span> <span class="n">ser</span><span class="o">.</span><span class="n">dumps</span><span class="p">(</span><span class="n">broadcast</span><span class="p">)</span>
    <span class="c"># There is a bug in py4j.java_gateway.JavaClass with auto_convert</span>
    <span class="c"># https://github.com/bartdag/py4j/issues/161</span>
    <span class="c"># TODO: use auto_convert once py4j fix the bug</span>
    <span class="n">broadcast_vars</span> <span class="o">=</span> <span class="n">ListConverter</span><span class="p">()</span><span class="o">.</span><span class="n">convert</span><span class="p">(</span>
        <span class="p">[</span><span class="n">x</span><span class="o">.</span><span class="n">_jbroadcast</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">sc</span><span class="o">.</span><span class="n">_pickled_broadcast_vars</span><span class="p">],</span>
        <span class="n">sc</span><span class="o">.</span><span class="n">_gateway</span><span class="o">.</span><span class="n">_gateway_client</span><span class="p">)</span>
    <span class="n">sc</span><span class="o">.</span><span class="n">_pickled_broadcast_vars</span><span class="o">.</span><span class="n">clear</span><span class="p">()</span>
    <span class="n">env</span> <span class="o">=</span> <span class="n">MapConverter</span><span class="p">()</span><span class="o">.</span><span class="n">convert</span><span class="p">(</span><span class="n">sc</span><span class="o">.</span><span class="n">environment</span><span class="p">,</span> <span class="n">sc</span><span class="o">.</span><span class="n">_gateway</span><span class="o">.</span><span class="n">_gateway_client</span><span class="p">)</span>
    <span class="n">includes</span> <span class="o">=</span> <span class="n">ListConverter</span><span class="p">()</span><span class="o">.</span><span class="n">convert</span><span class="p">(</span><span class="n">sc</span><span class="o">.</span><span class="n">_python_includes</span><span class="p">,</span> <span class="n">sc</span><span class="o">.</span><span class="n">_gateway</span><span class="o">.</span><span class="n">_gateway_client</span><span class="p">)</span>
    <span class="k">return</span> <span class="n">pickled_command</span><span class="p">,</span> <span class="n">broadcast_vars</span><span class="p">,</span> <span class="n">env</span><span class="p">,</span> <span class="n">includes</span>


<span class="k">class</span> <span class="nc">PipelinedRDD</span><span class="p">(</span><span class="n">RDD</span><span class="p">):</span>

    <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">    Pipelined maps:</span>

<span class="sd">    &gt;&gt;&gt; rdd = sc.parallelize([1, 2, 3, 4])</span>
<span class="sd">    &gt;&gt;&gt; rdd.map(lambda x: 2 * x).cache().map(lambda x: 2 * x).collect()</span>
<span class="sd">    [4, 8, 12, 16]</span>
<span class="sd">    &gt;&gt;&gt; rdd.map(lambda x: 2 * x).map(lambda x: 2 * x).collect()</span>
<span class="sd">    [4, 8, 12, 16]</span>

<span class="sd">    Pipelined reduces:</span>
<span class="sd">    &gt;&gt;&gt; from operator import add</span>
<span class="sd">    &gt;&gt;&gt; rdd.map(lambda x: 2 * x).reduce(add)</span>
<span class="sd">    20</span>
<span class="sd">    &gt;&gt;&gt; rdd.flatMap(lambda x: [x, x]).reduce(add)</span>
<span class="sd">    20</span>
<span class="sd">    &quot;&quot;&quot;</span>

    <span class="k">def</span> <span class="nf">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">prev</span><span class="p">,</span> <span class="n">func</span><span class="p">,</span> <span class="n">preservesPartitioning</span><span class="o">=</span><span class="bp">False</span><span class="p">):</span>
        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">prev</span><span class="p">,</span> <span class="n">PipelinedRDD</span><span class="p">)</span> <span class="ow">or</span> <span class="ow">not</span> <span class="n">prev</span><span class="o">.</span><span class="n">_is_pipelinable</span><span class="p">():</span>
            <span class="c"># This transformation is the first in its stage:</span>
            <span class="bp">self</span><span class="o">.</span><span class="n">func</span> <span class="o">=</span> <span class="n">func</span>
            <span class="bp">self</span><span class="o">.</span><span class="n">preservesPartitioning</span> <span class="o">=</span> <span class="n">preservesPartitioning</span>
            <span class="bp">self</span><span class="o">.</span><span class="n">_prev_jrdd</span> <span class="o">=</span> <span class="n">prev</span><span class="o">.</span><span class="n">_jrdd</span>
            <span class="bp">self</span><span class="o">.</span><span class="n">_prev_jrdd_deserializer</span> <span class="o">=</span> <span class="n">prev</span><span class="o">.</span><span class="n">_jrdd_deserializer</span>
        <span class="k">else</span><span class="p">:</span>
            <span class="n">prev_func</span> <span class="o">=</span> <span class="n">prev</span><span class="o">.</span><span class="n">func</span>

            <span class="k">def</span> <span class="nf">pipeline_func</span><span class="p">(</span><span class="n">split</span><span class="p">,</span> <span class="n">iterator</span><span class="p">):</span>
                <span class="k">return</span> <span class="n">func</span><span class="p">(</span><span class="n">split</span><span class="p">,</span> <span class="n">prev_func</span><span class="p">(</span><span class="n">split</span><span class="p">,</span> <span class="n">iterator</span><span class="p">))</span>
            <span class="bp">self</span><span class="o">.</span><span class="n">func</span> <span class="o">=</span> <span class="n">pipeline_func</span>
            <span class="bp">self</span><span class="o">.</span><span class="n">preservesPartitioning</span> <span class="o">=</span> \
                <span class="n">prev</span><span class="o">.</span><span class="n">preservesPartitioning</span> <span class="ow">and</span> <span class="n">preservesPartitioning</span>
            <span class="bp">self</span><span class="o">.</span><span class="n">_prev_jrdd</span> <span class="o">=</span> <span class="n">prev</span><span class="o">.</span><span class="n">_prev_jrdd</span>  <span class="c"># maintain the pipeline</span>
            <span class="bp">self</span><span class="o">.</span><span class="n">_prev_jrdd_deserializer</span> <span class="o">=</span> <span class="n">prev</span><span class="o">.</span><span class="n">_prev_jrdd_deserializer</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">is_cached</span> <span class="o">=</span> <span class="bp">False</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">is_checkpointed</span> <span class="o">=</span> <span class="bp">False</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span> <span class="o">=</span> <span class="n">prev</span><span class="o">.</span><span class="n">ctx</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">prev</span> <span class="o">=</span> <span class="n">prev</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">_jrdd_val</span> <span class="o">=</span> <span class="bp">None</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">_id</span> <span class="o">=</span> <span class="bp">None</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">_jrdd_deserializer</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="o">.</span><span class="n">serializer</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">_bypass_serializer</span> <span class="o">=</span> <span class="bp">False</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">partitioner</span> <span class="o">=</span> <span class="n">prev</span><span class="o">.</span><span class="n">partitioner</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">preservesPartitioning</span> <span class="k">else</span> <span class="bp">None</span>

    <span class="k">def</span> <span class="nf">getNumPartitions</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_prev_jrdd</span><span class="o">.</span><span class="n">partitions</span><span class="p">()</span><span class="o">.</span><span class="n">size</span><span class="p">()</span>

    <span class="nd">@property</span>
    <span class="k">def</span> <span class="nf">_jrdd</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_jrdd_val</span><span class="p">:</span>
            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_jrdd_val</span>
        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_bypass_serializer</span><span class="p">:</span>
            <span class="bp">self</span><span class="o">.</span><span class="n">_jrdd_deserializer</span> <span class="o">=</span> <span class="n">NoOpSerializer</span><span class="p">()</span>

        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="o">.</span><span class="n">profiler_collector</span><span class="p">:</span>
            <span class="n">profiler</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="o">.</span><span class="n">profiler_collector</span><span class="o">.</span><span class="n">new_profiler</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="p">)</span>
        <span class="k">else</span><span class="p">:</span>
            <span class="n">profiler</span> <span class="o">=</span> <span class="bp">None</span>

        <span class="n">command</span> <span class="o">=</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">func</span><span class="p">,</span> <span class="n">profiler</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_prev_jrdd_deserializer</span><span class="p">,</span>
                   <span class="bp">self</span><span class="o">.</span><span class="n">_jrdd_deserializer</span><span class="p">)</span>
        <span class="n">pickled_cmd</span><span class="p">,</span> <span class="n">bvars</span><span class="p">,</span> <span class="n">env</span><span class="p">,</span> <span class="n">includes</span> <span class="o">=</span> <span class="n">_prepare_for_python_RDD</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="p">,</span> <span class="n">command</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span>
        <span class="n">python_rdd</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="o">.</span><span class="n">_jvm</span><span class="o">.</span><span class="n">PythonRDD</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_prev_jrdd</span><span class="o">.</span><span class="n">rdd</span><span class="p">(),</span>
                                             <span class="nb">bytearray</span><span class="p">(</span><span class="n">pickled_cmd</span><span class="p">),</span>
                                             <span class="n">env</span><span class="p">,</span> <span class="n">includes</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">preservesPartitioning</span><span class="p">,</span>
                                             <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="o">.</span><span class="n">pythonExec</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="o">.</span><span class="n">pythonVer</span><span class="p">,</span>
                                             <span class="n">bvars</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="o">.</span><span class="n">_javaAccumulator</span><span class="p">)</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">_jrdd_val</span> <span class="o">=</span> <span class="n">python_rdd</span><span class="o">.</span><span class="n">asJavaRDD</span><span class="p">()</span>

        <span class="k">if</span> <span class="n">profiler</span><span class="p">:</span>
            <span class="bp">self</span><span class="o">.</span><span class="n">_id</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_jrdd_val</span><span class="o">.</span><span class="n">id</span><span class="p">()</span>
            <span class="bp">self</span><span class="o">.</span><span class="n">ctx</span><span class="o">.</span><span class="n">profiler_collector</span><span class="o">.</span><span class="n">add_profiler</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_id</span><span class="p">,</span> <span class="n">profiler</span><span class="p">)</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_jrdd_val</span>

    <span class="k">def</span> <span class="nf">id</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_id</span> <span class="ow">is</span> <span class="bp">None</span><span class="p">:</span>
            <span class="bp">self</span><span class="o">.</span><span class="n">_id</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_jrdd</span><span class="o">.</span><span class="n">id</span><span class="p">()</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_id</span>

    <span class="k">def</span> <span class="nf">_is_pipelinable</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="k">return</span> <span class="ow">not</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">is_cached</span> <span class="ow">or</span> <span class="bp">self</span><span class="o">.</span><span class="n">is_checkpointed</span><span class="p">)</span>


<span class="k">def</span> <span class="nf">_test</span><span class="p">():</span>
    <span class="kn">import</span> <span class="nn">doctest</span>
    <span class="kn">from</span> <span class="nn">pyspark.context</span> <span class="kn">import</span> <span class="n">SparkContext</span>
    <span class="n">globs</span> <span class="o">=</span> <span class="nb">globals</span><span class="p">()</span><span class="o">.</span><span class="n">copy</span><span class="p">()</span>
    <span class="c"># The small batch size here ensures that we see multiple batches,</span>
    <span class="c"># even in these small test examples:</span>
    <span class="n">globs</span><span class="p">[</span><span class="s">&#39;sc&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">SparkContext</span><span class="p">(</span><span class="s">&#39;local[4]&#39;</span><span class="p">,</span> <span class="s">&#39;PythonTest&#39;</span><span class="p">)</span>
    <span class="p">(</span><span class="n">failure_count</span><span class="p">,</span> <span class="n">test_count</span><span class="p">)</span> <span class="o">=</span> <span class="n">doctest</span><span class="o">.</span><span class="n">testmod</span><span class="p">(</span>
        <span class="n">globs</span><span class="o">=</span><span class="n">globs</span><span class="p">,</span> <span class="n">optionflags</span><span class="o">=</span><span class="n">doctest</span><span class="o">.</span><span class="n">ELLIPSIS</span><span class="p">)</span>
    <span class="n">globs</span><span class="p">[</span><span class="s">&#39;sc&#39;</span><span class="p">]</span><span class="o">.</span><span class="n">stop</span><span class="p">()</span>
    <span class="k">if</span> <span class="n">failure_count</span><span class="p">:</span>
        <span class="nb">exit</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">)</span>


<span class="k">if</span> <span class="n">__name__</span> <span class="o">==</span> <span class="s">&quot;__main__&quot;</span><span class="p">:</span>
    <span class="n">_test</span><span class="p">()</span>
</pre></div>

          </div>
        </div>
      </div>
      <div class="sphinxsidebar" role="navigation" aria-label="main navigation">
        <div class="sphinxsidebarwrapper">
            <p class="logo"><a href="../../index.html">
              <img class="logo" src="../../_static/spark-logo-hd.png" alt="Logo"/>
            </a></p>
<div id="searchbox" style="display: none" role="search">
  <h3>Quick search</h3>
    <form class="search" action="../../search.html" method="get">
      <input type="text" name="q" />
      <input type="submit" value="Go" />
      <input type="hidden" name="check_keywords" value="yes" />
      <input type="hidden" name="area" value="default" />
    </form>
    <p class="searchtip" style="font-size: 90%">
    Enter search terms or a module, class or function name.
    </p>
</div>
<script type="text/javascript">$('#searchbox').show(0);</script>
        </div>
      </div>
      <div class="clearer"></div>
    </div>
    <div class="related" role="navigation" aria-label="related navigation">
      <h3>Navigation</h3>
      <ul>
        <li class="nav-item nav-item-0"><a href="../../index.html">PySpark master documentation</a> &raquo;</li>
          <li class="nav-item nav-item-1"><a href="../index.html" >Module code</a> &raquo;</li> 
      </ul>
    </div>
    <div class="footer" role="contentinfo">
        &copy; Copyright .
      Created using <a href="http://sphinx-doc.org/">Sphinx</a> 1.3.1.
    </div>
  </body>
</html>
generated by cgit v1.2.3 (git 2.39.1) at 2024-08-15 20:55:36 +0000