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
|
// Protocol Buffers - Google's data interchange format
// Copyright 2008 Google Inc. All rights reserved.
// http://code.google.com/p/protobuf/
//
// Redistribution and use in source and binary forms, with or without
// modification, are permitted provided that the following conditions are
// met:
//
// * Redistributions of source code must retain the above copyright
// notice, this list of conditions and the following disclaimer.
// * Redistributions in binary form must reproduce the above
// copyright notice, this list of conditions and the following disclaimer
// in the documentation and/or other materials provided with the
// distribution.
// * Neither the name of Google Inc. nor the names of its
// contributors may be used to endorse or promote products derived from
// this software without specific prior written permission.
//
// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
// Author: kenton@google.com (Kenton Varda)
// Based on original Protocol Buffers design by
// Sanjay Ghemawat, Jeff Dean, and others.
#include <google/protobuf/compiler/command_line_interface.h>
#include <stdio.h>
#include <sys/types.h>
#include <sys/stat.h>
#include <fcntl.h>
#ifdef _MSC_VER
#include <io.h>
#include <direct.h>
#else
#include <unistd.h>
#endif
#include <errno.h>
#include <iostream>
#include <ctype.h>
#include <google/protobuf/compiler/importer.h>
#include <google/protobuf/compiler/code_generator.h>
#include <google/protobuf/compiler/plugin.pb.h>
#include <google/protobuf/compiler/subprocess.h>
#include <google/protobuf/compiler/zip_writer.h>
#include <google/protobuf/descriptor.h>
#include <google/protobuf/text_format.h>
#include <google/protobuf/dynamic_message.h>
#include <google/protobuf/io/zero_copy_stream_impl.h>
#include <google/protobuf/io/printer.h>
#include <google/protobuf/stubs/common.h>
#include <google/protobuf/stubs/strutil.h>
#include <google/protobuf/stubs/substitute.h>
#include <google/protobuf/stubs/map-util.h>
#include <google/protobuf/stubs/stl_util-inl.h>
#include <google/protobuf/stubs/hash.h>
namespace google {
namespace protobuf {
namespace compiler {
#if defined(_WIN32)
#define mkdir(name, mode) mkdir(name)
#ifndef W_OK
#define W_OK 02 // not defined by MSVC for whatever reason
#endif
#ifndef F_OK
#define F_OK 00 // not defined by MSVC for whatever reason
#endif
#ifndef STDIN_FILENO
#define STDIN_FILENO 0
#endif
#ifndef STDOUT_FILENO
#define STDOUT_FILENO 1
#endif
#endif
#ifndef O_BINARY
#ifdef _O_BINARY
#define O_BINARY _O_BINARY
#else
#define O_BINARY 0 // If this isn't defined, the platform doesn't need it.
#endif
#endif
namespace {
#if defined(_WIN32) && !defined(__CYGWIN__)
static const char* kPathSeparator = ";";
#else
static const char* kPathSeparator = ":";
#endif
// Returns true if the text looks like a Windows-style absolute path, starting
// with a drive letter. Example: "C:\foo". TODO(kenton): Share this with
// copy in importer.cc?
static bool IsWindowsAbsolutePath(const string& text) {
#if defined(_WIN32) || defined(__CYGWIN__)
return text.size() >= 3 && text[1] == ':' &&
isalpha(text[0]) &&
(text[2] == '/' || text[2] == '\\') &&
text.find_last_of(':') == 1;
#else
return false;
#endif
}
void SetFdToTextMode(int fd) {
#ifdef _WIN32
if (_setmode(fd, _O_TEXT) == -1) {
// This should never happen, I think.
GOOGLE_LOG(WARNING) << "_setmode(" << fd << ", _O_TEXT): " << strerror(errno);
}
#endif
// (Text and binary are the same on non-Windows platforms.)
}
void SetFdToBinaryMode(int fd) {
#ifdef _WIN32
if (_setmode(fd, _O_BINARY) == -1) {
// This should never happen, I think.
GOOGLE_LOG(WARNING) << "_setmode(" << fd << ", _O_BINARY): " << strerror(errno);
}
#endif
// (Text and binary are the same on non-Windows platforms.)
}
void AddTrailingSlash(string* path) {
if (!path->empty() && path->at(path->size() - 1) != '/') {
path->push_back('/');
}
}
bool VerifyDirectoryExists(const string& path) {
if (path.empty()) return true;
if (access(path.c_str(), W_OK) == -1) {
cerr << path << ": " << strerror(errno) << endl;
return false;
} else {
return true;
}
}
// Try to create the parent directory of the given file, creating the parent's
// parent if necessary, and so on. The full file name is actually
// (prefix + filename), but we assume |prefix| already exists and only create
// directories listed in |filename|.
bool TryCreateParentDirectory(const string& prefix, const string& filename) {
// Recursively create parent directories to the output file.
vector<string> parts;
SplitStringUsing(filename, "/", &parts);
string path_so_far = prefix;
for (int i = 0; i < parts.size() - 1; i++) {
path_so_far += parts[i];
if (mkdir(path_so_far.c_str(), 0777) != 0) {
if (errno != EEXIST) {
cerr << filename << ": while trying to create directory "
<< path_so_far << ": " << strerror(errno) << endl;
return false;
}
}
path_so_far += '/';
}
return true;
}
} // namespace
// A MultiFileErrorCollector that prints errors to stderr.
class CommandLineInterface::ErrorPrinter : public MultiFileErrorCollector,
public io::ErrorCollector {
public:
ErrorPrinter(ErrorFormat format) : format_(format) {}
~ErrorPrinter() {}
// implements MultiFileErrorCollector ------------------------------
void AddError(const string& filename, int line, int column,
const string& message) {
cerr << filename;
// Users typically expect 1-based line/column numbers, so we add 1
// to each here.
if (line != -1) {
// Allow for both GCC- and Visual-Studio-compatible output.
switch (format_) {
case CommandLineInterface::ERROR_FORMAT_GCC:
cerr << ":" << (line + 1) << ":" << (column + 1);
break;
case CommandLineInterface::ERROR_FORMAT_MSVS:
cerr << "(" << (line + 1) << ") : error in column=" << (column + 1);
break;
}
}
cerr << ": " << message << endl;
}
// implements io::ErrorCollector -----------------------------------
void AddError(int line, int column, const string& message) {
AddError("input", line, column, message);
}
private:
const ErrorFormat format_;
};
// -------------------------------------------------------------------
// An OutputDirectory implementation that buffers files in memory, then dumps
// them all to disk on demand.
class CommandLineInterface::MemoryOutputDirectory : public OutputDirectory {
public:
MemoryOutputDirectory();
~MemoryOutputDirectory();
// Write all files in the directory to disk at the given output location,
// which must end in a '/'.
bool WriteAllToDisk(const string& prefix);
// Write the contents of this directory to a ZIP-format archive with the
// given name.
bool WriteAllToZip(const string& filename);
// Add a boilerplate META-INF/MANIFEST.MF file as required by the Java JAR
// format, unless one has already been written.
void AddJarManifest();
// implements OutputDirectory --------------------------------------
io::ZeroCopyOutputStream* Open(const string& filename);
io::ZeroCopyOutputStream* OpenForInsert(
const string& filename, const string& insertion_point);
private:
friend class MemoryOutputStream;
// map instead of hash_map so that files are written in order (good when
// writing zips).
map<string, string*> files_;
bool had_error_;
};
class CommandLineInterface::MemoryOutputStream
: public io::ZeroCopyOutputStream {
public:
MemoryOutputStream(MemoryOutputDirectory* directory, const string& filename);
MemoryOutputStream(MemoryOutputDirectory* directory, const string& filename,
const string& insertion_point);
virtual ~MemoryOutputStream();
// implements ZeroCopyOutputStream ---------------------------------
virtual bool Next(void** data, int* size) { return inner_->Next(data, size); }
virtual void BackUp(int count) { inner_->BackUp(count); }
virtual int64 ByteCount() const { return inner_->ByteCount(); }
private:
// Where to insert the string when it's done.
MemoryOutputDirectory* directory_;
string filename_;
string insertion_point_;
// The string we're building.
string data_;
// StringOutputStream writing to data_.
scoped_ptr<io::StringOutputStream> inner_;
};
// -------------------------------------------------------------------
CommandLineInterface::MemoryOutputDirectory::MemoryOutputDirectory()
: had_error_(false) {}
CommandLineInterface::MemoryOutputDirectory::~MemoryOutputDirectory() {
STLDeleteValues(&files_);
}
bool CommandLineInterface::MemoryOutputDirectory::WriteAllToDisk(
const string& prefix) {
if (had_error_) {
return false;
}
if (!VerifyDirectoryExists(prefix)) {
return false;
}
for (map<string, string*>::const_iterator iter = files_.begin();
iter != files_.end(); ++iter) {
const string& relative_filename = iter->first;
const char* data = iter->second->data();
int size = iter->second->size();
if (!TryCreateParentDirectory(prefix, relative_filename)) {
return false;
}
string filename = prefix + relative_filename;
// Create the output file.
int file_descriptor;
do {
file_descriptor =
open(filename.c_str(), O_WRONLY | O_CREAT | O_TRUNC | O_BINARY, 0666);
} while (file_descriptor < 0 && errno == EINTR);
if (file_descriptor < 0) {
int error = errno;
cerr << filename << ": " << strerror(error);
return false;
}
// Write the file.
while (size > 0) {
int write_result;
do {
write_result = write(file_descriptor, data, size);
} while (write_result < 0 && errno == EINTR);
if (write_result <= 0) {
// Write error.
// FIXME(kenton): According to the man page, if write() returns zero,
// there was no error; write() simply did not write anything. It's
// unclear under what circumstances this might happen, but presumably
// errno won't be set in this case. I am confused as to how such an
// event should be handled. For now I'm treating it as an error,
// since retrying seems like it could lead to an infinite loop. I
// suspect this never actually happens anyway.
if (write_result < 0) {
int error = errno;
cerr << filename << ": write: " << strerror(error);
} else {
cerr << filename << ": write() returned zero?" << endl;
}
return false;
}
data += write_result;
size -= write_result;
}
if (close(file_descriptor) != 0) {
int error = errno;
cerr << filename << ": close: " << strerror(error);
return false;
}
}
return true;
}
bool CommandLineInterface::MemoryOutputDirectory::WriteAllToZip(
const string& filename) {
if (had_error_) {
return false;
}
// Create the output file.
int file_descriptor;
do {
file_descriptor =
open(filename.c_str(), O_WRONLY | O_CREAT | O_TRUNC | O_BINARY, 0666);
} while (file_descriptor < 0 && errno == EINTR);
if (file_descriptor < 0) {
int error = errno;
cerr << filename << ": " << strerror(error);
return false;
}
// Create the ZipWriter
io::FileOutputStream stream(file_descriptor);
ZipWriter zip_writer(&stream);
for (map<string, string*>::const_iterator iter = files_.begin();
iter != files_.end(); ++iter) {
zip_writer.Write(iter->first, *iter->second);
}
zip_writer.WriteDirectory();
if (stream.GetErrno() != 0) {
cerr << filename << ": " << strerror(stream.GetErrno()) << endl;
}
if (!stream.Close()) {
cerr << filename << ": " << strerror(stream.GetErrno()) << endl;
}
return true;
}
void CommandLineInterface::MemoryOutputDirectory::AddJarManifest() {
string** map_slot = &files_["META-INF/MANIFEST.MF"];
if (*map_slot == NULL) {
*map_slot = new string(
"Manifest-Version: 1.0\n"
"Created-By: 1.6.0 (protoc)\n"
"\n");
}
}
io::ZeroCopyOutputStream* CommandLineInterface::MemoryOutputDirectory::Open(
const string& filename) {
return new MemoryOutputStream(this, filename);
}
io::ZeroCopyOutputStream*
CommandLineInterface::MemoryOutputDirectory::OpenForInsert(
const string& filename, const string& insertion_point) {
return new MemoryOutputStream(this, filename, insertion_point);
}
// -------------------------------------------------------------------
CommandLineInterface::MemoryOutputStream::MemoryOutputStream(
MemoryOutputDirectory* directory, const string& filename)
: directory_(directory),
filename_(filename),
inner_(new io::StringOutputStream(&data_)) {
}
CommandLineInterface::MemoryOutputStream::MemoryOutputStream(
MemoryOutputDirectory* directory, const string& filename,
const string& insertion_point)
: directory_(directory),
filename_(filename),
insertion_point_(insertion_point),
inner_(new io::StringOutputStream(&data_)) {
}
CommandLineInterface::MemoryOutputStream::~MemoryOutputStream() {
// Make sure all data has been written.
inner_.reset();
// Insert into the directory.
string** map_slot = &directory_->files_[filename_];
if (insertion_point_.empty()) {
// This was just a regular Open().
if (*map_slot != NULL) {
cerr << filename_ << ": Tried to write the same file twice." << endl;
directory_->had_error_ = true;
return;
}
*map_slot = new string;
(*map_slot)->swap(data_);
} else {
// This was an OpenForInsert().
// If the data doens't end with a clean line break, add one.
if (!data_.empty() && data_[data_.size() - 1] != '\n') {
data_.push_back('\n');
}
// Find the file we are going to insert into.
if (*map_slot == NULL) {
cerr << filename_ << ": Tried to insert into file that doesn't exist."
<< endl;
directory_->had_error_ = true;
return;
}
string* target = *map_slot;
// Find the insertion point.
string magic_string = strings::Substitute(
"@@protoc_insertion_point($0)", insertion_point_);
string::size_type pos = target->find(magic_string);
if (pos == string::npos) {
cerr << filename_ << ": insertion point \"" << insertion_point_
<< "\" not found." << endl;
directory_->had_error_ = true;
return;
}
// Seek backwards to the beginning of the line, which is where we will
// insert the data. Note that this has the effect of pushing the insertion
// point down, so the data is inserted before it. This is intentional
// because it means that multiple insertions at the same point will end
// up in the expected order in the final output.
pos = target->find_last_of('\n', pos);
if (pos == string::npos) {
// Insertion point is on the first line.
pos = 0;
} else {
// Advance to character after '\n'.
++pos;
}
// Extract indent.
string indent_(*target, pos, target->find_first_not_of(" \t", pos) - pos);
if (indent_.empty()) {
// No indent. This makes things easier.
target->insert(pos, data_);
} else {
// Calculate how much space we need.
int indent_size = 0;
for (int i = 0; i < data_.size(); i++) {
if (data_[i] == '\n') indent_size += indent_.size();
}
// Make a hole for it.
target->insert(pos, data_.size() + indent_size, '\0');
// Now copy in the data.
string::size_type data_pos = 0;
char* target_ptr = string_as_array(target) + pos;
while (data_pos < data_.size()) {
// Copy indent.
memcpy(target_ptr, indent_.data(), indent_.size());
target_ptr += indent_.size();
// Copy line from data_.
// We already guaranteed that data_ ends with a newline (above), so this
// search can't fail.
string::size_type line_length =
data_.find_first_of('\n', data_pos) + 1 - data_pos;
memcpy(target_ptr, data_.data() + data_pos, line_length);
target_ptr += line_length;
data_pos += line_length;
}
GOOGLE_CHECK_EQ(target_ptr,
string_as_array(target) + pos + data_.size() + indent_size);
}
}
}
// ===================================================================
CommandLineInterface::CommandLineInterface()
: mode_(MODE_COMPILE),
error_format_(ERROR_FORMAT_GCC),
imports_in_descriptor_set_(false),
disallow_services_(false),
inputs_are_proto_path_relative_(false) {}
CommandLineInterface::~CommandLineInterface() {}
void CommandLineInterface::RegisterGenerator(const string& flag_name,
CodeGenerator* generator,
const string& help_text) {
GeneratorInfo info;
info.generator = generator;
info.help_text = help_text;
generators_[flag_name] = info;
}
void CommandLineInterface::AllowPlugins(const string& exe_name_prefix) {
plugin_prefix_ = exe_name_prefix;
}
int CommandLineInterface::Run(int argc, const char* const argv[]) {
Clear();
if (!ParseArguments(argc, argv)) return 1;
// Set up the source tree.
DiskSourceTree source_tree;
for (int i = 0; i < proto_path_.size(); i++) {
source_tree.MapPath(proto_path_[i].first, proto_path_[i].second);
}
// Map input files to virtual paths if necessary.
if (!inputs_are_proto_path_relative_) {
if (!MakeInputsBeProtoPathRelative(&source_tree)) {
return 1;
}
}
// Allocate the Importer.
ErrorPrinter error_collector(error_format_);
Importer importer(&source_tree, &error_collector);
vector<const FileDescriptor*> parsed_files;
// Parse each file.
for (int i = 0; i < input_files_.size(); i++) {
// Import the file.
const FileDescriptor* parsed_file = importer.Import(input_files_[i]);
if (parsed_file == NULL) return 1;
parsed_files.push_back(parsed_file);
// Enforce --disallow_services.
if (disallow_services_ && parsed_file->service_count() > 0) {
cerr << parsed_file->name() << ": This file contains services, but "
"--disallow_services was used." << endl;
return 1;
}
}
// We construct a separate OutputDirectory for each output location. Note
// that two code generators may output to the same location, in which case
// they should share a single OutputDirectory (so that OpenForInsert() works).
typedef hash_map<string, MemoryOutputDirectory*> OutputDirectoryMap;
OutputDirectoryMap output_directories_;
// Generate output.
MemoryOutputDirectory output_directory;
if (mode_ == MODE_COMPILE) {
for (int i = 0; i < output_directives_.size(); i++) {
string output_location = output_directives_[i].output_location;
if (!HasSuffixString(output_location, ".zip") &&
!HasSuffixString(output_location, ".jar")) {
AddTrailingSlash(&output_location);
}
MemoryOutputDirectory** map_slot = &output_directories_[output_location];
if (*map_slot == NULL) {
// First time we've seen this output location.
*map_slot = new MemoryOutputDirectory;
}
if (!GenerateOutput(parsed_files, output_directives_[i], *map_slot)) {
return 1;
}
}
}
// Write all output to disk.
for (OutputDirectoryMap::iterator iter = output_directories_.begin();
iter != output_directories_.end(); ++iter) {
const string& location = iter->first;
MemoryOutputDirectory* directory = iter->second;
if (HasSuffixString(location, "/")) {
if (!directory->WriteAllToDisk(location)) {
return 1;
}
} else {
if (HasSuffixString(location, ".jar")) {
directory->AddJarManifest();
}
if (!directory->WriteAllToZip(location)) {
return 1;
}
}
}
if (!descriptor_set_name_.empty()) {
if (!WriteDescriptorSet(parsed_files)) {
return 1;
}
}
if (mode_ == MODE_ENCODE || mode_ == MODE_DECODE) {
if (codec_type_.empty()) {
// HACK: Define an EmptyMessage type to use for decoding.
DescriptorPool pool;
FileDescriptorProto file;
file.set_name("empty_message.proto");
file.add_message_type()->set_name("EmptyMessage");
GOOGLE_CHECK(pool.BuildFile(file) != NULL);
codec_type_ = "EmptyMessage";
if (!EncodeOrDecode(&pool)) {
return 1;
}
} else {
if (!EncodeOrDecode(importer.pool())) {
return 1;
}
}
}
return 0;
}
void CommandLineInterface::Clear() {
// Clear all members that are set by Run(). Note that we must not clear
// members which are set by other methods before Run() is called.
executable_name_.clear();
proto_path_.clear();
input_files_.clear();
output_directives_.clear();
codec_type_.clear();
descriptor_set_name_.clear();
mode_ = MODE_COMPILE;
imports_in_descriptor_set_ = false;
disallow_services_ = false;
}
bool CommandLineInterface::MakeInputsBeProtoPathRelative(
DiskSourceTree* source_tree) {
for (int i = 0; i < input_files_.size(); i++) {
string virtual_file, shadowing_disk_file;
switch (source_tree->DiskFileToVirtualFile(
input_files_[i], &virtual_file, &shadowing_disk_file)) {
case DiskSourceTree::SUCCESS:
input_files_[i] = virtual_file;
break;
case DiskSourceTree::SHADOWED:
cerr << input_files_[i] << ": Input is shadowed in the --proto_path "
"by \"" << shadowing_disk_file << "\". Either use the latter "
"file as your input or reorder the --proto_path so that the "
"former file's location comes first." << endl;
return false;
case DiskSourceTree::CANNOT_OPEN:
cerr << input_files_[i] << ": " << strerror(errno) << endl;
return false;
case DiskSourceTree::NO_MAPPING:
// First check if the file exists at all.
if (access(input_files_[i].c_str(), F_OK) < 0) {
// File does not even exist.
cerr << input_files_[i] << ": " << strerror(ENOENT) << endl;
} else {
cerr << input_files_[i] << ": File does not reside within any path "
"specified using --proto_path (or -I). You must specify a "
"--proto_path which encompasses this file. Note that the "
"proto_path must be an exact prefix of the .proto file "
"names -- protoc is too dumb to figure out when two paths "
"(e.g. absolute and relative) are equivalent (it's harder "
"than you think)." << endl;
}
return false;
}
}
return true;
}
bool CommandLineInterface::ParseArguments(int argc, const char* const argv[]) {
executable_name_ = argv[0];
// Iterate through all arguments and parse them.
for (int i = 1; i < argc; i++) {
string name, value;
if (ParseArgument(argv[i], &name, &value)) {
// Returned true => Use the next argument as the flag value.
if (i + 1 == argc || argv[i+1][0] == '-') {
cerr << "Missing value for flag: " << name << endl;
if (name == "--decode") {
cerr << "To decode an unknown message, use --decode_raw." << endl;
}
return false;
} else {
++i;
value = argv[i];
}
}
if (!InterpretArgument(name, value)) return false;
}
// If no --proto_path was given, use the current working directory.
if (proto_path_.empty()) {
proto_path_.push_back(make_pair("", "."));
}
// Check some errror cases.
bool decoding_raw = (mode_ == MODE_DECODE) && codec_type_.empty();
if (decoding_raw && !input_files_.empty()) {
cerr << "When using --decode_raw, no input files should be given." << endl;
return false;
} else if (!decoding_raw && input_files_.empty()) {
cerr << "Missing input file." << endl;
return false;
}
if (mode_ == MODE_COMPILE && output_directives_.empty() &&
descriptor_set_name_.empty()) {
cerr << "Missing output directives." << endl;
return false;
}
if (imports_in_descriptor_set_ && descriptor_set_name_.empty()) {
cerr << "--include_imports only makes sense when combined with "
"--descriptor_set_out." << endl;
}
return true;
}
bool CommandLineInterface::ParseArgument(const char* arg,
string* name, string* value) {
bool parsed_value = false;
if (arg[0] != '-') {
// Not a flag.
name->clear();
parsed_value = true;
*value = arg;
} else if (arg[1] == '-') {
// Two dashes: Multi-character name, with '=' separating name and
// value.
const char* equals_pos = strchr(arg, '=');
if (equals_pos != NULL) {
*name = string(arg, equals_pos - arg);
*value = equals_pos + 1;
parsed_value = true;
} else {
*name = arg;
}
} else {
// One dash: One-character name, all subsequent characters are the
// value.
if (arg[1] == '\0') {
// arg is just "-". We treat this as an input file, except that at
// present this will just lead to a "file not found" error.
name->clear();
*value = arg;
parsed_value = true;
} else {
*name = string(arg, 2);
*value = arg + 2;
parsed_value = !value->empty();
}
}
// Need to return true iff the next arg should be used as the value for this
// one, false otherwise.
if (parsed_value) {
// We already parsed a value for this flag.
return false;
}
if (*name == "-h" || *name == "--help" ||
*name == "--disallow_services" ||
*name == "--include_imports" ||
*name == "--version" ||
*name == "--decode_raw") {
// HACK: These are the only flags that don't take a value.
// They probably should not be hard-coded like this but for now it's
// not worth doing better.
return false;
}
// Next argument is the flag value.
return true;
}
bool CommandLineInterface::InterpretArgument(const string& name,
const string& value) {
if (name.empty()) {
// Not a flag. Just a filename.
if (value.empty()) {
cerr << "You seem to have passed an empty string as one of the "
"arguments to " << executable_name_ << ". This is actually "
"sort of hard to do. Congrats. Unfortunately it is not valid "
"input so the program is going to die now." << endl;
return false;
}
input_files_.push_back(value);
} else if (name == "-I" || name == "--proto_path") {
// Java's -classpath (and some other languages) delimits path components
// with colons. Let's accept that syntax too just to make things more
// intuitive.
vector<string> parts;
SplitStringUsing(value, kPathSeparator, &parts);
for (int i = 0; i < parts.size(); i++) {
string virtual_path;
string disk_path;
int equals_pos = parts[i].find_first_of('=');
if (equals_pos == string::npos) {
virtual_path = "";
disk_path = parts[i];
} else {
virtual_path = parts[i].substr(0, equals_pos);
disk_path = parts[i].substr(equals_pos + 1);
}
if (disk_path.empty()) {
cerr << "--proto_path passed empty directory name. (Use \".\" for "
"current directory.)" << endl;
return false;
}
// Make sure disk path exists, warn otherwise.
if (access(disk_path.c_str(), F_OK) < 0) {
cerr << disk_path << ": warning: directory does not exist." << endl;
}
proto_path_.push_back(make_pair(virtual_path, disk_path));
}
} else if (name == "-o" || name == "--descriptor_set_out") {
if (!descriptor_set_name_.empty()) {
cerr << name << " may only be passed once." << endl;
return false;
}
if (value.empty()) {
cerr << name << " requires a non-empty value." << endl;
return false;
}
if (mode_ != MODE_COMPILE) {
cerr << "Cannot use --encode or --decode and generate descriptors at the "
"same time." << endl;
return false;
}
descriptor_set_name_ = value;
} else if (name == "--include_imports") {
if (imports_in_descriptor_set_) {
cerr << name << " may only be passed once." << endl;
return false;
}
imports_in_descriptor_set_ = true;
} else if (name == "-h" || name == "--help") {
PrintHelpText();
return false; // Exit without running compiler.
} else if (name == "--version") {
if (!version_info_.empty()) {
cout << version_info_ << endl;
}
cout << "libprotoc "
<< protobuf::internal::VersionString(GOOGLE_PROTOBUF_VERSION)
<< endl;
return false; // Exit without running compiler.
} else if (name == "--disallow_services") {
disallow_services_ = true;
} else if (name == "--encode" || name == "--decode" ||
name == "--decode_raw") {
if (mode_ != MODE_COMPILE) {
cerr << "Only one of --encode and --decode can be specified." << endl;
return false;
}
if (!output_directives_.empty() || !descriptor_set_name_.empty()) {
cerr << "Cannot use " << name
<< " and generate code or descriptors at the same time." << endl;
return false;
}
mode_ = (name == "--encode") ? MODE_ENCODE : MODE_DECODE;
if (value.empty() && name != "--decode_raw") {
cerr << "Type name for " << name << " cannot be blank." << endl;
if (name == "--decode") {
cerr << "To decode an unknown message, use --decode_raw." << endl;
}
return false;
} else if (!value.empty() && name == "--decode_raw") {
cerr << "--decode_raw does not take a parameter." << endl;
return false;
}
codec_type_ = value;
} else if (name == "--error_format") {
if (value == "gcc") {
error_format_ = ERROR_FORMAT_GCC;
} else if (value == "msvs") {
error_format_ = ERROR_FORMAT_MSVS;
} else {
cerr << "Unknown error format: " << value << endl;
return false;
}
} else if (name == "--plugin") {
if (plugin_prefix_.empty()) {
cerr << "This compiler does not support plugins." << endl;
return false;
}
string name;
string path;
string::size_type equals_pos = value.find_first_of('=');
if (equals_pos == string::npos) {
// Use the basename of the file.
string::size_type slash_pos = value.find_last_of('/');
if (slash_pos == string::npos) {
name = value;
} else {
name = value.substr(slash_pos + 1);
}
path = value;
} else {
name = value.substr(0, equals_pos);
path = value.substr(equals_pos + 1);
}
plugins_[name] = path;
} else {
// Some other flag. Look it up in the generators list.
const GeneratorInfo* generator_info = FindOrNull(generators_, name);
if (generator_info == NULL &&
(plugin_prefix_.empty() || !HasSuffixString(name, "_out"))) {
cerr << "Unknown flag: " << name << endl;
return false;
}
// It's an output flag. Add it to the output directives.
if (mode_ != MODE_COMPILE) {
cerr << "Cannot use --encode or --decode and generate code at the "
"same time." << endl;
return false;
}
OutputDirective directive;
directive.name = name;
if (generator_info == NULL) {
directive.generator = NULL;
} else {
directive.generator = generator_info->generator;
}
// Split value at ':' to separate the generator parameter from the
// filename. However, avoid doing this if the colon is part of a valid
// Windows-style absolute path.
string::size_type colon_pos = value.find_first_of(':');
if (colon_pos == string::npos || IsWindowsAbsolutePath(value)) {
directive.output_location = value;
} else {
directive.parameter = value.substr(0, colon_pos);
directive.output_location = value.substr(colon_pos + 1);
}
output_directives_.push_back(directive);
}
return true;
}
void CommandLineInterface::PrintHelpText() {
// Sorry for indentation here; line wrapping would be uglier.
cerr <<
"Usage: " << executable_name_ << " [OPTION] PROTO_FILES\n"
"Parse PROTO_FILES and generate output based on the options given:\n"
" -IPATH, --proto_path=PATH Specify the directory in which to search for\n"
" imports. May be specified multiple times;\n"
" directories will be searched in order. If not\n"
" given, the current working directory is used.\n"
" --version Show version info and exit.\n"
" -h, --help Show this text and exit.\n"
" --encode=MESSAGE_TYPE Read a text-format message of the given type\n"
" from standard input and write it in binary\n"
" to standard output. The message type must\n"
" be defined in PROTO_FILES or their imports.\n"
" --decode=MESSAGE_TYPE Read a binary message of the given type from\n"
" standard input and write it in text format\n"
" to standard output. The message type must\n"
" be defined in PROTO_FILES or their imports.\n"
" --decode_raw Read an arbitrary protocol message from\n"
" standard input and write the raw tag/value\n"
" pairs in text format to standard output. No\n"
" PROTO_FILES should be given when using this\n"
" flag.\n"
" -oFILE, Writes a FileDescriptorSet (a protocol buffer,\n"
" --descriptor_set_out=FILE defined in descriptor.proto) containing all of\n"
" the input files to FILE.\n"
" --include_imports When using --descriptor_set_out, also include\n"
" all dependencies of the input files in the\n"
" set, so that the set is self-contained.\n"
" --error_format=FORMAT Set the format in which to print errors.\n"
" FORMAT may be 'gcc' (the default) or 'msvs'\n"
" (Microsoft Visual Studio format)." << endl;
if (!plugin_prefix_.empty()) {
cerr <<
" --plugin=EXECUTABLE Specifies a plugin executable to use.\n"
" Normally, protoc searches the PATH for\n"
" plugins, but you may specify additional\n"
" executables not in the path using this flag.\n"
" Additionally, EXECUTABLE may be of the form\n"
" NAME=PATH, in which case the given plugin name\n"
" is mapped to the given executable even if\n"
" the executable's own name differs." << endl;
}
for (GeneratorMap::iterator iter = generators_.begin();
iter != generators_.end(); ++iter) {
// FIXME(kenton): If the text is long enough it will wrap, which is ugly,
// but fixing this nicely (e.g. splitting on spaces) is probably more
// trouble than it's worth.
cerr << " " << iter->first << "=OUT_DIR "
<< string(19 - iter->first.size(), ' ') // Spaces for alignment.
<< iter->second.help_text << endl;
}
}
bool CommandLineInterface::GenerateOutput(
const vector<const FileDescriptor*>& parsed_files,
const OutputDirective& output_directive,
OutputDirectory* output_directory) {
// Call the generator.
string error;
if (output_directive.generator == NULL) {
// This is a plugin.
GOOGLE_CHECK(HasPrefixString(output_directive.name, "--") &&
HasSuffixString(output_directive.name, "_out"))
<< "Bad name for plugin generator: " << output_directive.name;
// Strip the "--" and "_out" and add the plugin prefix.
string plugin_name = plugin_prefix_ + "gen-" +
output_directive.name.substr(2, output_directive.name.size() - 6);
if (!GeneratePluginOutput(parsed_files, plugin_name,
output_directive.parameter,
output_directory, &error)) {
cerr << output_directive.name << ": " << error << endl;
return false;
}
} else {
// Regular generator.
for (int i = 0; i < parsed_files.size(); i++) {
if (!output_directive.generator->Generate(
parsed_files[i], output_directive.parameter,
output_directory, &error)) {
// Generator returned an error.
cerr << output_directive.name << ": " << parsed_files[i]->name() << ": "
<< error << endl;
return false;
}
}
}
return true;
}
bool CommandLineInterface::GeneratePluginOutput(
const vector<const FileDescriptor*>& parsed_files,
const string& plugin_name,
const string& parameter,
OutputDirectory* output_directory,
string* error) {
CodeGeneratorRequest request;
CodeGeneratorResponse response;
// Build the request.
if (!parameter.empty()) {
request.set_parameter(parameter);
}
set<const FileDescriptor*> already_seen;
for (int i = 0; i < parsed_files.size(); i++) {
request.add_file_to_generate(parsed_files[i]->name());
GetTransitiveDependencies(parsed_files[i], &already_seen,
request.mutable_proto_file());
}
// Invoke the plugin.
Subprocess subprocess;
if (plugins_.count(plugin_name) > 0) {
subprocess.Start(plugins_[plugin_name], Subprocess::EXACT_NAME);
} else {
subprocess.Start(plugin_name, Subprocess::SEARCH_PATH);
}
string communicate_error;
if (!subprocess.Communicate(request, &response, &communicate_error)) {
*error = strings::Substitute("$0: $1", plugin_name, communicate_error);
return false;
}
// Write the files. We do this even if there was a generator error in order
// to match the behavior of a compiled-in generator.
scoped_ptr<io::ZeroCopyOutputStream> current_output;
for (int i = 0; i < response.file_size(); i++) {
const CodeGeneratorResponse::File& output_file = response.file(i);
if (!output_file.insertion_point().empty()) {
// Open a file for insert.
// We reset current_output to NULL first so that the old file is closed
// before the new one is opened.
current_output.reset();
current_output.reset(output_directory->OpenForInsert(
output_file.name(), output_file.insertion_point()));
} else if (!output_file.name().empty()) {
// Starting a new file. Open it.
// We reset current_output to NULL first so that the old file is closed
// before the new one is opened.
current_output.reset();
current_output.reset(output_directory->Open(output_file.name()));
} else if (current_output == NULL) {
*error = strings::Substitute(
"$0: First file chunk returned by plugin did not specify a file name.",
plugin_name);
return false;
}
// Use CodedOutputStream for convenience; otherwise we'd need to provide
// our own buffer-copying loop.
io::CodedOutputStream writer(current_output.get());
writer.WriteString(output_file.content());
}
// Check for errors.
if (!response.error().empty()) {
// Generator returned an error.
*error = response.error();
return false;
}
return true;
}
bool CommandLineInterface::EncodeOrDecode(const DescriptorPool* pool) {
// Look up the type.
const Descriptor* type = pool->FindMessageTypeByName(codec_type_);
if (type == NULL) {
cerr << "Type not defined: " << codec_type_ << endl;
return false;
}
DynamicMessageFactory dynamic_factory(pool);
scoped_ptr<Message> message(dynamic_factory.GetPrototype(type)->New());
if (mode_ == MODE_ENCODE) {
SetFdToTextMode(STDIN_FILENO);
SetFdToBinaryMode(STDOUT_FILENO);
} else {
SetFdToBinaryMode(STDIN_FILENO);
SetFdToTextMode(STDOUT_FILENO);
}
io::FileInputStream in(STDIN_FILENO);
io::FileOutputStream out(STDOUT_FILENO);
if (mode_ == MODE_ENCODE) {
// Input is text.
ErrorPrinter error_collector(error_format_);
TextFormat::Parser parser;
parser.RecordErrorsTo(&error_collector);
parser.AllowPartialMessage(true);
if (!parser.Parse(&in, message.get())) {
cerr << "Failed to parse input." << endl;
return false;
}
} else {
// Input is binary.
if (!message->ParsePartialFromZeroCopyStream(&in)) {
cerr << "Failed to parse input." << endl;
return false;
}
}
if (!message->IsInitialized()) {
cerr << "warning: Input message is missing required fields: "
<< message->InitializationErrorString() << endl;
}
if (mode_ == MODE_ENCODE) {
// Output is binary.
if (!message->SerializePartialToZeroCopyStream(&out)) {
cerr << "output: I/O error." << endl;
return false;
}
} else {
// Output is text.
if (!TextFormat::Print(*message, &out)) {
cerr << "output: I/O error." << endl;
return false;
}
}
return true;
}
bool CommandLineInterface::WriteDescriptorSet(
const vector<const FileDescriptor*> parsed_files) {
FileDescriptorSet file_set;
if (imports_in_descriptor_set_) {
set<const FileDescriptor*> already_seen;
for (int i = 0; i < parsed_files.size(); i++) {
GetTransitiveDependencies(
parsed_files[i], &already_seen, file_set.mutable_file());
}
} else {
for (int i = 0; i < parsed_files.size(); i++) {
parsed_files[i]->CopyTo(file_set.add_file());
}
}
int fd;
do {
fd = open(descriptor_set_name_.c_str(),
O_WRONLY | O_CREAT | O_TRUNC | O_BINARY, 0666);
} while (fd < 0 && errno == EINTR);
if (fd < 0) {
perror(descriptor_set_name_.c_str());
return false;
}
io::FileOutputStream out(fd);
if (!file_set.SerializeToZeroCopyStream(&out)) {
cerr << descriptor_set_name_ << ": " << strerror(out.GetErrno()) << endl;
out.Close();
return false;
}
if (!out.Close()) {
cerr << descriptor_set_name_ << ": " << strerror(out.GetErrno()) << endl;
return false;
}
return true;
}
void CommandLineInterface::GetTransitiveDependencies(
const FileDescriptor* file,
set<const FileDescriptor*>* already_seen,
RepeatedPtrField<FileDescriptorProto>* output) {
if (!already_seen->insert(file).second) {
// Already saw this file. Skip.
return;
}
// Add all dependencies.
for (int i = 0; i < file->dependency_count(); i++) {
GetTransitiveDependencies(file->dependency(i), already_seen, output);
}
// Add this file.
file->CopyTo(output->Add());
}
} // namespace compiler
} // namespace protobuf
} // namespace google
|