Source

setuptools / setuptools / package_index.py

Full commit
PJ Eby 32f6551 
Vinay Sajip 66e9505 
Jason R. Coombs 2b7a3ac 
Jason R. Coombs 938b84c 
Jason R. Coombs 970ad90 
PJ Eby 32f6551 
PJ Eby f65be02 
PJ Eby ab401fd 
Vinay Sajip 66e9505 



Jason R. Coombs ce3db66 
PJ Eby 22ee3b4 



PJ Eby 99ca944 
Jason R. Coombs 2b7a3ac 
Jason R. Coombs 635f2c3 
Jason R. Coombs 5d53c90 
PJ Eby e1315c5 
PJ Eby ba233d2 
PJ Eby 5608177 

PJ Eby 503ef29 

PJ Eby 5608177 
PJ Eby 32f6551 



PJ Eby 7194270 

PJ Eby 32f6551 

tarek bafde0d 

PJ Eby 7194270 

PJ Eby 32f6551 
PJ Eby 7194270 
agroszer 0808246 
PJ Eby 32f6551 
PJ Eby 7194270 


agroszer 0808246 
PJ Eby 2dcc98a 
PJ Eby 33eac32 
PJ Eby 7194270 
agroszer 0808246 








PJ Eby 32f6551 

PJ Eby 0f1ec85 
Vinay Sajip 66e9505 

PJ Eby d24edbe 
Jason R. Coombs 333193d 
PJ Eby f4ff636 
PJ Eby 0f1ec85 




PJ Eby ef8708e 

PJ Eby ba233d2 

PJ Eby ef8708e 
PJ Eby ba233d2 
PJ Eby ef8708e 

PJ Eby 6c23340 
PJ Eby ba233d2 
PJ Eby cd4efd5 


PJ Eby ef8708e 

PJ Eby ba233d2 
PJ Eby cd4efd5 
agroszer 0808246 
PJ Eby 7194270 

agroszer 0808246 
PJ Eby 7194270 

PJ Eby 6c23340 
PJ Eby 32f6551 
PJ Eby cd4efd5 

PJ Eby ba233d2 
PJ Eby 7194270 

PJ Eby ba233d2 




PJ Eby 7194270 

PJ Eby ba233d2 



PJ Eby 7194270 
PJ Eby ba233d2 



PJ Eby 32f6551 











PJ Eby cd4efd5 
PJ Eby 23e42f6 




PJ Eby 32f6551 

PJ Eby ba233d2 

PJ Eby 7194270 
PJ Eby 32f6551 

Jason R. Coombs f9c8b1d 







Jason R. Coombs ce3db66 
Jason R. Coombs f9c8b1d 








Jason R. Coombs 5d53c90 









PJ Eby aa3d0c0 


Jason R. Coombs 5d53c90 
PJ Eby aa3d0c0 




Vinay Sajip 066b3f3 
PJ Eby aa3d0c0 

Vinay Sajip 66e9505 
PJ Eby aa3d0c0 





Vinay Sajip 66e9505 
Dirley Rodrigues d5451f3 
PJ Eby 09896de 
Jason R. Coombs 3180d86 
PJ Eby 8c12806 
PJ Eby aa3d0c0 
PJ Eby 71cbc4c 
PJ Eby 32f6551 

pje 185d7f0 
pje 274bbe3 
PJ Eby de07aa0 
PJ Eby 71cbc4c 
PJ Eby 32f6551 



PJ Eby 99ca944 
PJ Eby 6b544b6 
pje 274bbe3 


PJ Eby 32f6551 

PJ Eby cd4efd5 
PJ Eby 32f6551 


PJ Eby cd4efd5 
PJ Eby 6b544b6 

PJ Eby cd4efd5 

PJ Eby 99ca944 



PJ Eby cd4efd5 
PJ Eby 6c23340 
Vinay Sajip 66e9505 
PJ Eby 99ca944 


PJ Eby c5eda1f 
PJ Eby 32f6551 

PJ Eby 700ab10 
PJ Eby d24edbe 
Jason R. Coombs bac790a 
PJ Eby 79eac6a 
PJ Eby d24edbe 
PJ Eby a6f7800 
PJ Eby 32f6551 




Lennart Regebro d4f21a0 
Vinay Sajip 66e9505 
Lennart Regebro d4f21a0 



Martin von Löwis 34b5ca6 
PJ Eby 32f6551 
PJ Eby 2599f21 
Vinay Sajip 66e9505 
PJ Eby 2599f21 
Christophe Combe… d04bea7 

PJ Eby 32f6551 
PJ Eby 6b544b6 


PJ Eby 105ac4f 
PJ Eby 6b544b6 

PJ Eby 3a3cf34 
PJ Eby 6b544b6 






Vinay Sajip 66e9505 
PJ Eby cd4efd5 
PJ Eby 99ca944 
PJ Eby fdda506 
Vinay Sajip 66e9505 
PJ Eby 99ca944 





PJ Eby cd4efd5 
PJ Eby 16718b7 





PJ Eby cd4efd5 
PJ Eby 16718b7 
Vinay Sajip 66e9505 

PJ Eby 16718b7 




PJ Eby de07aa0 
PJ Eby 32f6551 
PJ Eby 700ab10 
PJ Eby 32f6551 
PJ Eby 700ab10 
PJ Eby 32f6551 
Vinay Sajip 66e9505 


PJ Eby aa3d0c0 
PJ Eby 32f6551 



PJ Eby 0f1ec85 

PJ Eby 700ab10 
PJ Eby aa3d0c0 

Tarek Ziadé 7d06383 
Vinay Sajip 66e9505 
Tarek Ziadé 7d06383 

PJ Eby aa3d0c0 
















PJ Eby 32f6551 
PJ Eby aa3d0c0 


PJ Eby 700ab10 
PJ Eby 0f1ec85 




PJ Eby 700ab10 
PJ Eby d028455 
PJ Eby 0f1ec85 

PJ Eby 37aded6 
PJ Eby 0f1ec85 


PJ Eby 700ab10 
PJ Eby 27083ee 
PJ Eby c4481bd 
PJ Eby 105ac4f 
PJ Eby c4481bd 


PJ Eby 6b544b6 
PJ Eby 700ab10 

PJ Eby b32b399 
PJ Eby 700ab10 
PJ Eby aa3d0c0 
PJ Eby 700ab10 


PJ Eby 27083ee 
PJ Eby 6b544b6 
PJ Eby d38e94e 
PJ Eby 32f6551 

PJ Eby 6c23340 
PJ Eby 27083ee 
PJ Eby 700ab10 
PJ Eby b32b399 



PJ Eby 5608177 


Vinay Sajip 66e9505 
PJ Eby 5608177 





PJ Eby 700ab10 
PJ Eby 6b544b6 













PJ Eby f93703a 
PJ Eby 6b544b6 


Vinay Sajip 66e9505 
PJ Eby 6b544b6 
PJ Eby f93703a 
PJ Eby b32b399 







PJ Eby f93703a 
PJ Eby 32f6551 
PJ Eby 27083ee 
PJ Eby 32f6551 

PJ Eby 27083ee 
PJ Eby 0f1ec85 



PJ Eby 32f6551 
PJ Eby 27083ee 
PJ Eby 4ec3166 





PJ Eby 32f6551 




PJ Eby 0f1ec85 




PJ Eby 32f6551 






PJ Eby ab401fd 
PJ Eby 32f6551 


PJ Eby 4ec3166 
PJ Eby 27083ee 

PJ Eby 4ec3166 
David Cournapeau d7edb68 

PJ Eby 4ec3166 

PJ Eby 27083ee 




PJ Eby 4ec3166 



PJ Eby 86ac3e2 

PJ Eby 4ec3166 


PJ Eby 27083ee 
PJ Eby 32f6551 
PJ Eby 27083ee 
PJ Eby 4ec3166 
David Cournapeau d7edb68 
PJ Eby 27083ee 
David Cournapeau d7edb68 


PJ Eby 4ec3166 

David Cournapeau d7edb68 
PJ Eby 4ec3166 






PJ Eby 86ac3e2 
PJ Eby d24edbe 


PJ Eby 5608177 
PJ Eby 27083ee 
PJ Eby 6b544b6 

David Cournapeau d7edb68 
PJ Eby 6b544b6 
David Cournapeau d7edb68 


PJ Eby 6b544b6 




PJ Eby 27083ee 
PJ Eby 86ac3e2 
PJ Eby 27083ee 
PJ Eby 86ac3e2 





PJ Eby 8953930 


PJ Eby d24edbe 
PJ Eby 27083ee 
PJ Eby 4ec3166 







PJ Eby e2a81bc 
PJ Eby 4ec3166 




PJ Eby 0f1ec85 
PJ Eby 105ac4f 
PJ Eby 0f1ec85 





































PJ Eby 105ac4f 
PJ Eby 32f6551 

PJ Eby 700ab10 
PJ Eby 32f6551 
PJ Eby 5608177 
PJ Eby 32f6551 
PJ Eby 5608177 

PJ Eby 32f6551 
Vinay Sajip 66e9505 
PJ Eby ab401fd 
PJ Eby 32f6551 

PJ Eby 5608177 
PJ Eby 32f6551 




Tarek Ziadé 4a17a74 
Jason R. Coombs 635f2c3 

PJ Eby 32f6551 




PJ Eby 5608177 
PJ Eby 32f6551 




PJ Eby 5608177 
PJ Eby 32f6551 







PJ Eby ffc6fb3 
PJ Eby 79eac6a 
PJ Eby f5627d0 

PJ Eby 32f6551 
pje 274bbe3 
Vinay Sajip 66e9505 

tarek 347c767 




Vinay Sajip 66e9505 

PJ Eby 32f6551 
Vinay Sajip 66e9505 

tarek 347c767 

Jim Fulton 74f940a 


Vinay Sajip 66e9505 

tarek 347c767 





Vinay Sajip 66e9505 

Hanno Schlichtin… 202458e 




PJ Eby 32f6551 



PJ Eby 056ddad 
PJ Eby 32f6551 





PJ Eby 6c23340 


PJ Eby 32f6551 





Paulo Köch ef917ac 

Paulo Köch a13066a 

PJ Eby 105ac4f 
Vinay Sajip 6015632 
PJ Eby 32f6551 
PJ Eby 3e13956 

PJ Eby 32f6551 
PJ Eby 700ab10 

PJ Eby 32f6551 
PJ Eby 9c4a8d9 
PJ Eby 3e13956 

PJ Eby 42aa009 
PJ Eby 3e13956 



PJ Eby ffc6fb3 
PJ Eby 32f6551 



PJ Eby 09896de 
PJ Eby 32f6551 





PJ Eby a4c73c5 
PJ Eby ab401fd 
PJ Eby 32f6551 
PJ Eby 66f6a26 














PJ Eby 32f6551 
PJ Eby ba233d2 
PJ Eby 66f6a26 

Vinay Sajip 6015632 
PJ Eby 66f6a26 

Vinay Sajip 066b3f3 
PJ Eby 66f6a26 






Vinay Sajip 6015632 
PJ Eby 700ab10 
PJ Eby 66f6a26 
PJ Eby 32f6551 

Paulo Köch bb0cc7d 
Vinay Sajip 6015632 
Paulo Köch bb0cc7d 










Vinay Sajip 6015632 
Paulo Köch bb0cc7d 


Paulo Köch ef917ac 
Paulo Köch bb0cc7d 


Paulo Köch ef917ac 
Paulo Köch bb0cc7d 








Paulo Köch ef917ac 

Paulo Köch a13066a 
Paulo Köch bb0cc7d 


Paulo Köch a13066a 

Paulo Köch bb0cc7d 







Paulo Köch a13066a 

PJ Eby 700ab10 
PJ Eby f65be02 
PJ Eby 32f6551 
PJ Eby 700ab10 
PJ Eby f65be02 
PJ Eby 6c23340 
PJ Eby 700ab10 
PJ Eby f65be02 
PJ Eby 32f6551 
PJ Eby 66f6a26 











PJ Eby da2a851 




































PJ Eby 3e13956 
tarek bafde0d 










PJ Eby 3e13956 
Jason R. Coombs 938b84c 






Jason R. Coombs c0510ee 
Jason R. Coombs 938b84c 







PJ Eby 3e13956 
pje 274bbe3 
PJ Eby 9c4a8d9 

Vinay Sajip 66e9505 
PJ Eby 9c4a8d9 
Stefan H. Holek 386239a 

Stefan H. Holek 4825beb 

Stefan H. Holek 386239a 
PJ Eby 9c4a8d9 
Vinay Sajip 66e9505 
PJ Eby 9c4a8d9 



Jason R. Coombs 938b84c 
Vinay Sajip 66e9505 
PJ Eby 9c4a8d9 





pje 274bbe3 
PJ Eby 9c4a8d9 



Vinay Sajip 66e9505 
PJ Eby 9c4a8d9 
Vinay Sajip 66e9505 
PJ Eby 9c4a8d9 


tarek bafde0d 

PJ Eby 9c4a8d9 










PJ Eby f93703a 
PJ Eby 3e13956 
PJ Eby a4c73c5 
PJ Eby f5627d0 

Vinay Sajip 66e9505 

PJ Eby f5627d0 





Hanno Schlichtin… 02cdc04 


PJ Eby f5627d0 









PJ Eby 32f6551 
Vinay Sajip 66e9505 

PJ Eby f93703a 


PJ Eby de07aa0 









Stefan H. Holek 386239a 
  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
"""PyPI and direct package downloading"""
import sys, os.path, re, shutil, random, socket
import itertools
import base64
from setuptools import ssl_support
from pkg_resources import *
from distutils import log
from distutils.errors import DistutilsError
from setuptools.compat import (urllib2, httplib, StringIO, HTTPError,
                               urlparse, urlunparse, unquote, splituser,
                               url2pathname, name2codepoint,
                               unichr, urljoin)
from setuptools.compat import filterfalse
try:
    from hashlib import md5
except ImportError:
    from md5 import md5
from fnmatch import translate
from setuptools.py24compat import wraps
from setuptools.py27compat import get_all_headers

EGG_FRAGMENT = re.compile(r'^egg=([-A-Za-z0-9_.]+)$')
HREF = re.compile("""href\\s*=\\s*['"]?([^'"> ]+)""", re.I)
# this is here to fix emacs' cruddy broken syntax highlighting
PYPI_MD5 = re.compile(
    '<a href="([^"#]+)">([^<]+)</a>\n\s+\\(<a (?:title="MD5 hash"\n\s+)'
    'href="[^?]+\?:action=show_md5&amp;digest=([0-9a-f]{32})">md5</a>\\)'
)
URL_SCHEME = re.compile('([-+.a-z0-9]{2,}):',re.I).match
EXTENSIONS = ".tar.gz .tar.bz2 .tar .zip .tgz".split()

__all__ = [
    'PackageIndex', 'distros_for_url', 'parse_bdist_wininst',
    'interpret_distro_name',
]

_SOCKET_TIMEOUT = 15

def parse_bdist_wininst(name):
    """Return (base,pyversion) or (None,None) for possible .exe name"""

    lower = name.lower()
    base, py_ver, plat = None, None, None

    if lower.endswith('.exe'):
        if lower.endswith('.win32.exe'):
            base = name[:-10]
            plat = 'win32'
        elif lower.startswith('.win32-py',-16):
            py_ver = name[-7:-4]
            base = name[:-16]
            plat = 'win32'
        elif lower.endswith('.win-amd64.exe'):
            base = name[:-14]
            plat = 'win-amd64'
        elif lower.startswith('.win-amd64-py',-20):
            py_ver = name[-7:-4]
            base = name[:-20]
            plat = 'win-amd64'
    return base,py_ver,plat


def egg_info_for_url(url):
    scheme, server, path, parameters, query, fragment = urlparse(url)
    base = unquote(path.split('/')[-1])
    if server=='sourceforge.net' and base=='download':    # XXX Yuck
        base = unquote(path.split('/')[-2])
    if '#' in base: base, fragment = base.split('#',1)
    return base,fragment

def distros_for_url(url, metadata=None):
    """Yield egg or source distribution objects that might be found at a URL"""
    base, fragment = egg_info_for_url(url)
    for dist in distros_for_location(url, base, metadata): yield dist
    if fragment:
        match = EGG_FRAGMENT.match(fragment)
        if match:
            for dist in interpret_distro_name(
                url, match.group(1), metadata, precedence = CHECKOUT_DIST
            ):
                yield dist

def distros_for_location(location, basename, metadata=None):
    """Yield egg or source distribution objects based on basename"""
    if basename.endswith('.egg.zip'):
        basename = basename[:-4]    # strip the .zip
    if basename.endswith('.egg') and '-' in basename:
        # only one, unambiguous interpretation
        return [Distribution.from_location(location, basename, metadata)]
    if basename.endswith('.exe'):
        win_base, py_ver, platform = parse_bdist_wininst(basename)
        if win_base is not None:
            return interpret_distro_name(
                location, win_base, metadata, py_ver, BINARY_DIST, platform
            )
    # Try source distro extensions (.zip, .tgz, etc.)
    #
    for ext in EXTENSIONS:
        if basename.endswith(ext):
            basename = basename[:-len(ext)]
            return interpret_distro_name(location, basename, metadata)
    return []  # no extension matched

def distros_for_filename(filename, metadata=None):
    """Yield possible egg or source distribution objects based on a filename"""
    return distros_for_location(
        normalize_path(filename), os.path.basename(filename), metadata
    )


def interpret_distro_name(location, basename, metadata,
    py_version=None, precedence=SOURCE_DIST, platform=None
):
    """Generate alternative interpretations of a source distro name

    Note: if `location` is a filesystem filename, you should call
    ``pkg_resources.normalize_path()`` on it before passing it to this
    routine!
    """
    # Generate alternative interpretations of a source distro name
    # Because some packages are ambiguous as to name/versions split
    # e.g. "adns-python-1.1.0", "egenix-mx-commercial", etc.
    # So, we generate each possible interepretation (e.g. "adns, python-1.1.0"
    # "adns-python, 1.1.0", and "adns-python-1.1.0, no version").  In practice,
    # the spurious interpretations should be ignored, because in the event
    # there's also an "adns" package, the spurious "python-1.1.0" version will
    # compare lower than any numeric version number, and is therefore unlikely
    # to match a request for it.  It's still a potential problem, though, and
    # in the long run PyPI and the distutils should go for "safe" names and
    # versions in distribution archive names (sdist and bdist).

    parts = basename.split('-')
    if not py_version:
        for i,p in enumerate(parts[2:]):
            if len(p)==5 and p.startswith('py2.'):
                return # It's a bdist_dumb, not an sdist -- bail out

    for p in range(1,len(parts)+1):
        yield Distribution(
            location, metadata, '-'.join(parts[:p]), '-'.join(parts[p:]),
            py_version=py_version, precedence = precedence,
            platform = platform
        )

# From Python 2.7 docs
def unique_everseen(iterable, key=None):
    "List unique elements, preserving order. Remember all elements ever seen."
    # unique_everseen('AAAABBBCCDAABBB') --> A B C D
    # unique_everseen('ABBCcAD', str.lower) --> A B C D
    seen = set()
    seen_add = seen.add
    if key is None:
        for element in filterfalse(seen.__contains__, iterable):
            seen_add(element)
            yield element
    else:
        for element in iterable:
            k = key(element)
            if k not in seen:
                seen_add(k)
                yield element

def unique_values(func):
    """
    Wrap a function returning an iterable such that the resulting iterable
    only ever yields unique items.
    """
    @wraps(func)
    def wrapper(*args, **kwargs):
        return unique_everseen(func(*args, **kwargs))
    return wrapper

REL = re.compile("""<([^>]*\srel\s*=\s*['"]?([^'">]+)[^>]*)>""", re.I)
# this line is here to fix emacs' cruddy broken syntax highlighting

@unique_values
def find_external_links(url, page):
    """Find rel="homepage" and rel="download" links in `page`, yielding URLs"""

    for match in REL.finditer(page):
        tag, rel = match.groups()
        rels = set(map(str.strip, rel.lower().split(',')))
        if 'homepage' in rels or 'download' in rels:
            for match in HREF.finditer(tag):
                yield urljoin(url, htmldecode(match.group(1)))

    for tag in ("<th>Home Page", "<th>Download URL"):
        pos = page.find(tag)
        if pos!=-1:
            match = HREF.search(page,pos)
            if match:
                yield urljoin(url, htmldecode(match.group(1)))

user_agent = "Python-urllib/%s setuptools/%s" % (
    sys.version[:3], require('setuptools')[0].version
)

class PackageIndex(Environment):
    """A distribution index that scans web pages for download URLs"""

    def __init__(self, index_url="https://pypi.python.org/simple", hosts=('*',),
        ca_bundle=None, verify_ssl=True, *args, **kw
    ):
        Environment.__init__(self,*args,**kw)
        self.index_url = index_url + "/"[:not index_url.endswith('/')]
        self.scanned_urls = {}
        self.fetched_urls = {}
        self.package_pages = {}
        self.allows = re.compile('|'.join(map(translate,hosts))).match
        self.to_scan = []
        if verify_ssl and ssl_support.is_available and (ca_bundle or ssl_support.find_ca_bundle()):
            self.opener = ssl_support.opener_for(ca_bundle)
        else: self.opener = urllib2.urlopen

    def process_url(self, url, retrieve=False):
        """Evaluate a URL as a possible download, and maybe retrieve it"""
        if url in self.scanned_urls and not retrieve:
            return
        self.scanned_urls[url] = True
        if not URL_SCHEME(url):
            self.process_filename(url)
            return
        else:
            dists = list(distros_for_url(url))
            if dists:
                if not self.url_ok(url):
                    return
                self.debug("Found link: %s", url)

        if dists or not retrieve or url in self.fetched_urls:
            list(map(self.add, dists))
            return  # don't need the actual page

        if not self.url_ok(url):
            self.fetched_urls[url] = True
            return

        self.info("Reading %s", url)
        self.fetched_urls[url] = True   # prevent multiple fetch attempts
        f = self.open_url(url, "Download error on %s: %%s -- Some packages may not be found!" % url)
        if f is None: return
        self.fetched_urls[f.url] = True
        if 'html' not in f.headers.get('content-type', '').lower():
            f.close()   # not html, we can't process it
            return

        base = f.url     # handle redirects
        page = f.read()
        if not isinstance(page, str): # We are in Python 3 and got bytes. We want str.
            if isinstance(f, HTTPError):
                # Errors have no charset, assume latin1:
                charset = 'latin-1'
            else:
                charset = f.headers.get_param('charset') or 'latin-1'
            page = page.decode(charset, "ignore")
        f.close()
        for match in HREF.finditer(page):
            link = urljoin(base, htmldecode(match.group(1)))
            self.process_url(link)
        if url.startswith(self.index_url) and getattr(f,'code',None)!=404:
            page = self.process_index(url, page)

    def process_filename(self, fn, nested=False):
        # process filenames or directories
        if not os.path.exists(fn):
            self.warn("Not found: %s", fn)
            return

        if os.path.isdir(fn) and not nested:
            path = os.path.realpath(fn)
            for item in os.listdir(path):
                self.process_filename(os.path.join(path,item), True)

        dists = distros_for_filename(fn)
        if dists:
            self.debug("Found: %s", fn)
            list(map(self.add, dists))

    def url_ok(self, url, fatal=False):
        s = URL_SCHEME(url)
        if (s and s.group(1).lower()=='file') or self.allows(urlparse(url)[1]):
            return True
        msg = "\nLink to % s ***BLOCKED*** by --allow-hosts\n"
        if fatal:
            raise DistutilsError(msg % url)
        else:
            self.warn(msg, url)

    def scan_egg_links(self, search_path):
        for item in search_path:
            if os.path.isdir(item):
                for entry in os.listdir(item):
                    if entry.endswith('.egg-link'):
                        self.scan_egg_link(item, entry)

    def scan_egg_link(self, path, entry):
        lines = [_f for _f in map(str.strip,
                                  open(os.path.join(path, entry))) if _f]
        if len(lines)==2:
            for dist in find_distributions(os.path.join(path, lines[0])):
                dist.location = os.path.join(path, *lines)
                dist.precedence = SOURCE_DIST
                self.add(dist)

    def process_index(self,url,page):
        """Process the contents of a PyPI page"""
        def scan(link):
            # Process a URL to see if it's for a package page
            if link.startswith(self.index_url):
                parts = list(map(
                    unquote, link[len(self.index_url):].split('/')
                ))
                if len(parts)==2 and '#' not in parts[1]:
                    # it's a package page, sanitize and index it
                    pkg = safe_name(parts[0])
                    ver = safe_version(parts[1])
                    self.package_pages.setdefault(pkg.lower(),{})[link] = True
                    return to_filename(pkg), to_filename(ver)
            return None, None

        # process an index page into the package-page index
        for match in HREF.finditer(page):
            try:
                scan( urljoin(url, htmldecode(match.group(1))) )
            except ValueError:
                pass

        pkg, ver = scan(url)   # ensure this page is in the page index
        if pkg:
            # process individual package page
            for new_url in find_external_links(url, page):
                # Process the found URL
                base, frag = egg_info_for_url(new_url)
                if base.endswith('.py') and not frag:
                    if ver:
                        new_url+='#egg=%s-%s' % (pkg,ver)
                    else:
                        self.need_version_info(url)
                self.scan_url(new_url)

            return PYPI_MD5.sub(
                lambda m: '<a href="%s#md5=%s">%s</a>' % m.group(1,3,2), page
            )
        else:
            return ""   # no sense double-scanning non-package pages



    def need_version_info(self, url):
        self.scan_all(
            "Page at %s links to .py file(s) without version info; an index "
            "scan is required.", url
        )

    def scan_all(self, msg=None, *args):
        if self.index_url not in self.fetched_urls:
            if msg: self.warn(msg,*args)
            self.info(
                "Scanning index of all packages (this may take a while)"
            )
        self.scan_url(self.index_url)

    def find_packages(self, requirement):
        self.scan_url(self.index_url + requirement.unsafe_name+'/')

        if not self.package_pages.get(requirement.key):
            # Fall back to safe version of the name
            self.scan_url(self.index_url + requirement.project_name+'/')

        if not self.package_pages.get(requirement.key):
            # We couldn't find the target package, so search the index page too
            self.not_found_in_index(requirement)

        for url in list(self.package_pages.get(requirement.key,())):
            # scan each page that might be related to the desired package
            self.scan_url(url)

    def obtain(self, requirement, installer=None):
        self.prescan(); self.find_packages(requirement)
        for dist in self[requirement.key]:
            if dist in requirement:
                return dist
            self.debug("%s does not match %s", requirement, dist)
        return super(PackageIndex, self).obtain(requirement,installer)





    def check_md5(self, cs, info, filename, tfp):
        if re.match('md5=[0-9a-f]{32}$', info):
            self.debug("Validating md5 checksum for %s", filename)
            if cs.hexdigest() != info[4:]:
                tfp.close()
                os.unlink(filename)
                raise DistutilsError(
                    "MD5 validation failed for "+os.path.basename(filename)+
                    "; possible download problem?"
                )

    def add_find_links(self, urls):
        """Add `urls` to the list that will be prescanned for searches"""
        for url in urls:
            if (
                self.to_scan is None        # if we have already "gone online"
                or not URL_SCHEME(url)      # or it's a local file/directory
                or url.startswith('file:')
                or list(distros_for_url(url))   # or a direct package link
            ):
                # then go ahead and process it now
                self.scan_url(url)
            else:
                # otherwise, defer retrieval till later
                self.to_scan.append(url)

    def prescan(self):
        """Scan urls scheduled for prescanning (e.g. --find-links)"""
        if self.to_scan:
            list(map(self.scan_url, self.to_scan))
        self.to_scan = None     # from now on, go ahead and process immediately

    def not_found_in_index(self, requirement):
        if self[requirement.key]:   # we've seen at least one distro
            meth, msg = self.info, "Couldn't retrieve index page for %r"
        else:   # no distros seen for this name, might be misspelled
            meth, msg = (self.warn,
                "Couldn't find index page for %r (maybe misspelled?)")
        meth(msg, requirement.unsafe_name)
        self.scan_all()

    def download(self, spec, tmpdir):
        """Locate and/or download `spec` to `tmpdir`, returning a local path

        `spec` may be a ``Requirement`` object, or a string containing a URL,
        an existing local filename, or a project/version requirement spec
        (i.e. the string form of a ``Requirement`` object).  If it is the URL
        of a .py file with an unambiguous ``#egg=name-version`` tag (i.e., one
        that escapes ``-`` as ``_`` throughout), a trivial ``setup.py`` is
        automatically created alongside the downloaded file.

        If `spec` is a ``Requirement`` object or a string containing a
        project/version requirement spec, this method returns the location of
        a matching distribution (possibly after downloading it to `tmpdir`).
        If `spec` is a locally existing file or directory name, it is simply
        returned unchanged.  If `spec` is a URL, it is downloaded to a subpath
        of `tmpdir`, and the local filename is returned.  Various errors may be
        raised if a problem occurs during downloading.
        """
        if not isinstance(spec,Requirement):
            scheme = URL_SCHEME(spec)
            if scheme:
                # It's a url, download it to tmpdir
                found = self._download_url(scheme.group(1), spec, tmpdir)
                base, fragment = egg_info_for_url(spec)
                if base.endswith('.py'):
                    found = self.gen_setup(found,fragment,tmpdir)
                return found
            elif os.path.exists(spec):
                # Existing file or directory, just return it
                return spec
            else:
                try:
                    spec = Requirement.parse(spec)
                except ValueError:
                    raise DistutilsError(
                        "Not a URL, existing file, or requirement spec: %r" %
                        (spec,)
                    )
        return getattr(self.fetch_distribution(spec, tmpdir),'location',None)


    def fetch_distribution(self,
        requirement, tmpdir, force_scan=False, source=False, develop_ok=False,
        local_index=None
    ):
        """Obtain a distribution suitable for fulfilling `requirement`

        `requirement` must be a ``pkg_resources.Requirement`` instance.
        If necessary, or if the `force_scan` flag is set, the requirement is
        searched for in the (online) package index as well as the locally
        installed packages.  If a distribution matching `requirement` is found,
        the returned distribution's ``location`` is the value you would have
        gotten from calling the ``download()`` method with the matching
        distribution's URL or filename.  If no matching distribution is found,
        ``None`` is returned.

        If the `source` flag is set, only source distributions and source
        checkout links will be considered.  Unless the `develop_ok` flag is
        set, development and system eggs (i.e., those using the ``.egg-info``
        format) will be ignored.
        """
        # process a Requirement
        self.info("Searching for %s", requirement)
        skipped = {}
        dist = None

        def find(req, env=None):
            if env is None:
                env = self
            # Find a matching distribution; may be called more than once

            for dist in env[req.key]:

                if dist.precedence==DEVELOP_DIST and not develop_ok:
                    if dist not in skipped:
                        self.warn("Skipping development or system egg: %s",dist)
                        skipped[dist] = 1
                    continue

                if dist in req and (dist.precedence<=SOURCE_DIST or not source):
                    return dist



        if force_scan:
            self.prescan()
            self.find_packages(requirement)
            dist = find(requirement)

        if local_index is not None:
            dist = dist or find(requirement, local_index)

        if dist is None and self.to_scan is not None:
            self.prescan()
            dist = find(requirement)

        if dist is None and not force_scan:
            self.find_packages(requirement)
            dist = find(requirement)

        if dist is None:
            self.warn(
                "No local packages or download links found for %s%s",
                (source and "a source distribution of " or ""),
                requirement,
            )
        else:
            self.info("Best match: %s", dist)
            return dist.clone(location=self.download(dist.location, tmpdir))


    def fetch(self, requirement, tmpdir, force_scan=False, source=False):
        """Obtain a file suitable for fulfilling `requirement`

        DEPRECATED; use the ``fetch_distribution()`` method now instead.  For
        backward compatibility, this routine is identical but returns the
        ``location`` of the downloaded distribution instead of a distribution
        object.
        """
        dist = self.fetch_distribution(requirement,tmpdir,force_scan,source)
        if dist is not None:
            return dist.location
        return None


    def gen_setup(self, filename, fragment, tmpdir):
        match = EGG_FRAGMENT.match(fragment)
        dists = match and [d for d in
            interpret_distro_name(filename, match.group(1), None) if d.version
        ] or []

        if len(dists)==1:   # unambiguous ``#egg`` fragment
            basename = os.path.basename(filename)

            # Make sure the file has been downloaded to the temp dir.
            if os.path.dirname(filename) != tmpdir:
                dst = os.path.join(tmpdir, basename)
                from setuptools.command.easy_install import samefile
                if not samefile(filename, dst):
                    shutil.copy2(filename, dst)
                    filename=dst

            file = open(os.path.join(tmpdir, 'setup.py'), 'w')
            file.write(
                "from setuptools import setup\n"
                "setup(name=%r, version=%r, py_modules=[%r])\n"
                % (
                    dists[0].project_name, dists[0].version,
                    os.path.splitext(basename)[0]
                )
            )
            file.close()
            return filename

        elif match:
            raise DistutilsError(
                "Can't unambiguously interpret project/version identifier %r; "
                "any dashes in the name or version should be escaped using "
                "underscores. %r" % (fragment,dists)
            )
        else:
            raise DistutilsError(
                "Can't process plain .py files without an '#egg=name-version'"
                " suffix to enable automatic setup script generation."
            )

    dl_blocksize = 8192
    def _download_to(self, url, filename):
        self.info("Downloading %s", url)
        # Download the file
        fp, tfp, info = None, None, None
        try:
            if '#' in url:
                url, info = url.split('#', 1)
            fp = self.open_url(url)
            if isinstance(fp, HTTPError):
                raise DistutilsError(
                    "Can't download %s: %s %s" % (url, fp.code,fp.msg)
                )
            cs = md5()
            headers = fp.info()
            blocknum = 0
            bs = self.dl_blocksize
            size = -1
            if "content-length" in headers:
                # Some servers return multiple Content-Length headers :(
                sizes = get_all_headers(headers, 'Content-Length')
                size = max(map(int, sizes))
                self.reporthook(url, filename, blocknum, bs, size)
            tfp = open(filename,'wb')
            while True:
                block = fp.read(bs)
                if block:
                    cs.update(block)
                    tfp.write(block)
                    blocknum += 1
                    self.reporthook(url, filename, blocknum, bs, size)
                else:
                    break
            if info: self.check_md5(cs, info, filename, tfp)
            return headers
        finally:
            if fp: fp.close()
            if tfp: tfp.close()

    def reporthook(self, url, filename, blocknum, blksize, size):
        pass    # no-op


    def open_url(self, url, warning=None):
        if url.startswith('file:'):
            return local_open(url)
        try:
            return open_with_auth(url, self.opener)
        except (ValueError, httplib.InvalidURL):
            v = sys.exc_info()[1]
            msg = ' '.join([str(arg) for arg in v.args])
            if warning:
                self.warn(warning, msg)
            else:
                raise DistutilsError('%s %s' % (url, msg))
        except urllib2.HTTPError:
            v = sys.exc_info()[1]
            return v
        except urllib2.URLError:
            v = sys.exc_info()[1]
            if warning:
                self.warn(warning, v.reason)
            else:
                raise DistutilsError("Download error for %s: %s"
                                     % (url, v.reason))
        except httplib.BadStatusLine:
            v = sys.exc_info()[1]
            if warning:
                self.warn(warning, v.line)
            else:
                raise DistutilsError('%s returned a bad status line. '
                                     'The server might be down, %s' % \
                                             (url, v.line))
        except httplib.HTTPException:
            v = sys.exc_info()[1]
            if warning:
                self.warn(warning, v)
            else:
                raise DistutilsError("Download error for %s: %s"
                                     % (url, v))

    def _download_url(self, scheme, url, tmpdir):
        # Determine download filename
        #
        name, fragment = egg_info_for_url(url)
        if name:
            while '..' in name:
                name = name.replace('..','.').replace('\\','_')
        else:
            name = "__downloaded__"    # default if URL has no path contents

        if name.endswith('.egg.zip'):
            name = name[:-4]    # strip the extra .zip before download

        filename = os.path.join(tmpdir,name)

        # Download the file
        #
        if scheme=='svn' or scheme.startswith('svn+'):
            return self._download_svn(url, filename)
        elif scheme=='git' or scheme.startswith('git+'):
            return self._download_git(url, filename)
        elif scheme.startswith('hg+'):
            return self._download_hg(url, filename)
        elif scheme=='file':
            return url2pathname(urlparse(url)[2])
        else:
            self.url_ok(url, True)   # raises error if not allowed
            return self._attempt_download(url, filename)

    def scan_url(self, url):
        self.process_url(url, True)


    def _attempt_download(self, url, filename):
        headers = self._download_to(url, filename)
        if 'html' in headers.get('content-type','').lower():
            return self._download_html(url, headers, filename)
        else:
            return filename

    def _download_html(self, url, headers, filename):
        file = open(filename)
        for line in file:
            if line.strip():
                # Check for a subversion index page
                if re.search(r'<title>([^- ]+ - )?Revision \d+:', line):
                    # it's a subversion index page:
                    file.close()
                    os.unlink(filename)
                    return self._download_svn(url, filename)
                break   # not an index page
        file.close()
        os.unlink(filename)
        raise DistutilsError("Unexpected HTML page found at "+url)
















    def _download_svn(self, url, filename):
        url = url.split('#',1)[0]   # remove any fragment for svn's sake
        creds = ''
        if url.lower().startswith('svn:') and '@' in url:
            scheme, netloc, path, p, q, f = urlparse(url)
            if not netloc and path.startswith('//') and '/' in path[2:]:
                netloc, path = path[2:].split('/',1)
                auth, host = splituser(netloc)
                if auth:
                    if ':' in auth:
                        user, pw = auth.split(':',1)
                        creds = " --username=%s --password=%s" % (user, pw)
                    else:
                        creds = " --username="+auth
                    netloc = host
                    url = urlunparse((scheme, netloc, url, p, q, f))
        self.info("Doing subversion checkout from %s to %s", url, filename)
        os.system("svn checkout%s -q %s %s" % (creds, url, filename))
        return filename

    def _vcs_split_rev_from_url(self, url, pop_prefix=False):
        scheme, netloc, path, query, frag = urlsplit(url)

        scheme = scheme.split('+', 1)[-1]

        # Some fragment identification fails
        path = path.split('#',1)[0]

        rev = None
        if '@' in path:
            path, rev = path.rsplit('@', 1)

        # Also, discard fragment
        url = urlunsplit((scheme, netloc, path, query, ''))

        return url, rev

    def _download_git(self, url, filename):
        filename = filename.split('#',1)[0]
        url, rev = self._vcs_split_rev_from_url(url, pop_prefix=True)

        self.info("Doing git clone from %s to %s", url, filename)
        os.system("git clone --quiet %s %s" % (url, filename))

        if rev is not None:
            self.info("Checking out %s", rev)
            os.system("(cd %s && git checkout --quiet %s)" % (
                filename,
                rev,
            ))

        return filename

    def _download_hg(self, url, filename):
        filename = filename.split('#',1)[0]
        url, rev = self._vcs_split_rev_from_url(url, pop_prefix=True)

        self.info("Doing hg clone from %s to %s", url, filename)
        os.system("hg clone --quiet %s %s" % (url, filename))

        if rev is not None:
            self.info("Updating to %s", rev)
            os.system("(cd %s && hg up -C -r %s >&-)" % (
                filename,
                rev,
            ))

        return filename

    def debug(self, msg, *args):
        log.debug(msg, *args)

    def info(self, msg, *args):
        log.info(msg, *args)

    def warn(self, msg, *args):
        log.warn(msg, *args)













# This pattern matches a character entity reference (a decimal numeric
# references, a hexadecimal numeric reference, or a named reference).
entity_sub = re.compile(r'&(#(\d+|x[\da-fA-F]+)|[\w.:-]+);?').sub

def uchr(c):
    if not isinstance(c, int):
        return c
    if c>255: return unichr(c)
    return chr(c)

def decode_entity(match):
    what = match.group(1)
    if what.startswith('#x'):
        what = int(what[2:], 16)
    elif what.startswith('#'):
        what = int(what[1:])
    else:
        what = name2codepoint.get(what, match.group(0))
    return uchr(what)

def htmldecode(text):
    """Decode HTML entities in the given text."""
    return entity_sub(decode_entity, text)















def socket_timeout(timeout=15):
    def _socket_timeout(func):
        def _socket_timeout(*args, **kwargs):
            old_timeout = socket.getdefaulttimeout()
            socket.setdefaulttimeout(timeout)
            try:
                return func(*args, **kwargs)
            finally:
                socket.setdefaulttimeout(old_timeout)
        return _socket_timeout
    return _socket_timeout

def _encode_auth(auth):
    """
    A function compatible with Python 2.3-3.3 that will encode
    auth from a URL suitable for an HTTP header.
    >>> _encode_auth('username%3Apassword')
    u'dXNlcm5hbWU6cGFzc3dvcmQ='
    """
    auth_s = unquote(auth)
    # convert to bytes
    auth_bytes = auth_s.encode()
    # use the legacy interface for Python 2.3 support
    encoded_bytes = base64.encodestring(auth_bytes)
    # convert back to a string
    encoded = encoded_bytes.decode()
    # strip the trailing carriage return
    return encoded.rstrip()

def open_with_auth(url, opener=urllib2.urlopen):
    """Open a urllib2 request, handling HTTP authentication"""

    scheme, netloc, path, params, query, frag = urlparse(url)

    # Double scheme does not raise on Mac OS X as revealed by a
    # failing test. We would expect "nonnumeric port". Refs #20.
    if netloc.endswith(':'):
        raise httplib.InvalidURL("nonnumeric port: ''")

    if scheme in ('http', 'https'):
        auth, host = splituser(netloc)
    else:
        auth = None

    if auth:
        auth = "Basic " + _encode_auth(auth)
        new_url = urlunparse((scheme,host,path,params,query,frag))
        request = urllib2.Request(new_url)
        request.add_header("Authorization", auth)
    else:
        request = urllib2.Request(url)

    request.add_header('User-Agent', user_agent)
    fp = opener(request)

    if auth:
        # Put authentication info back into request URL if same host,
        # so that links found on the page will work
        s2, h2, path2, param2, query2, frag2 = urlparse(fp.url)
        if s2==scheme and h2==host:
            fp.url = urlunparse((s2,netloc,path2,param2,query2,frag2))

    return fp

# adding a timeout to avoid freezing package_index
open_with_auth = socket_timeout(_SOCKET_TIMEOUT)(open_with_auth)











def fix_sf_url(url):
    return url      # backward compatibility

def local_open(url):
    """Read a local path, with special support for directories"""
    scheme, server, path, param, query, frag = urlparse(url)
    filename = url2pathname(path)
    if os.path.isfile(filename):
        return urllib2.urlopen(url)
    elif path.endswith('/') and os.path.isdir(filename):
        files = []
        for f in os.listdir(filename):
            if f=='index.html':
                fp = open(os.path.join(filename,f),'rb')
                body = fp.read()
                fp.close()
                break
            elif os.path.isdir(os.path.join(filename,f)):
                f+='/'
            files.append("<a href=%r>%s</a>" % (f,f))
        else:
            body = ("<html><head><title>%s</title>" % url) + \
                "</head><body>%s</body></html>" % '\n'.join(files)
        status, message = 200, "OK"
    else:
        status, message, body = 404, "Path not found", "Not found"

    return HTTPError(url, status, message,
            {'content-type':'text/html'}, StringIO(body))













# this line is a kludge to keep the trailing blank lines for pje's editor