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
|
#
# Licensed to the Apache Software Foundation (ASF) under one or more
# contributor license agreements. See the NOTICE file distributed with
# this work for additional information regarding copyright ownership.
# The ASF licenses this file to You under the Apache License, Version 2.0
# (the "License"); you may not use this file except in compliance with
# the License. You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
#
from base64 import standard_b64encode as b64enc
import copy
from collections import defaultdict
from collections import namedtuple
from itertools import chain, ifilter, imap
import operator
import os
import sys
import shlex
import traceback
from subprocess import Popen, PIPE
from tempfile import NamedTemporaryFile
from threading import Thread
import warnings
import heapq
import bisect
from random import Random
from math import sqrt, log, isinf, isnan
from pyspark.serializers import NoOpSerializer, CartesianDeserializer, \
BatchedSerializer, CloudPickleSerializer, PairDeserializer, \
PickleSerializer, pack_long, CompressedSerializer
from pyspark.join import python_join, python_left_outer_join, \
python_right_outer_join, python_cogroup
from pyspark.statcounter import StatCounter
from pyspark.rddsampler import RDDSampler, RDDStratifiedSampler
from pyspark.storagelevel import StorageLevel
from pyspark.resultiterable import ResultIterable
from pyspark.shuffle import Aggregator, InMemoryMerger, ExternalMerger, \
get_used_memory
from py4j.java_collections import ListConverter, MapConverter
__all__ = ["RDD"]
# TODO: for Python 3.3+, PYTHONHASHSEED should be reset to disable randomized
# hash for string
def portable_hash(x):
"""
This function returns consistant hash code for builtin types, especially
for None and tuple with None.
The algrithm is similar to that one used by CPython 2.7
>>> portable_hash(None)
0
>>> portable_hash((None, 1))
219750521
"""
if x is None:
return 0
if isinstance(x, tuple):
h = 0x345678
for i in x:
h ^= portable_hash(i)
h *= 1000003
h &= 0xffffffff
h ^= len(x)
if h == -1:
h = -2
return h
return hash(x)
def _extract_concise_traceback():
"""
This function returns the traceback info for a callsite, returns a dict
with function name, file name and line number
"""
tb = traceback.extract_stack()
callsite = namedtuple("Callsite", "function file linenum")
if len(tb) == 0:
return None
file, line, module, what = tb[len(tb) - 1]
sparkpath = os.path.dirname(file)
first_spark_frame = len(tb) - 1
for i in range(0, len(tb)):
file, line, fun, what = tb[i]
if file.startswith(sparkpath):
first_spark_frame = i
break
if first_spark_frame == 0:
file, line, fun, what = tb[0]
return callsite(function=fun, file=file, linenum=line)
sfile, sline, sfun, swhat = tb[first_spark_frame]
ufile, uline, ufun, uwhat = tb[first_spark_frame - 1]
return callsite(function=sfun, file=ufile, linenum=uline)
_spark_stack_depth = 0
class _JavaStackTrace(object):
def __init__(self, sc):
tb = _extract_concise_traceback()
if tb is not None:
self._traceback = "%s at %s:%s" % (
tb.function, tb.file, tb.linenum)
else:
self._traceback = "Error! Could not extract traceback info"
self._context = sc
def __enter__(self):
global _spark_stack_depth
if _spark_stack_depth == 0:
self._context._jsc.setCallSite(self._traceback)
_spark_stack_depth += 1
def __exit__(self, type, value, tb):
global _spark_stack_depth
_spark_stack_depth -= 1
if _spark_stack_depth == 0:
self._context._jsc.setCallSite(None)
class MaxHeapQ(object):
"""
An implementation of MaxHeap.
>>> import pyspark.rdd
>>> heap = pyspark.rdd.MaxHeapQ(5)
>>> [heap.insert(i) for i in range(10)]
[None, None, None, None, None, None, None, None, None, None]
>>> sorted(heap.getElements())
[0, 1, 2, 3, 4]
>>> heap = pyspark.rdd.MaxHeapQ(5)
>>> [heap.insert(i) for i in range(9, -1, -1)]
[None, None, None, None, None, None, None, None, None, None]
>>> sorted(heap.getElements())
[0, 1, 2, 3, 4]
>>> heap = pyspark.rdd.MaxHeapQ(1)
>>> [heap.insert(i) for i in range(9, -1, -1)]
[None, None, None, None, None, None, None, None, None, None]
>>> heap.getElements()
[0]
"""
def __init__(self, maxsize):
# We start from q[1], so its children are always 2 * k
self.q = [0]
self.maxsize = maxsize
def _swim(self, k):
while (k > 1) and (self.q[k / 2] < self.q[k]):
self._swap(k, k / 2)
k = k / 2
def _swap(self, i, j):
t = self.q[i]
self.q[i] = self.q[j]
self.q[j] = t
def _sink(self, k):
N = self.size()
while 2 * k <= N:
j = 2 * k
# Here we test if both children are greater than parent
# if not swap with larger one.
if j < N and self.q[j] < self.q[j + 1]:
j = j + 1
if(self.q[k] > self.q[j]):
break
self._swap(k, j)
k = j
def size(self):
return len(self.q) - 1
def insert(self, value):
if (self.size()) < self.maxsize:
self.q.append(value)
self._swim(self.size())
else:
self._replaceRoot(value)
def getElements(self):
return self.q[1:]
def _replaceRoot(self, value):
if(self.q[1] > value):
self.q[1] = value
self._sink(1)
def _parse_memory(s):
"""
Parse a memory string in the format supported by Java (e.g. 1g, 200m) and
return the value in MB
>>> _parse_memory("256m")
256
>>> _parse_memory("2g")
2048
"""
units = {'g': 1024, 'm': 1, 't': 1 << 20, 'k': 1.0 / 1024}
if s[-1] not in units:
raise ValueError("invalid format: " + s)
return int(float(s[:-1]) * units[s[-1].lower()])
class RDD(object):
"""
A Resilient Distributed Dataset (RDD), the basic abstraction in Spark.
Represents an immutable, partitioned collection of elements that can be
operated on in parallel.
"""
def __init__(self, jrdd, ctx, jrdd_deserializer):
self._jrdd = jrdd
self.is_cached = False
self.is_checkpointed = False
self.ctx = ctx
self._jrdd_deserializer = jrdd_deserializer
self._id = jrdd.id()
def _toPickleSerialization(self):
if (self._jrdd_deserializer == PickleSerializer() or
self._jrdd_deserializer == BatchedSerializer(PickleSerializer())):
return self
else:
return self._reserialize(BatchedSerializer(PickleSerializer(), 10))
def id(self):
"""
A unique ID for this RDD (within its SparkContext).
"""
return self._id
def __repr__(self):
return self._jrdd.toString()
@property
def context(self):
"""
The L{SparkContext} that this RDD was created on.
"""
return self.ctx
def cache(self):
"""
Persist this RDD with the default storage level (C{MEMORY_ONLY_SER}).
"""
self.is_cached = True
self.persist(StorageLevel.MEMORY_ONLY_SER)
return self
def persist(self, storageLevel):
"""
Set this RDD's storage level to persist its values across operations
after the first time it is computed. This can only be used to assign
a new storage level if the RDD does not have a storage level set yet.
"""
self.is_cached = True
javaStorageLevel = self.ctx._getJavaStorageLevel(storageLevel)
self._jrdd.persist(javaStorageLevel)
return self
def unpersist(self):
"""
Mark the RDD as non-persistent, and remove all blocks for it from
memory and disk.
"""
self.is_cached = False
self._jrdd.unpersist()
return self
def checkpoint(self):
"""
Mark this RDD for checkpointing. It will be saved to a file inside the
checkpoint directory set with L{SparkContext.setCheckpointDir()} and
all references to its parent RDDs will be removed. This function must
be called before any job has been executed on this RDD. It is strongly
recommended that this RDD is persisted in memory, otherwise saving it
on a file will require recomputation.
"""
self.is_checkpointed = True
self._jrdd.rdd().checkpoint()
def isCheckpointed(self):
"""
Return whether this RDD has been checkpointed or not
"""
return self._jrdd.rdd().isCheckpointed()
def getCheckpointFile(self):
"""
Gets the name of the file to which this RDD was checkpointed
"""
checkpointFile = self._jrdd.rdd().getCheckpointFile()
if checkpointFile.isDefined():
return checkpointFile.get()
else:
return None
def map(self, f, preservesPartitioning=False):
"""
Return a new RDD by applying a function to each element of this RDD.
>>> rdd = sc.parallelize(["b", "a", "c"])
>>> sorted(rdd.map(lambda x: (x, 1)).collect())
[('a', 1), ('b', 1), ('c', 1)]
"""
def func(_, iterator):
return imap(f, iterator)
return self.mapPartitionsWithIndex(func, preservesPartitioning)
def flatMap(self, f, preservesPartitioning=False):
"""
Return a new RDD by first applying a function to all elements of this
RDD, and then flattening the results.
>>> rdd = sc.parallelize([2, 3, 4])
>>> sorted(rdd.flatMap(lambda x: range(1, x)).collect())
[1, 1, 1, 2, 2, 3]
>>> sorted(rdd.flatMap(lambda x: [(x, x), (x, x)]).collect())
[(2, 2), (2, 2), (3, 3), (3, 3), (4, 4), (4, 4)]
"""
def func(s, iterator):
return chain.from_iterable(imap(f, iterator))
return self.mapPartitionsWithIndex(func, preservesPartitioning)
def mapPartitions(self, f, preservesPartitioning=False):
"""
Return a new RDD by applying a function to each partition of this RDD.
>>> rdd = sc.parallelize([1, 2, 3, 4], 2)
>>> def f(iterator): yield sum(iterator)
>>> rdd.mapPartitions(f).collect()
[3, 7]
"""
def func(s, iterator):
return f(iterator)
return self.mapPartitionsWithIndex(func)
def mapPartitionsWithIndex(self, f, preservesPartitioning=False):
"""
Return a new RDD by applying a function to each partition of this RDD,
while tracking the index of the original partition.
>>> rdd = sc.parallelize([1, 2, 3, 4], 4)
>>> def f(splitIndex, iterator): yield splitIndex
>>> rdd.mapPartitionsWithIndex(f).sum()
6
"""
return PipelinedRDD(self, f, preservesPartitioning)
def mapPartitionsWithSplit(self, f, preservesPartitioning=False):
"""
Deprecated: use mapPartitionsWithIndex instead.
Return a new RDD by applying a function to each partition of this RDD,
while tracking the index of the original partition.
>>> rdd = sc.parallelize([1, 2, 3, 4], 4)
>>> def f(splitIndex, iterator): yield splitIndex
>>> rdd.mapPartitionsWithSplit(f).sum()
6
"""
warnings.warn("mapPartitionsWithSplit is deprecated; "
"use mapPartitionsWithIndex instead", DeprecationWarning, stacklevel=2)
return self.mapPartitionsWithIndex(f, preservesPartitioning)
def getNumPartitions(self):
"""
Returns the number of partitions in RDD
>>> rdd = sc.parallelize([1, 2, 3, 4], 2)
>>> rdd.getNumPartitions()
2
"""
return self._jrdd.partitions().size()
def filter(self, f):
"""
Return a new RDD containing only the elements that satisfy a predicate.
>>> rdd = sc.parallelize([1, 2, 3, 4, 5])
>>> rdd.filter(lambda x: x % 2 == 0).collect()
[2, 4]
"""
def func(iterator):
return ifilter(f, iterator)
return self.mapPartitions(func)
def distinct(self):
"""
Return a new RDD containing the distinct elements in this RDD.
>>> sorted(sc.parallelize([1, 1, 2, 3]).distinct().collect())
[1, 2, 3]
"""
return self.map(lambda x: (x, None)) \
.reduceByKey(lambda x, _: x) \
.map(lambda (x, _): x)
def sample(self, withReplacement, fraction, seed=None):
"""
Return a sampled subset of this RDD (relies on numpy and falls back
on default random generator if numpy is unavailable).
>>> sc.parallelize(range(0, 100)).sample(False, 0.1, 2).collect() #doctest: +SKIP
[2, 3, 20, 21, 24, 41, 42, 66, 67, 89, 90, 98]
"""
assert fraction >= 0.0, "Negative fraction value: %s" % fraction
return self.mapPartitionsWithIndex(RDDSampler(withReplacement, fraction, seed).func, True)
# this is ported from scala/spark/RDD.scala
def takeSample(self, withReplacement, num, seed=None):
"""
Return a fixed-size sampled subset of this RDD (currently requires
numpy).
>>> rdd = sc.parallelize(range(0, 10))
>>> len(rdd.takeSample(True, 20, 1))
20
>>> len(rdd.takeSample(False, 5, 2))
5
>>> len(rdd.takeSample(False, 15, 3))
10
"""
numStDev = 10.0
if num < 0:
raise ValueError("Sample size cannot be negative.")
elif num == 0:
return []
initialCount = self.count()
if initialCount == 0:
return []
rand = Random(seed)
if (not withReplacement) and num >= initialCount:
# shuffle current RDD and return
samples = self.collect()
rand.shuffle(samples)
return samples
maxSampleSize = sys.maxint - int(numStDev * sqrt(sys.maxint))
if num > maxSampleSize:
raise ValueError(
"Sample size cannot be greater than %d." % maxSampleSize)
fraction = RDD._computeFractionForSampleSize(
num, initialCount, withReplacement)
samples = self.sample(withReplacement, fraction, seed).collect()
# If the first sample didn't turn out large enough, keep trying to take samples;
# this shouldn't happen often because we use a big multiplier for their initial size.
# See: scala/spark/RDD.scala
while len(samples) < num:
# TODO: add log warning for when more than one iteration was run
seed = rand.randint(0, sys.maxint)
samples = self.sample(withReplacement, fraction, seed).collect()
rand.shuffle(samples)
return samples[0:num]
@staticmethod
def _computeFractionForSampleSize(sampleSizeLowerBound, total, withReplacement):
"""
Returns a sampling rate that guarantees a sample of
size >= sampleSizeLowerBound 99.99% of the time.
How the sampling rate is determined:
Let p = num / total, where num is the sample size and total is the
total number of data points in the RDD. We're trying to compute
q > p such that
- when sampling with replacement, we're drawing each data point
with prob_i ~ Pois(q), where we want to guarantee
Pr[s < num] < 0.0001 for s = sum(prob_i for i from 0 to
total), i.e. the failure rate of not having a sufficiently large
sample < 0.0001. Setting q = p + 5 * sqrt(p/total) is sufficient
to guarantee 0.9999 success rate for num > 12, but we need a
slightly larger q (9 empirically determined).
- when sampling without replacement, we're drawing each data point
with prob_i ~ Binomial(total, fraction) and our choice of q
guarantees 1-delta, or 0.9999 success rate, where success rate is
defined the same as in sampling with replacement.
"""
fraction = float(sampleSizeLowerBound) / total
if withReplacement:
numStDev = 5
if (sampleSizeLowerBound < 12):
numStDev = 9
return fraction + numStDev * sqrt(fraction / total)
else:
delta = 0.00005
gamma = - log(delta) / total
return min(1, fraction + gamma + sqrt(gamma * gamma + 2 * gamma * fraction))
def union(self, other):
"""
Return the union of this RDD and another one.
>>> rdd = sc.parallelize([1, 1, 2, 3])
>>> rdd.union(rdd).collect()
[1, 1, 2, 3, 1, 1, 2, 3]
"""
if self._jrdd_deserializer == other._jrdd_deserializer:
rdd = RDD(self._jrdd.union(other._jrdd), self.ctx,
self._jrdd_deserializer)
return rdd
else:
# These RDDs contain data in different serialized formats, so we
# must normalize them to the default serializer.
self_copy = self._reserialize()
other_copy = other._reserialize()
return RDD(self_copy._jrdd.union(other_copy._jrdd), self.ctx,
self.ctx.serializer)
def intersection(self, other):
"""
Return the intersection of this RDD and another one. The output will
not contain any duplicate elements, even if the input RDDs did.
Note that this method performs a shuffle internally.
>>> rdd1 = sc.parallelize([1, 10, 2, 3, 4, 5])
>>> rdd2 = sc.parallelize([1, 6, 2, 3, 7, 8])
>>> rdd1.intersection(rdd2).collect()
[1, 2, 3]
"""
return self.map(lambda v: (v, None)) \
.cogroup(other.map(lambda v: (v, None))) \
.filter(lambda x: (len(x[1][0]) != 0) and (len(x[1][1]) != 0)) \
.keys()
def _reserialize(self, serializer=None):
serializer = serializer or self.ctx.serializer
if self._jrdd_deserializer == serializer:
return self
else:
converted = self.map(lambda x: x, preservesPartitioning=True)
converted._jrdd_deserializer = serializer
return converted
def __add__(self, other):
"""
Return the union of this RDD and another one.
>>> rdd = sc.parallelize([1, 1, 2, 3])
>>> (rdd + rdd).collect()
[1, 1, 2, 3, 1, 1, 2, 3]
"""
if not isinstance(other, RDD):
raise TypeError
return self.union(other)
def sortByKey(self, ascending=True, numPartitions=None, keyfunc=lambda x: x):
"""
Sorts this RDD, which is assumed to consist of (key, value) pairs.
# noqa
>>> tmp = [('a', 1), ('b', 2), ('1', 3), ('d', 4), ('2', 5)]
>>> sc.parallelize(tmp).sortByKey().first()
('1', 3)
>>> sc.parallelize(tmp).sortByKey(True, 1).collect()
[('1', 3), ('2', 5), ('a', 1), ('b', 2), ('d', 4)]
>>> sc.parallelize(tmp).sortByKey(True, 2).collect()
[('1', 3), ('2', 5), ('a', 1), ('b', 2), ('d', 4)]
>>> tmp2 = [('Mary', 1), ('had', 2), ('a', 3), ('little', 4), ('lamb', 5)]
>>> tmp2.extend([('whose', 6), ('fleece', 7), ('was', 8), ('white', 9)])
>>> sc.parallelize(tmp2).sortByKey(True, 3, keyfunc=lambda k: k.lower()).collect()
[('a', 3), ('fleece', 7), ('had', 2), ('lamb', 5),...('white', 9), ('whose', 6)]
"""
if numPartitions is None:
numPartitions = self._defaultReducePartitions()
def sortPartition(iterator):
return iter(sorted(iterator, key=lambda (k, v): keyfunc(k), reverse=not ascending))
if numPartitions == 1:
if self.getNumPartitions() > 1:
self = self.coalesce(1)
return self.mapPartitions(sortPartition)
# first compute the boundary of each part via sampling: we want to partition
# the key-space into bins such that the bins have roughly the same
# number of (key, value) pairs falling into them
rddSize = self.count()
maxSampleSize = numPartitions * 20.0 # constant from Spark's RangePartitioner
fraction = min(maxSampleSize / max(rddSize, 1), 1.0)
samples = self.sample(False, fraction, 1).map(lambda (k, v): k).collect()
samples = sorted(samples, reverse=(not ascending), key=keyfunc)
# we have numPartitions many parts but one of the them has
# an implicit boundary
bounds = [samples[len(samples) * (i + 1) / numPartitions]
for i in range(0, numPartitions - 1)]
def rangePartitioner(k):
p = bisect.bisect_left(bounds, keyfunc(k))
if ascending:
return p
else:
return numPartitions - 1 - p
return self.partitionBy(numPartitions, rangePartitioner).mapPartitions(sortPartition, True)
def sortBy(self, keyfunc, ascending=True, numPartitions=None):
"""
Sorts this RDD by the given keyfunc
>>> tmp = [('a', 1), ('b', 2), ('1', 3), ('d', 4), ('2', 5)]
>>> sc.parallelize(tmp).sortBy(lambda x: x[0]).collect()
[('1', 3), ('2', 5), ('a', 1), ('b', 2), ('d', 4)]
>>> sc.parallelize(tmp).sortBy(lambda x: x[1]).collect()
[('a', 1), ('b', 2), ('1', 3), ('d', 4), ('2', 5)]
"""
return self.keyBy(keyfunc).sortByKey(ascending, numPartitions).values()
def glom(self):
"""
Return an RDD created by coalescing all elements within each partition
into a list.
>>> rdd = sc.parallelize([1, 2, 3, 4], 2)
>>> sorted(rdd.glom().collect())
[[1, 2], [3, 4]]
"""
def func(iterator):
yield list(iterator)
return self.mapPartitions(func)
def cartesian(self, other):
"""
Return the Cartesian product of this RDD and another one, that is, the
RDD of all pairs of elements C{(a, b)} where C{a} is in C{self} and
C{b} is in C{other}.
>>> rdd = sc.parallelize([1, 2])
>>> sorted(rdd.cartesian(rdd).collect())
[(1, 1), (1, 2), (2, 1), (2, 2)]
"""
# Due to batching, we can't use the Java cartesian method.
deserializer = CartesianDeserializer(self._jrdd_deserializer,
other._jrdd_deserializer)
return RDD(self._jrdd.cartesian(other._jrdd), self.ctx, deserializer)
def groupBy(self, f, numPartitions=None):
"""
Return an RDD of grouped items.
>>> rdd = sc.parallelize([1, 1, 2, 3, 5, 8])
>>> result = rdd.groupBy(lambda x: x % 2).collect()
>>> sorted([(x, sorted(y)) for (x, y) in result])
[(0, [2, 8]), (1, [1, 1, 3, 5])]
"""
return self.map(lambda x: (f(x), x)).groupByKey(numPartitions)
def pipe(self, command, env={}):
"""
Return an RDD created by piping elements to a forked external process.
>>> sc.parallelize(['1', '2', '', '3']).pipe('cat').collect()
['1', '2', '', '3']
"""
def func(iterator):
pipe = Popen(
shlex.split(command), env=env, stdin=PIPE, stdout=PIPE)
def pipe_objs(out):
for obj in iterator:
out.write(str(obj).rstrip('\n') + '\n')
out.close()
Thread(target=pipe_objs, args=[pipe.stdin]).start()
return (x.rstrip('\n') for x in iter(pipe.stdout.readline, ''))
return self.mapPartitions(func)
def foreach(self, f):
"""
Applies a function to all elements of this RDD.
>>> def f(x): print x
>>> sc.parallelize([1, 2, 3, 4, 5]).foreach(f)
"""
def processPartition(iterator):
for x in iterator:
f(x)
yield None
self.mapPartitions(processPartition).collect() # Force evaluation
def foreachPartition(self, f):
"""
Applies a function to each partition of this RDD.
>>> def f(iterator):
... for x in iterator:
... print x
... yield None
>>> sc.parallelize([1, 2, 3, 4, 5]).foreachPartition(f)
"""
self.mapPartitions(f).collect() # Force evaluation
def collect(self):
"""
Return a list that contains all of the elements in this RDD.
"""
with _JavaStackTrace(self.context) as st:
bytesInJava = self._jrdd.collect().iterator()
return list(self._collect_iterator_through_file(bytesInJava))
def _collect_iterator_through_file(self, iterator):
# Transferring lots of data through Py4J can be slow because
# socket.readline() is inefficient. Instead, we'll dump the data to a
# file and read it back.
tempFile = NamedTemporaryFile(delete=False, dir=self.ctx._temp_dir)
tempFile.close()
self.ctx._writeToFile(iterator, tempFile.name)
# Read the data into Python and deserialize it:
with open(tempFile.name, 'rb') as tempFile:
for item in self._jrdd_deserializer.load_stream(tempFile):
yield item
os.unlink(tempFile.name)
def reduce(self, f):
"""
Reduces the elements of this RDD using the specified commutative and
associative binary operator. Currently reduces partitions locally.
>>> from operator import add
>>> sc.parallelize([1, 2, 3, 4, 5]).reduce(add)
15
>>> sc.parallelize((2 for _ in range(10))).map(lambda x: 1).cache().reduce(add)
10
"""
def func(iterator):
acc = None
for obj in iterator:
if acc is None:
acc = obj
else:
acc = f(obj, acc)
if acc is not None:
yield acc
vals = self.mapPartitions(func).collect()
return reduce(f, vals)
def fold(self, zeroValue, op):
"""
Aggregate the elements of each partition, and then the results for all
the partitions, using a given associative function and a neutral "zero
value."
The function C{op(t1, t2)} is allowed to modify C{t1} and return it
as its result value to avoid object allocation; however, it should not
modify C{t2}.
>>> from operator import add
>>> sc.parallelize([1, 2, 3, 4, 5]).fold(0, add)
15
"""
def func(iterator):
acc = zeroValue
for obj in iterator:
acc = op(obj, acc)
yield acc
vals = self.mapPartitions(func).collect()
return reduce(op, vals, zeroValue)
def aggregate(self, zeroValue, seqOp, combOp):
"""
Aggregate the elements of each partition, and then the results for all
the partitions, using a given combine functions and a neutral "zero
value."
The functions C{op(t1, t2)} is allowed to modify C{t1} and return it
as its result value to avoid object allocation; however, it should not
modify C{t2}.
The first function (seqOp) can return a different result type, U, than
the type of this RDD. Thus, we need one operation for merging a T into
an U and one operation for merging two U
>>> seqOp = (lambda x, y: (x[0] + y, x[1] + 1))
>>> combOp = (lambda x, y: (x[0] + y[0], x[1] + y[1]))
>>> sc.parallelize([1, 2, 3, 4]).aggregate((0, 0), seqOp, combOp)
(10, 4)
>>> sc.parallelize([]).aggregate((0, 0), seqOp, combOp)
(0, 0)
"""
def func(iterator):
acc = zeroValue
for obj in iterator:
acc = seqOp(acc, obj)
yield acc
return self.mapPartitions(func).fold(zeroValue, combOp)
def max(self):
"""
Find the maximum item in this RDD.
>>> sc.parallelize([1.0, 5.0, 43.0, 10.0]).max()
43.0
"""
return self.reduce(max)
def min(self):
"""
Find the minimum item in this RDD.
>>> sc.parallelize([1.0, 5.0, 43.0, 10.0]).min()
1.0
"""
return self.reduce(min)
def sum(self):
"""
Add up the elements in this RDD.
>>> sc.parallelize([1.0, 2.0, 3.0]).sum()
6.0
"""
return self.mapPartitions(lambda x: [sum(x)]).reduce(operator.add)
def count(self):
"""
Return the number of elements in this RDD.
>>> sc.parallelize([2, 3, 4]).count()
3
"""
return self.mapPartitions(lambda i: [sum(1 for _ in i)]).sum()
def stats(self):
"""
Return a L{StatCounter} object that captures the mean, variance
and count of the RDD's elements in one operation.
"""
def redFunc(left_counter, right_counter):
return left_counter.mergeStats(right_counter)
return self.mapPartitions(lambda i: [StatCounter(i)]).reduce(redFunc)
def histogram(self, buckets):
"""
Compute a histogram using the provided buckets. The buckets
are all open to the right except for the last which is closed.
e.g. [1,10,20,50] means the buckets are [1,10) [10,20) [20,50],
which means 1<=x<10, 10<=x<20, 20<=x<=50. And on the input of 1
and 50 we would have a histogram of 1,0,1.
If your histogram is evenly spaced (e.g. [0, 10, 20, 30]),
this can be switched from an O(log n) inseration to O(1) per
element(where n = # buckets).
Buckets must be sorted and not contain any duplicates, must be
at least two elements.
If `buckets` is a number, it will generates buckets which are
evenly spaced between the minimum and maximum of the RDD. For
example, if the min value is 0 and the max is 100, given buckets
as 2, the resulting buckets will be [0,50) [50,100]. buckets must
be at least 1 If the RDD contains infinity, NaN throws an exception
If the elements in RDD do not vary (max == min) always returns
a single bucket.
It will return an tuple of buckets and histogram.
>>> rdd = sc.parallelize(range(51))
>>> rdd.histogram(2)
([0, 25, 50], [25, 26])
>>> rdd.histogram([0, 5, 25, 50])
([0, 5, 25, 50], [5, 20, 26])
>>> rdd.histogram([0, 15, 30, 45, 60]) # evenly spaced buckets
([0, 15, 30, 45, 60], [15, 15, 15, 6])
>>> rdd = sc.parallelize(["ab", "ac", "b", "bd", "ef"])
>>> rdd.histogram(("a", "b", "c"))
(('a', 'b', 'c'), [2, 2])
"""
if isinstance(buckets, (int, long)):
if buckets < 1:
raise ValueError("number of buckets must be >= 1")
# filter out non-comparable elements
def comparable(x):
if x is None:
return False
if type(x) is float and isnan(x):
return False
return True
filtered = self.filter(comparable)
# faster than stats()
def minmax(a, b):
return min(a[0], b[0]), max(a[1], b[1])
try:
minv, maxv = filtered.map(lambda x: (x, x)).reduce(minmax)
except TypeError as e:
if " empty " in str(e):
raise ValueError("can not generate buckets from empty RDD")
raise
if minv == maxv or buckets == 1:
return [minv, maxv], [filtered.count()]
try:
inc = (maxv - minv) / buckets
except TypeError:
raise TypeError("Can not generate buckets with non-number in RDD")
if isinf(inc):
raise ValueError("Can not generate buckets with infinite value")
# keep them as integer if possible
if inc * buckets != maxv - minv:
inc = (maxv - minv) * 1.0 / buckets
buckets = [i * inc + minv for i in range(buckets)]
buckets.append(maxv) # fix accumulated error
even = True
elif isinstance(buckets, (list, tuple)):
if len(buckets) < 2:
raise ValueError("buckets should have more than one value")
if any(i is None or isinstance(i, float) and isnan(i) for i in buckets):
raise ValueError("can not have None or NaN in buckets")
if sorted(buckets) != list(buckets):
raise ValueError("buckets should be sorted")
if len(set(buckets)) != len(buckets):
raise ValueError("buckets should not contain duplicated values")
minv = buckets[0]
maxv = buckets[-1]
even = False
inc = None
try:
steps = [buckets[i + 1] - buckets[i] for i in range(len(buckets) - 1)]
except TypeError:
pass # objects in buckets do not support '-'
else:
if max(steps) - min(steps) < 1e-10: # handle precision errors
even = True
inc = (maxv - minv) / (len(buckets) - 1)
else:
raise TypeError("buckets should be a list or tuple or number(int or long)")
def histogram(iterator):
counters = [0] * len(buckets)
for i in iterator:
if i is None or (type(i) is float and isnan(i)) or i > maxv or i < minv:
continue
t = (int((i - minv) / inc) if even
else bisect.bisect_right(buckets, i) - 1)
counters[t] += 1
# add last two together
last = counters.pop()
counters[-1] += last
return [counters]
def mergeCounters(a, b):
return [i + j for i, j in zip(a, b)]
return buckets, self.mapPartitions(histogram).reduce(mergeCounters)
def mean(self):
"""
Compute the mean of this RDD's elements.
>>> sc.parallelize([1, 2, 3]).mean()
2.0
"""
return self.stats().mean()
def variance(self):
"""
Compute the variance of this RDD's elements.
>>> sc.parallelize([1, 2, 3]).variance()
0.666...
"""
return self.stats().variance()
def stdev(self):
"""
Compute the standard deviation of this RDD's elements.
>>> sc.parallelize([1, 2, 3]).stdev()
0.816...
"""
return self.stats().stdev()
def sampleStdev(self):
"""
Compute the sample standard deviation of this RDD's elements (which
corrects for bias in estimating the standard deviation by dividing by
N-1 instead of N).
>>> sc.parallelize([1, 2, 3]).sampleStdev()
1.0
"""
return self.stats().sampleStdev()
def sampleVariance(self):
"""
Compute the sample variance of this RDD's elements (which corrects
for bias in estimating the variance by dividing by N-1 instead of N).
>>> sc.parallelize([1, 2, 3]).sampleVariance()
1.0
"""
return self.stats().sampleVariance()
def countByValue(self):
"""
Return the count of each unique value in this RDD as a dictionary of
(value, count) pairs.
>>> sorted(sc.parallelize([1, 2, 1, 2, 2], 2).countByValue().items())
[(1, 2), (2, 3)]
"""
def countPartition(iterator):
counts = defaultdict(int)
for obj in iterator:
counts[obj] += 1
yield counts
def mergeMaps(m1, m2):
for (k, v) in m2.iteritems():
m1[k] += v
return m1
return self.mapPartitions(countPartition).reduce(mergeMaps)
def top(self, num):
"""
Get the top N elements from a RDD.
Note: It returns the list sorted in descending order.
>>> sc.parallelize([10, 4, 2, 12, 3]).top(1)
[12]
>>> sc.parallelize([2, 3, 4, 5, 6], 2).top(2)
[6, 5]
"""
def topIterator(iterator):
q = []
for k in iterator:
if len(q) < num:
heapq.heappush(q, k)
else:
heapq.heappushpop(q, k)
yield q
def merge(a, b):
return next(topIterator(a + b))
return sorted(self.mapPartitions(topIterator).reduce(merge), reverse=True)
def takeOrdered(self, num, key=None):
"""
Get the N elements from a RDD ordered in ascending order or as
specified by the optional key function.
>>> sc.parallelize([10, 1, 2, 9, 3, 4, 5, 6, 7]).takeOrdered(6)
[1, 2, 3, 4, 5, 6]
>>> sc.parallelize([10, 1, 2, 9, 3, 4, 5, 6, 7], 2).takeOrdered(6, key=lambda x: -x)
[10, 9, 7, 6, 5, 4]
"""
def topNKeyedElems(iterator, key_=None):
q = MaxHeapQ(num)
for k in iterator:
if key_ is not None:
k = (key_(k), k)
q.insert(k)
yield q.getElements()
def unKey(x, key_=None):
if key_ is not None:
x = [i[1] for i in x]
return x
def merge(a, b):
return next(topNKeyedElems(a + b))
result = self.mapPartitions(
lambda i: topNKeyedElems(i, key)).reduce(merge)
return sorted(unKey(result, key), key=key)
def take(self, num):
"""
Take the first num elements of the RDD.
It works by first scanning one partition, and use the results from
that partition to estimate the number of additional partitions needed
to satisfy the limit.
Translated from the Scala implementation in RDD#take().
>>> sc.parallelize([2, 3, 4, 5, 6]).cache().take(2)
[2, 3]
>>> sc.parallelize([2, 3, 4, 5, 6]).take(10)
[2, 3, 4, 5, 6]
>>> sc.parallelize(range(100), 100).filter(lambda x: x > 90).take(3)
[91, 92, 93]
"""
items = []
totalParts = self._jrdd.partitions().size()
partsScanned = 0
while len(items) < num and partsScanned < totalParts:
# The number of partitions to try in this iteration.
# It is ok for this number to be greater than totalParts because
# we actually cap it at totalParts in runJob.
numPartsToTry = 1
if partsScanned > 0:
# If we didn't find any rows after the previous iteration,
# quadruple and retry. Otherwise, interpolate the number of
# partitions we need to try, but overestimate it by 50%.
if len(items) == 0:
numPartsToTry = partsScanned * 4
else:
numPartsToTry = int(1.5 * num * partsScanned / len(items))
left = num - len(items)
def takeUpToNumLeft(iterator):
taken = 0
while taken < left:
yield next(iterator)
taken += 1
p = range(
partsScanned, min(partsScanned + numPartsToTry, totalParts))
res = self.context.runJob(self, takeUpToNumLeft, p, True)
items += res
partsScanned += numPartsToTry
return items[:num]
def first(self):
"""
Return the first element in this RDD.
>>> sc.parallelize([2, 3, 4]).first()
2
"""
return self.take(1)[0]
def saveAsNewAPIHadoopDataset(self, conf, keyConverter=None, valueConverter=None):
"""
Output a Python RDD of key-value pairs (of form C{RDD[(K, V)]}) to any Hadoop file
system, using the new Hadoop OutputFormat API (mapreduce package). Keys/values are
converted for output using either user specified converters or, by default,
L{org.apache.spark.api.python.JavaToWritableConverter}.
@param conf: Hadoop job configuration, passed in as a dict
@param keyConverter: (None by default)
@param valueConverter: (None by default)
"""
jconf = self.ctx._dictToJavaMap(conf)
pickledRDD = self._toPickleSerialization()
batched = isinstance(pickledRDD._jrdd_deserializer, BatchedSerializer)
self.ctx._jvm.PythonRDD.saveAsHadoopDataset(pickledRDD._jrdd, batched, jconf,
keyConverter, valueConverter, True)
def saveAsNewAPIHadoopFile(self, path, outputFormatClass, keyClass=None, valueClass=None,
keyConverter=None, valueConverter=None, conf=None):
"""
Output a Python RDD of key-value pairs (of form C{RDD[(K, V)]}) to any Hadoop file
system, using the new Hadoop OutputFormat API (mapreduce package). Key and value types
will be inferred if not specified. Keys and values are converted for output using either
user specified converters or L{org.apache.spark.api.python.JavaToWritableConverter}. The
C{conf} is applied on top of the base Hadoop conf associated with the SparkContext
of this RDD to create a merged Hadoop MapReduce job configuration for saving the data.
@param path: path to Hadoop file
@param outputFormatClass: fully qualified classname of Hadoop OutputFormat
(e.g. "org.apache.hadoop.mapreduce.lib.output.SequenceFileOutputFormat")
@param keyClass: fully qualified classname of key Writable class
(e.g. "org.apache.hadoop.io.IntWritable", None by default)
@param valueClass: fully qualified classname of value Writable class
(e.g. "org.apache.hadoop.io.Text", None by default)
@param keyConverter: (None by default)
@param valueConverter: (None by default)
@param conf: Hadoop job configuration, passed in as a dict (None by default)
"""
jconf = self.ctx._dictToJavaMap(conf)
pickledRDD = self._toPickleSerialization()
batched = isinstance(pickledRDD._jrdd_deserializer, BatchedSerializer)
self.ctx._jvm.PythonRDD.saveAsNewAPIHadoopFile(pickledRDD._jrdd, batched, path,
outputFormatClass,
keyClass, valueClass,
keyConverter, valueConverter, jconf)
def saveAsHadoopDataset(self, conf, keyConverter=None, valueConverter=None):
"""
Output a Python RDD of key-value pairs (of form C{RDD[(K, V)]}) to any Hadoop file
system, using the old Hadoop OutputFormat API (mapred package). Keys/values are
converted for output using either user specified converters or, by default,
L{org.apache.spark.api.python.JavaToWritableConverter}.
@param conf: Hadoop job configuration, passed in as a dict
@param keyConverter: (None by default)
@param valueConverter: (None by default)
"""
jconf = self.ctx._dictToJavaMap(conf)
pickledRDD = self._toPickleSerialization()
batched = isinstance(pickledRDD._jrdd_deserializer, BatchedSerializer)
self.ctx._jvm.PythonRDD.saveAsHadoopDataset(pickledRDD._jrdd, batched, jconf,
keyConverter, valueConverter, False)
def saveAsHadoopFile(self, path, outputFormatClass, keyClass=None, valueClass=None,
keyConverter=None, valueConverter=None, conf=None,
compressionCodecClass=None):
"""
Output a Python RDD of key-value pairs (of form C{RDD[(K, V)]}) to any Hadoop file
system, using the old Hadoop OutputFormat API (mapred package). Key and value types
will be inferred if not specified. Keys and values are converted for output using either
user specified converters or L{org.apache.spark.api.python.JavaToWritableConverter}. The
C{conf} is applied on top of the base Hadoop conf associated with the SparkContext
of this RDD to create a merged Hadoop MapReduce job configuration for saving the data.
@param path: path to Hadoop file
@param outputFormatClass: fully qualified classname of Hadoop OutputFormat
(e.g. "org.apache.hadoop.mapred.SequenceFileOutputFormat")
@param keyClass: fully qualified classname of key Writable class
(e.g. "org.apache.hadoop.io.IntWritable", None by default)
@param valueClass: fully qualified classname of value Writable class
(e.g. "org.apache.hadoop.io.Text", None by default)
@param keyConverter: (None by default)
@param valueConverter: (None by default)
@param conf: (None by default)
@param compressionCodecClass: (None by default)
"""
jconf = self.ctx._dictToJavaMap(conf)
pickledRDD = self._toPickleSerialization()
batched = isinstance(pickledRDD._jrdd_deserializer, BatchedSerializer)
self.ctx._jvm.PythonRDD.saveAsHadoopFile(pickledRDD._jrdd, batched, path,
outputFormatClass,
keyClass, valueClass,
keyConverter, valueConverter,
jconf, compressionCodecClass)
def saveAsSequenceFile(self, path, compressionCodecClass=None):
"""
Output a Python RDD of key-value pairs (of form C{RDD[(K, V)]}) to any Hadoop file
system, using the L{org.apache.hadoop.io.Writable} types that we convert from the
RDD's key and value types. The mechanism is as follows:
1. Pyrolite is used to convert pickled Python RDD into RDD of Java objects.
2. Keys and values of this Java RDD are converted to Writables and written out.
@param path: path to sequence file
@param compressionCodecClass: (None by default)
"""
pickledRDD = self._toPickleSerialization()
batched = isinstance(pickledRDD._jrdd_deserializer, BatchedSerializer)
self.ctx._jvm.PythonRDD.saveAsSequenceFile(pickledRDD._jrdd, batched,
path, compressionCodecClass)
def saveAsPickleFile(self, path, batchSize=10):
"""
Save this RDD as a SequenceFile of serialized objects. The serializer
used is L{pyspark.serializers.PickleSerializer}, default batch size
is 10.
>>> tmpFile = NamedTemporaryFile(delete=True)
>>> tmpFile.close()
>>> sc.parallelize([1, 2, 'spark', 'rdd']).saveAsPickleFile(tmpFile.name, 3)
>>> sorted(sc.pickleFile(tmpFile.name, 5).collect())
[1, 2, 'rdd', 'spark']
"""
self._reserialize(BatchedSerializer(PickleSerializer(),
batchSize))._jrdd.saveAsObjectFile(path)
def saveAsTextFile(self, path):
"""
Save this RDD as a text file, using string representations of elements.
>>> tempFile = NamedTemporaryFile(delete=True)
>>> tempFile.close()
>>> sc.parallelize(range(10)).saveAsTextFile(tempFile.name)
>>> from fileinput import input
>>> from glob import glob
>>> ''.join(sorted(input(glob(tempFile.name + "/part-0000*"))))
'0\\n1\\n2\\n3\\n4\\n5\\n6\\n7\\n8\\n9\\n'
Empty lines are tolerated when saving to text files.
>>> tempFile2 = NamedTemporaryFile(delete=True)
>>> tempFile2.close()
>>> sc.parallelize(['', 'foo', '', 'bar', '']).saveAsTextFile(tempFile2.name)
>>> ''.join(sorted(input(glob(tempFile2.name + "/part-0000*"))))
'\\n\\n\\nbar\\nfoo\\n'
"""
def func(split, iterator):
for x in iterator:
if not isinstance(x, basestring):
x = unicode(x)
if isinstance(x, unicode):
x = x.encode("utf-8")
yield x
keyed = self.mapPartitionsWithIndex(func)
keyed._bypass_serializer = True
keyed._jrdd.map(self.ctx._jvm.BytesToString()).saveAsTextFile(path)
# Pair functions
def collectAsMap(self):
"""
Return the key-value pairs in this RDD to the master as a dictionary.
>>> m = sc.parallelize([(1, 2), (3, 4)]).collectAsMap()
>>> m[1]
2
>>> m[3]
4
"""
return dict(self.collect())
def keys(self):
"""
Return an RDD with the keys of each tuple.
>>> m = sc.parallelize([(1, 2), (3, 4)]).keys()
>>> m.collect()
[1, 3]
"""
return self.map(lambda (k, v): k)
def values(self):
"""
Return an RDD with the values of each tuple.
>>> m = sc.parallelize([(1, 2), (3, 4)]).values()
>>> m.collect()
[2, 4]
"""
return self.map(lambda (k, v): v)
def reduceByKey(self, func, numPartitions=None):
"""
Merge the values for each key using an associative reduce function.
This will also perform the merging locally on each mapper before
sending results to a reducer, similarly to a "combiner" in MapReduce.
Output will be hash-partitioned with C{numPartitions} partitions, or
the default parallelism level if C{numPartitions} is not specified.
>>> from operator import add
>>> rdd = sc.parallelize([("a", 1), ("b", 1), ("a", 1)])
>>> sorted(rdd.reduceByKey(add).collect())
[('a', 2), ('b', 1)]
"""
return self.combineByKey(lambda x: x, func, func, numPartitions)
def reduceByKeyLocally(self, func):
"""
Merge the values for each key using an associative reduce function, but
return the results immediately to the master as a dictionary.
This will also perform the merging locally on each mapper before
sending results to a reducer, similarly to a "combiner" in MapReduce.
>>> from operator import add
>>> rdd = sc.parallelize([("a", 1), ("b", 1), ("a", 1)])
>>> sorted(rdd.reduceByKeyLocally(add).items())
[('a', 2), ('b', 1)]
"""
def reducePartition(iterator):
m = {}
for (k, v) in iterator:
m[k] = v if k not in m else func(m[k], v)
yield m
def mergeMaps(m1, m2):
for (k, v) in m2.iteritems():
m1[k] = v if k not in m1 else func(m1[k], v)
return m1
return self.mapPartitions(reducePartition).reduce(mergeMaps)
def countByKey(self):
"""
Count the number of elements for each key, and return the result to the
master as a dictionary.
>>> rdd = sc.parallelize([("a", 1), ("b", 1), ("a", 1)])
>>> sorted(rdd.countByKey().items())
[('a', 2), ('b', 1)]
"""
return self.map(lambda x: x[0]).countByValue()
def join(self, other, numPartitions=None):
"""
Return an RDD containing all pairs of elements with matching keys in
C{self} and C{other}.
Each pair of elements will be returned as a (k, (v1, v2)) tuple, where
(k, v1) is in C{self} and (k, v2) is in C{other}.
Performs a hash join across the cluster.
>>> x = sc.parallelize([("a", 1), ("b", 4)])
>>> y = sc.parallelize([("a", 2), ("a", 3)])
>>> sorted(x.join(y).collect())
[('a', (1, 2)), ('a', (1, 3))]
"""
return python_join(self, other, numPartitions)
def leftOuterJoin(self, other, numPartitions=None):
"""
Perform a left outer join of C{self} and C{other}.
For each element (k, v) in C{self}, the resulting RDD will either
contain all pairs (k, (v, w)) for w in C{other}, or the pair
(k, (v, None)) if no elements in other have key k.
Hash-partitions the resulting RDD into the given number of partitions.
>>> x = sc.parallelize([("a", 1), ("b", 4)])
>>> y = sc.parallelize([("a", 2)])
>>> sorted(x.leftOuterJoin(y).collect())
[('a', (1, 2)), ('b', (4, None))]
"""
return python_left_outer_join(self, other, numPartitions)
def rightOuterJoin(self, other, numPartitions=None):
"""
Perform a right outer join of C{self} and C{other}.
For each element (k, w) in C{other}, the resulting RDD will either
contain all pairs (k, (v, w)) for v in this, or the pair (k, (None, w))
if no elements in C{self} have key k.
Hash-partitions the resulting RDD into the given number of partitions.
>>> x = sc.parallelize([("a", 1), ("b", 4)])
>>> y = sc.parallelize([("a", 2)])
>>> sorted(y.rightOuterJoin(x).collect())
[('a', (2, 1)), ('b', (None, 4))]
"""
return python_right_outer_join(self, other, numPartitions)
# TODO: add option to control map-side combining
# portable_hash is used as default, because builtin hash of None is different
# cross machines.
def partitionBy(self, numPartitions, partitionFunc=portable_hash):
"""
Return a copy of the RDD partitioned using the specified partitioner.
>>> pairs = sc.parallelize([1, 2, 3, 4, 2, 4, 1]).map(lambda x: (x, x))
>>> sets = pairs.partitionBy(2).glom().collect()
>>> set(sets[0]).intersection(set(sets[1]))
set([])
"""
if numPartitions is None:
numPartitions = self._defaultReducePartitions()
# Transferring O(n) objects to Java is too expensive.
# Instead, we'll form the hash buckets in Python,
# transferring O(numPartitions) objects to Java.
# Each object is a (splitNumber, [objects]) pair.
# In order to avoid too huge objects, the objects are
# grouped into chunks.
outputSerializer = self.ctx._unbatched_serializer
limit = (_parse_memory(self.ctx._conf.get(
"spark.python.worker.memory", "512m")) / 2)
def add_shuffle_key(split, iterator):
buckets = defaultdict(list)
c, batch = 0, min(10 * numPartitions, 1000)
for (k, v) in iterator:
buckets[partitionFunc(k) % numPartitions].append((k, v))
c += 1
# check used memory and avg size of chunk of objects
if (c % 1000 == 0 and get_used_memory() > limit
or c > batch):
n, size = len(buckets), 0
for split in buckets.keys():
yield pack_long(split)
d = outputSerializer.dumps(buckets[split])
del buckets[split]
yield d
size += len(d)
avg = (size / n) >> 20
# let 1M < avg < 10M
if avg < 1:
batch *= 1.5
elif avg > 10:
batch = max(batch / 1.5, 1)
c = 0
for (split, items) in buckets.iteritems():
yield pack_long(split)
yield outputSerializer.dumps(items)
keyed = self.mapPartitionsWithIndex(add_shuffle_key)
keyed._bypass_serializer = True
with _JavaStackTrace(self.context) as st:
pairRDD = self.ctx._jvm.PairwiseRDD(
keyed._jrdd.rdd()).asJavaPairRDD()
partitioner = self.ctx._jvm.PythonPartitioner(numPartitions,
id(partitionFunc))
jrdd = pairRDD.partitionBy(partitioner).values()
rdd = RDD(jrdd, self.ctx, BatchedSerializer(outputSerializer))
# This is required so that id(partitionFunc) remains unique,
# even if partitionFunc is a lambda:
rdd._partitionFunc = partitionFunc
return rdd
# TODO: add control over map-side aggregation
def combineByKey(self, createCombiner, mergeValue, mergeCombiners,
numPartitions=None):
"""
Generic function to combine the elements for each key using a custom
set of aggregation functions.
Turns an RDD[(K, V)] into a result of type RDD[(K, C)], for a "combined
type" C. Note that V and C can be different -- for example, one might
group an RDD of type (Int, Int) into an RDD of type (Int, List[Int]).
Users provide three functions:
- C{createCombiner}, which turns a V into a C (e.g., creates
a one-element list)
- C{mergeValue}, to merge a V into a C (e.g., adds it to the end of
a list)
- C{mergeCombiners}, to combine two C's into a single one.
In addition, users can control the partitioning of the output RDD.
>>> x = sc.parallelize([("a", 1), ("b", 1), ("a", 1)])
>>> def f(x): return x
>>> def add(a, b): return a + str(b)
>>> sorted(x.combineByKey(str, add, add).collect())
[('a', '11'), ('b', '1')]
"""
if numPartitions is None:
numPartitions = self._defaultReducePartitions()
serializer = self.ctx.serializer
spill = (self.ctx._conf.get("spark.shuffle.spill", 'True').lower()
== 'true')
memory = _parse_memory(self.ctx._conf.get(
"spark.python.worker.memory", "512m"))
agg = Aggregator(createCombiner, mergeValue, mergeCombiners)
def combineLocally(iterator):
merger = ExternalMerger(agg, memory * 0.9, serializer) \
if spill else InMemoryMerger(agg)
merger.mergeValues(iterator)
return merger.iteritems()
locally_combined = self.mapPartitions(combineLocally)
shuffled = locally_combined.partitionBy(numPartitions)
def _mergeCombiners(iterator):
merger = ExternalMerger(agg, memory, serializer) \
if spill else InMemoryMerger(agg)
merger.mergeCombiners(iterator)
return merger.iteritems()
return shuffled.mapPartitions(_mergeCombiners)
def aggregateByKey(self, zeroValue, seqFunc, combFunc, numPartitions=None):
"""
Aggregate the values of each key, using given combine functions and a neutral
"zero value". This function can return a different result type, U, than the type
of the values in this RDD, V. Thus, we need one operation for merging a V into
a U and one operation for merging two U's, The former operation is used for merging
values within a partition, and the latter is used for merging values between
partitions. To avoid memory allocation, both of these functions are
allowed to modify and return their first argument instead of creating a new U.
"""
def createZero():
return copy.deepcopy(zeroValue)
return self.combineByKey(
lambda v: seqFunc(createZero(), v), seqFunc, combFunc, numPartitions)
def foldByKey(self, zeroValue, func, numPartitions=None):
"""
Merge the values for each key using an associative function "func"
and a neutral "zeroValue" which may be added to the result an
arbitrary number of times, and must not change the result
(e.g., 0 for addition, or 1 for multiplication.).
>>> rdd = sc.parallelize([("a", 1), ("b", 1), ("a", 1)])
>>> from operator import add
>>> rdd.foldByKey(0, add).collect()
[('a', 2), ('b', 1)]
"""
def createZero():
return copy.deepcopy(zeroValue)
return self.combineByKey(lambda v: func(createZero(), v), func, func, numPartitions)
# TODO: support variant with custom partitioner
def groupByKey(self, numPartitions=None):
"""
Group the values for each key in the RDD into a single sequence.
Hash-partitions the resulting RDD with into numPartitions partitions.
Note: If you are grouping in order to perform an aggregation (such as a
sum or average) over each key, using reduceByKey will provide much
better performance.
>>> x = sc.parallelize([("a", 1), ("b", 1), ("a", 1)])
>>> map((lambda (x,y): (x, list(y))), sorted(x.groupByKey().collect()))
[('a', [1, 1]), ('b', [1])]
"""
def createCombiner(x):
return [x]
def mergeValue(xs, x):
xs.append(x)
return xs
def mergeCombiners(a, b):
a.extend(b)
return a
return self.combineByKey(createCombiner, mergeValue, mergeCombiners,
numPartitions).mapValues(lambda x: ResultIterable(x))
# TODO: add tests
def flatMapValues(self, f):
"""
Pass each value in the key-value pair RDD through a flatMap function
without changing the keys; this also retains the original RDD's
partitioning.
>>> x = sc.parallelize([("a", ["x", "y", "z"]), ("b", ["p", "r"])])
>>> def f(x): return x
>>> x.flatMapValues(f).collect()
[('a', 'x'), ('a', 'y'), ('a', 'z'), ('b', 'p'), ('b', 'r')]
"""
flat_map_fn = lambda (k, v): ((k, x) for x in f(v))
return self.flatMap(flat_map_fn, preservesPartitioning=True)
def mapValues(self, f):
"""
Pass each value in the key-value pair RDD through a map function
without changing the keys; this also retains the original RDD's
partitioning.
>>> x = sc.parallelize([("a", ["apple", "banana", "lemon"]), ("b", ["grapes"])])
>>> def f(x): return len(x)
>>> x.mapValues(f).collect()
[('a', 3), ('b', 1)]
"""
map_values_fn = lambda (k, v): (k, f(v))
return self.map(map_values_fn, preservesPartitioning=True)
def groupWith(self, other, *others):
"""
Alias for cogroup but with support for multiple RDDs.
>>> w = sc.parallelize([("a", 5), ("b", 6)])
>>> x = sc.parallelize([("a", 1), ("b", 4)])
>>> y = sc.parallelize([("a", 2)])
>>> z = sc.parallelize([("b", 42)])
>>> map((lambda (x,y): (x, (list(y[0]), list(y[1]), list(y[2]), list(y[3])))), \
sorted(list(w.groupWith(x, y, z).collect())))
[('a', ([5], [1], [2], [])), ('b', ([6], [4], [], [42]))]
"""
return python_cogroup((self, other) + others, numPartitions=None)
# TODO: add variant with custom parittioner
def cogroup(self, other, numPartitions=None):
"""
For each key k in C{self} or C{other}, return a resulting RDD that
contains a tuple with the list of values for that key in C{self} as
well as C{other}.
>>> x = sc.parallelize([("a", 1), ("b", 4)])
>>> y = sc.parallelize([("a", 2)])
>>> map((lambda (x,y): (x, (list(y[0]), list(y[1])))), sorted(list(x.cogroup(y).collect())))
[('a', ([1], [2])), ('b', ([4], []))]
"""
return python_cogroup((self, other), numPartitions)
def sampleByKey(self, withReplacement, fractions, seed=None):
"""
Return a subset of this RDD sampled by key (via stratified sampling).
Create a sample of this RDD using variable sampling rates for
different keys as specified by fractions, a key to sampling rate map.
>>> fractions = {"a": 0.2, "b": 0.1}
>>> rdd = sc.parallelize(fractions.keys()).cartesian(sc.parallelize(range(0, 1000)))
>>> sample = dict(rdd.sampleByKey(False, fractions, 2).groupByKey().collect())
>>> 100 < len(sample["a"]) < 300 and 50 < len(sample["b"]) < 150
True
>>> max(sample["a"]) <= 999 and min(sample["a"]) >= 0
True
>>> max(sample["b"]) <= 999 and min(sample["b"]) >= 0
True
"""
for fraction in fractions.values():
assert fraction >= 0.0, "Negative fraction value: %s" % fraction
return self.mapPartitionsWithIndex(
RDDStratifiedSampler(withReplacement, fractions, seed).func, True)
def subtractByKey(self, other, numPartitions=None):
"""
Return each (key, value) pair in C{self} that has no pair with matching
key in C{other}.
>>> x = sc.parallelize([("a", 1), ("b", 4), ("b", 5), ("a", 2)])
>>> y = sc.parallelize([("a", 3), ("c", None)])
>>> sorted(x.subtractByKey(y).collect())
[('b', 4), ('b', 5)]
"""
def filter_func((key, vals)):
return len(vals[0]) > 0 and len(vals[1]) == 0
map_func = lambda (key, vals): [(key, val) for val in vals[0]]
return self.cogroup(other, numPartitions).filter(filter_func).flatMap(map_func)
def subtract(self, other, numPartitions=None):
"""
Return each value in C{self} that is not contained in C{other}.
>>> x = sc.parallelize([("a", 1), ("b", 4), ("b", 5), ("a", 3)])
>>> y = sc.parallelize([("a", 3), ("c", None)])
>>> sorted(x.subtract(y).collect())
[('a', 1), ('b', 4), ('b', 5)]
"""
# note: here 'True' is just a placeholder
rdd = other.map(lambda x: (x, True))
return self.map(lambda x: (x, True)).subtractByKey(rdd).map(lambda tpl: tpl[0])
def keyBy(self, f):
"""
Creates tuples of the elements in this RDD by applying C{f}.
>>> x = sc.parallelize(range(0,3)).keyBy(lambda x: x*x)
>>> y = sc.parallelize(zip(range(0,5), range(0,5)))
>>> map((lambda (x,y): (x, (list(y[0]), (list(y[1]))))), sorted(x.cogroup(y).collect()))
[(0, ([0], [0])), (1, ([1], [1])), (2, ([], [2])), (3, ([], [3])), (4, ([2], [4]))]
"""
return self.map(lambda x: (f(x), x))
def repartition(self, numPartitions):
"""
Return a new RDD that has exactly numPartitions partitions.
Can increase or decrease the level of parallelism in this RDD.
Internally, this uses a shuffle to redistribute data.
If you are decreasing the number of partitions in this RDD, consider
using `coalesce`, which can avoid performing a shuffle.
>>> rdd = sc.parallelize([1,2,3,4,5,6,7], 4)
>>> sorted(rdd.glom().collect())
[[1], [2, 3], [4, 5], [6, 7]]
>>> len(rdd.repartition(2).glom().collect())
2
>>> len(rdd.repartition(10).glom().collect())
10
"""
jrdd = self._jrdd.repartition(numPartitions)
return RDD(jrdd, self.ctx, self._jrdd_deserializer)
def coalesce(self, numPartitions, shuffle=False):
"""
Return a new RDD that is reduced into `numPartitions` partitions.
>>> sc.parallelize([1, 2, 3, 4, 5], 3).glom().collect()
[[1], [2, 3], [4, 5]]
>>> sc.parallelize([1, 2, 3, 4, 5], 3).coalesce(1).glom().collect()
[[1, 2, 3, 4, 5]]
"""
jrdd = self._jrdd.coalesce(numPartitions)
return RDD(jrdd, self.ctx, self._jrdd_deserializer)
def zip(self, other):
"""
Zips this RDD with another one, returning key-value pairs with the
first element in each RDD second element in each RDD, etc. Assumes
that the two RDDs have the same number of partitions and the same
number of elements in each partition (e.g. one was made through
a map on the other).
>>> x = sc.parallelize(range(0,5))
>>> y = sc.parallelize(range(1000, 1005))
>>> x.zip(y).collect()
[(0, 1000), (1, 1001), (2, 1002), (3, 1003), (4, 1004)]
"""
if self.getNumPartitions() != other.getNumPartitions():
raise ValueError("Can only zip with RDD which has the same number of partitions")
def get_batch_size(ser):
if isinstance(ser, BatchedSerializer):
return ser.batchSize
return 0
def batch_as(rdd, batchSize):
ser = rdd._jrdd_deserializer
if isinstance(ser, BatchedSerializer):
ser = ser.serializer
return rdd._reserialize(BatchedSerializer(ser, batchSize))
my_batch = get_batch_size(self._jrdd_deserializer)
other_batch = get_batch_size(other._jrdd_deserializer)
if my_batch != other_batch:
# use the greatest batchSize to batch the other one.
if my_batch > other_batch:
other = batch_as(other, my_batch)
else:
self = batch_as(self, other_batch)
# There will be an Exception in JVM if there are different number
# of items in each partitions.
pairRDD = self._jrdd.zip(other._jrdd)
deserializer = PairDeserializer(self._jrdd_deserializer,
other._jrdd_deserializer)
return RDD(pairRDD, self.ctx, deserializer)
def zipWithIndex(self):
"""
Zips this RDD with its element indices.
The ordering is first based on the partition index and then the
ordering of items within each partition. So the first item in
the first partition gets index 0, and the last item in the last
partition receives the largest index.
This method needs to trigger a spark job when this RDD contains
more than one partitions.
>>> sc.parallelize(["a", "b", "c", "d"], 3).zipWithIndex().collect()
[('a', 0), ('b', 1), ('c', 2), ('d', 3)]
"""
starts = [0]
if self.getNumPartitions() > 1:
nums = self.mapPartitions(lambda it: [sum(1 for i in it)]).collect()
for i in range(len(nums) - 1):
starts.append(starts[-1] + nums[i])
def func(k, it):
for i, v in enumerate(it, starts[k]):
yield v, i
return self.mapPartitionsWithIndex(func)
def zipWithUniqueId(self):
"""
Zips this RDD with generated unique Long ids.
Items in the kth partition will get ids k, n+k, 2*n+k, ..., where
n is the number of partitions. So there may exist gaps, but this
method won't trigger a spark job, which is different from
L{zipWithIndex}
>>> sc.parallelize(["a", "b", "c", "d", "e"], 3).zipWithUniqueId().collect()
[('a', 0), ('b', 1), ('c', 4), ('d', 2), ('e', 5)]
"""
n = self.getNumPartitions()
def func(k, it):
for i, v in enumerate(it):
yield v, i * n + k
return self.mapPartitionsWithIndex(func)
def name(self):
"""
Return the name of this RDD.
"""
name_ = self._jrdd.name()
if not name_:
return None
return name_.encode('utf-8')
def setName(self, name):
"""
Assign a name to this RDD.
>>> rdd1 = sc.parallelize([1,2])
>>> rdd1.setName('RDD1')
>>> rdd1.name()
'RDD1'
"""
self._jrdd.setName(name)
def toDebugString(self):
"""
A description of this RDD and its recursive dependencies for debugging.
"""
debug_string = self._jrdd.toDebugString()
if not debug_string:
return None
return debug_string.encode('utf-8')
def getStorageLevel(self):
"""
Get the RDD's current storage level.
>>> rdd1 = sc.parallelize([1,2])
>>> rdd1.getStorageLevel()
StorageLevel(False, False, False, False, 1)
>>> print(rdd1.getStorageLevel())
Serialized 1x Replicated
"""
java_storage_level = self._jrdd.getStorageLevel()
storage_level = StorageLevel(java_storage_level.useDisk(),
java_storage_level.useMemory(),
java_storage_level.useOffHeap(),
java_storage_level.deserialized(),
java_storage_level.replication())
return storage_level
def _defaultReducePartitions(self):
"""
Returns the default number of partitions to use during reduce tasks (e.g., groupBy).
If spark.default.parallelism is set, then we'll use the value from SparkContext
defaultParallelism, otherwise we'll use the number of partitions in this RDD.
This mirrors the behavior of the Scala Partitioner#defaultPartitioner, intended to reduce
the likelihood of OOMs. Once PySpark adopts Partitioner-based APIs, this behavior will
be inherent.
"""
if self.ctx._conf.contains("spark.default.parallelism"):
return self.ctx.defaultParallelism
else:
return self.getNumPartitions()
# TODO: `lookup` is disabled because we can't make direct comparisons based
# on the key; we need to compare the hash of the key to the hash of the
# keys in the pairs. This could be an expensive operation, since those
# hashes aren't retained.
class PipelinedRDD(RDD):
"""
Pipelined maps:
>>> rdd = sc.parallelize([1, 2, 3, 4])
>>> rdd.map(lambda x: 2 * x).cache().map(lambda x: 2 * x).collect()
[4, 8, 12, 16]
>>> rdd.map(lambda x: 2 * x).map(lambda x: 2 * x).collect()
[4, 8, 12, 16]
Pipelined reduces:
>>> from operator import add
>>> rdd.map(lambda x: 2 * x).reduce(add)
20
>>> rdd.flatMap(lambda x: [x, x]).reduce(add)
20
"""
def __init__(self, prev, func, preservesPartitioning=False):
if not isinstance(prev, PipelinedRDD) or not prev._is_pipelinable():
# This transformation is the first in its stage:
self.func = func
self.preservesPartitioning = preservesPartitioning
self._prev_jrdd = prev._jrdd
self._prev_jrdd_deserializer = prev._jrdd_deserializer
else:
prev_func = prev.func
def pipeline_func(split, iterator):
return func(split, prev_func(split, iterator))
self.func = pipeline_func
self.preservesPartitioning = \
prev.preservesPartitioning and preservesPartitioning
self._prev_jrdd = prev._prev_jrdd # maintain the pipeline
self._prev_jrdd_deserializer = prev._prev_jrdd_deserializer
self.is_cached = False
self.is_checkpointed = False
self.ctx = prev.ctx
self.prev = prev
self._jrdd_val = None
self._jrdd_deserializer = self.ctx.serializer
self._bypass_serializer = False
@property
def _jrdd(self):
if self._jrdd_val:
return self._jrdd_val
if self._bypass_serializer:
self._jrdd_deserializer = NoOpSerializer()
command = (self.func, self._prev_jrdd_deserializer,
self._jrdd_deserializer)
ser = CloudPickleSerializer()
pickled_command = ser.dumps(command)
broadcast_vars = ListConverter().convert(
[x._jbroadcast for x in self.ctx._pickled_broadcast_vars],
self.ctx._gateway._gateway_client)
self.ctx._pickled_broadcast_vars.clear()
env = MapConverter().convert(self.ctx.environment,
self.ctx._gateway._gateway_client)
includes = ListConverter().convert(self.ctx._python_includes,
self.ctx._gateway._gateway_client)
python_rdd = self.ctx._jvm.PythonRDD(self._prev_jrdd.rdd(),
bytearray(pickled_command),
env, includes, self.preservesPartitioning,
self.ctx.pythonExec,
broadcast_vars, self.ctx._javaAccumulator)
self._jrdd_val = python_rdd.asJavaRDD()
return self._jrdd_val
def _is_pipelinable(self):
return not (self.is_cached or self.is_checkpointed)
def _test():
import doctest
from pyspark.context import SparkContext
globs = globals().copy()
# The small batch size here ensures that we see multiple batches,
# even in these small test examples:
globs['sc'] = SparkContext('local[4]', 'PythonTest', batchSize=2)
(failure_count, test_count) = doctest.testmod(
globs=globs, optionflags=doctest.ELLIPSIS)
globs['sc'].stop()
if failure_count:
exit(-1)
if __name__ == "__main__":
_test()
|