-
Notifications
You must be signed in to change notification settings - Fork 18
/
Copy pathChanges
1363 lines (851 loc) · 45.5 KB
/
Changes
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
Revision history for Perl module CPAN::Reporter
{{$NEXT}}
1.2020 2024-08-13 01:44:56-03:00 America/Sao_Paulo
[ADDED]
- Add support for UTF-8 in config file
- Adapt to new bailout output in TAP::Harness 3.49
1.2019 2023-05-01 03:14:19+02:00 Europe/Paris
[ADDED]
- Record whether perl was built with taint support enabled or not
- Include Test2 version in toolchain report
- snip in the middle of the log, not a the end, so you get
potentially more interesting data
[CHANGED]
- replaced IO::CaptureOutput with Capture::Tiny
- Eliminate dependency on File::Copy::Recursive.
1.2018 2016-06-21 15:05:12-04:00 America/New_York
[CHANGED]
- Expand list of environment variables included in reports
1.2017 2016-03-02 15:49:43-05:00 America/New_York
[CHANGED]
- Distributions that appear to be in a Perl6/ directory are not
graded and reports are discarded.
1.2016 2016-02-28 22:39:04-05:00 America/New_York
[FIXED]
- Improve prerequisite version detection for dynamic versions
- Add support for static Config like cperl and XSConfig
1.2015 2015-08-21 17:53:31-04:00 America/New_York
- Allow custom comment strings to be used as the default comment, using
comment.txt in the config directory [JMASLAK]
1.2014 2015-04-06 12:08:28-04:00 America/New_York
- no changes from 1.2013-TRIAL
1.2013 2015-03-29 21:52:11-04:00 America/New_York (TRIAL RELEASE)
[ADDED]
- Added experimental option to save output of hanging modules [CHORNY]
- Added experimental option to save list of hanging modules [CHORNY]
[CHANGED]
- C::R::PrereqCheck will read <> before checking modules
[FIXED]
- Fixed incorrect report grade edge case [CHORNY]
- Fixed tests for Perls on Windows with spaces in the path
- Added Dancer::Plugin::FlashNote to list of prereq-check conflicts
[TESTING]
- Tests for C::R::PrereqCheck added
1.2011 2014-03-16 22:27:56+01:00 Europe/Paris
[ADDED]
- Added support for future optional prerequisite reporting
when CPAN.pm supports it
[CHANGED]
- Moved test distributions into corpus/ to avoid repetitive
tarball unpacking
- Modernized distribution meta and support files
1.2010 2013-04-12 15:35:38 Europe/London
[NEW FEATURES]
- new option 'retry_submission' tries to send the report one
extra time in case it fails. Defaults to undef (don't retry)
(Patch by Alexandr "chorny" Ciornii)
1.2009 2012-12-18 16:08:37 America/New_York
[BUG FIXES]
- Expanded prerequisite load check skip to all Acme::* modules; prereqs
still need to be installed and of sufficient version, but we don't
check that Acme modules load without error to avoid side effects.
1.2008 2012-12-15 07:26:11 America/New_York
[BUG FIXES]
- Don't try to check if Acme::Bleach can be loaded if it's listed
as a prereq
1.2007 2012-12-10 15:13:22 America/New_York
[BUG FIXES]
- Recognize another form of "Perl is too low" message [CHORNY]
1.2006 2012-04-10 18:44:20 America/New_York
[BUG FIXES]
- Changed how CPAN::Reporter::PrereqCheck attempts to load modules to
avoid module reloading, which caused some prereqs to show up as
'broken' [rt.cpan.org #76394 and #75559]
[OTHER]
- Bumped IPC::Cmd prereq to 0.76 to avoid failures on some platforms
1.2005 2012-03-02 13:21:36 EST5EDT
[BUG FIXES]
- Canonicalise the config path returned by bsd_glob [rt.cpan.org #70504]
[Kent Fredric]
1.2004 2012-02-13 16:25:07 EST5EDT
[BUG FIXES]
- Fix bug in testing tilde expansion on Windows
[rt.cpan.org #72051]
1.2003 2011-12-06 11:07:04 EST5EDT
[BUG FIXES]
- Fix bug in transport class validation to allow multi-level
class names [rt.cpan.org #73045]
1.2002 2011-08-10 17:52:08 America/New_York
[BUG FIXES]
- Fix failing tests on systems without
Test::Reporter::Transport::Metabase
- Minor test cleanups for use with 'prove'
1.2001 2011-08-07 05:05:58 America/New_York
[BUG FIXES]
- Fix failing test on Windows due to glob expansion of "~"
[Christian Walde]
1.20 2011-08-05 13:17:20 America/New_York
[BUG FIXES]
- Removed Test::Reporter::Transport::Metabase as a prerequisite.
New testers are recommended to install Task::CPAN::Reporter instead
of CPAN::Reporter.
1.19_04 2011-06-29 17:39:25 America/New_York
[NEW FEATURES]
- Metabase profile generation will use as much of the provided
email_from parameter as possible and will generate a random
secret API key. If user provides email_from in a standard form
like C<< "John Q. Public" <jqp@example.com> >> then profile
generation is entirely automatic without additional prompts
from metabase-profile [David Golden]
[DOCUMENTATION]
- Swapped in Task::CPAN::Reporter in the SYNOPSIS [David Golden]
[BUG FIXES]
- Expands "~" into user directory before attempting to locate
'relative' pathname in .cpanreporter directory [David Golden]
1.19_03 2011-06-28 14:44:19 America/New_York
[NEW FEATURES]
- During configuration, if 'Metabase' is the requested transport and
a Metabase profile file is not found, configuration will offer to
run metabase-profile to create it. [David Golden]
- For Metabase transport (only), the 'id_file' may have a relative
file path, treated as relative to the .cpanreporter directory.
[David Golden]
[BUG FIXES]
- updated CPAN::Reporter to go with Metabase by default, instead of
the deprecated SMTP transport system. This provides more up-to-date
information for new testers on how to setup their systems and
addresses RT#64316 (and maybe also RT#61735) [Breno G. de Oliveira]
- Replaced use of Tee.pm with Capture::Tiny as the two don't play
well together on some systems. [Christian Walde]
[DOCUMENTATION]
- Docs have been substantially re-written to reflect the new HTTP-based
transport instead of email-based transport. [Breno G. de Oliveira and
David Golden]
1.1902 2011-01-28 15:22:35 EST5EDT
[BUG FIXES]
- Updated Tee prereq to 0.14 to fix an instability on Windows
1.1901 2011-01-24 20:23:23 EST5EDT
- Fixes some space and backslash escaping for Windows [Christian Walde]
1.1900 2011-01-21 10:55:28 EST5EDT
- No changes from 1.18_06
1.18_06 2010-12-07 05:18:51 EST5EDT
Fixed:
- Reports with a "can't spawn" error from TAP::Parser::Iterator::Process
are discarded instead of reporting FAIL
Other:
- CPAN::Reporter is now released using Dist::Zilla. This required
some changes to various files help dzil find/skip various bits of data.
It should have no runtime effect.
1.18_05 Sun Oct 24 21:25:22 EDT 2010
Fixed:
- Fixed error in calling prerequisite detection script when there
are spaces in the path [RURBAN]
1.18_04 NOT RELEASED TO CPAN
Changed:
- Executed processes are given their own process group and timeouts are
now handled by sending SIGKILL to the entire child process group.
This is an attempt to address RT#61912
1.1803 Thu Sep 30 21:49:10 EDT 2010
Fixed:
- Fixed t/13_record_command.t to reflect removal of Proc::ProcessTable
as a conditional dependency and fixed a resulting bug in the test
- Removed mention of Proc::ProcessTable in CPAN::Reporter::Config
1.1802 Wed Sep 29 23:19:15 EDT 2010
Changed:
- Given the migration to CT2.0, the maximum report length before
truncation has been increased from 50 to 1,000 kB.
- Removed dependency on Proc::ProcessTable for using a command timeout
on non-Win32 platforms.
1.1801 Fri Sep 10 16:34:29 EDT 2010
Changed:
- Default config directory now uses File::HomeDir->my_home on
windows instead of ->my_documents. If that directory doesn't exist,
it will fall back to ->my_documents.
1.1800 Mon Jul 26 16:02:06 EDT 2010
Changed:
- "sending test report..." message now shows transport method, not
the old email list address for less confusion when using non-email
transports
- Enhances missing "email_from" prompt to clarify why an email is
still needed under non-email transports for legacy compatibililty
1.1711 Wed Feb 17 13:52:14 EST 2010
Bugs fixed:
- Tests for "no make test" were failing due to internationalization
of the error message. Rather than try to interpret every possible
make error message, CPAN::Reporter now just looks inside the Makefile
for a "test" target. Hopefully, that will prove more robust.
1.1710 Wed Feb 10 10:30:51 EST 2010
Bugs fixed:
- Tests for "no make test" were failing on some systems due to
the incredibly wide range of error messages possible from
make/nmake/dmake on various systems. The error message check is
now even more liberal.
1.1709 Mon Feb 1 22:39:59 EST 2010
Bugs fixed:
- Fixed PrereqCheck for modules like "if.pm" that could be confused
with Perl syntax (RT#53477 by Serguei Trouchelle)
- Added PrereqCheck special-case for Module::Install and improved
diagnostics when prereqs are missing or too low (RT#51257)
- Distributions with no 'make test' target should be UNKNOWN,
not FAIL (RT#52139)
- Improved detection of Test::Reporter capable of setting 'distfile'
parameter
Documentation changes:
- Added 'Port' example for transport options
- Adopted new format for Changes
1.1708 Mon Jun 1 23:31:58 EDT 2009
*** Emergency fix -- 1.1706/1.1707 can prevent subsequent module ***
*** installation if installed with an older Test::Reporter. ***
*** Disable test reporting (From CPAN, 'o conf test_report 0') ***
*** and install CPAN::Reporter 1.1708. Then re-enable test ***
*** reports. ***
- Fixed: in the case where an old Test::Reporter without the distfile()
method is installed, CPAN::Reporter would crash and prevent
installation of modules (including newer Test::Reporter); now checks
for a Test::Reporter
1.1707 Sun May 31 21:44:07 EDT 2009
- Prereq: bumped Test::Reporter to 1.54
1.1706 Sun May 31 14:00:43 EDT 2009
- Changed: URL of CPAN Testers wiki is now wiki.cpantesters.org
- Added: support for setting 'distfile' parameter for test reports
if supported by Test::Reporter (1.53_01+)
1.1705 Tue Jan 20 10:51:24 EST 2009
- Docs: CPAN Testers website updated to www.cpantesters.org
(David Westbrook)
1.1704 Sun Dec 7 06:24:07 EST 2008
- Fixed: protect CPAN::Reporter tests from older versions of itself
- Prereq: bumped Devel::Autoflush to 0.04 and CPAN to 1.9301
1.1703 Fri Dec 5 23:06:20 EST 2008
- Changed: Devel::Autoflush only added to PERL5OPT when capturing
output from Makefile.PL or Build.PL
- Prereq: bumped Devel::Autoflush to 0.03 for various fixes
1.1702 Sat Sep 13 07:20:04 EDT 2008
*** Emergency fix -- 1.1701 tickled a new Test::Harness bug ***
- Changed: reverted to old -I and -M flag order in PERL5OPT; -I flag
only added on Perl 5.8+; this should still fix Perl < 5.8 that can
only take a single flag without provoking the T::H bug
1.1701 Thu Sep 11 23:56:41 EDT 2008
- Fixed: changed order of -I and -M flags in PERL5OPT to have -M first
(since Perl 5.6.2 only supports a single flag); 5.6.2 testers may yet
again see complaints about missing Devel::Autoflush when authors call
'perl' instead of $^X in tests (oh, well...)
- Changed: bumped Test::Reporter prereq to 1.50
1.17 Sat Sep 6 19:35:44 EDT 2008
- Changed: In line with changes to CPAN Testers grade definitions,
failures during the PL or make/Build phase will now be reported as
UNKNOWN instead of FAIL
- Removed: In line with the CPAN Testers decision to no longer have
individual testers send reports to authors, the "cc_author" option
has been removed
- Fixed: "Build -j3" type errors are detected and reports discarded
- Fixed: Makefile out of date vs Makefile.PL errors detected and discarded
- Docs: updated FAQ
1.16_51 Wed Jul 16 12:41:48 EDT 2008
- Fixed: t/11_env_config was failing when tested under
CPAN::Reporter due to PERL5OPT manipulations
1.16_50 Wed Jul 16 09:41:20 EDT 2008
- Added: restored "-I" for PERL5OPT, but to a temporary lib
directory containing a copy of Devel::Autoflush.
- Fixed: enviroment variables section of report now reflects
PERL5OPT as used during build and testing
1.1601 Sat Jul 5 09:37:14 EDT 2008
- Fixed: reverted 1.15_56 PERL5OPT changes; Devel::Autoflush INC path
no longer added with "-I" as it causes old modules to be found during
testing. Testers will have to live with authors complaining about
Devel::Autoflush missing when authors call 'perl' instead of $^X
- Fixed: timeout support on *nix checks for *both* Proc::Killfam and
Proc::ProcessTable to protect against unauthorized copies of only
Proc::Killfam bundled in distributions like Tk-ExecuteCommand
1.16 Fri Jul 4 09:14:52 EDT 2008
- No changes since 1.15_56
- Summary of major features/fixes since 1.15:
- Added: support for 'configure_requires' in META.yml (adds
Parse::CPAN::Meta as dependency)
- Fixed: uses Devel::Autoflush to ensure buggy *.PL files that
prompt without a newline are always visible through the 'tee'
pipe
- Fixed: on Unix, changed command timeout approach to avoid
disconnecting STDIN from the terminal; timeouts now require
Proc::Killfam on *nix platforms
1.15_56 Thu Jun 19 22:19:33 EDT 2008
- Fixed: when multiple versions of perl are installed, distributions
that called 'perl' during testing instead of '$^X' would die trying
to load Devel::Autoflush if the first perl in PATH didn't have
Devel::Autoflush installed; fixed by added -I flag to PERL5OPT with
the path to Devel::Autoflush to ensure it can be found; (calling
'perl' is actually a subtle testing bug, but this way it doesn't look
like it's the tester's fault)
- Documented: removed suggestion to subscribe to cpan-testers since
that no longer is necessary now that cpan-testers is just a
drop-box (noted by Jose Luis Martinez)
1.15_55 Mon Jun 9 08:11:01 EDT 2008
- Fixed: catch META.yml parse failures from Parse::CPAN::Meta
(Reported by Andreas Koenig)
1.15_54 Fri Jun 6 14:06:57 EDT 2008
- Fixed: would disconnect from terminal input when running interactively
with an inactivity timeout; fixed by avoiding use of setpgrp(), but
now requires Proc::Killfam to support inactivity timeouts on non Win32
platforms. (Bug reported by Andy Armstrong)
1.15_53 Thu Jun 5 06:37:13 EDT 2008
- Fixed: PERL5OPT could was being set with a leading space, which can
break distributions being tested under Test::Harness
1.15_52 Wed Jun 4 22:23:39 EDT 2008
- Fixed: t/11_env_config.t had a bug that would fail incorrectly if
an environment variable had a leading space
1.15_51 Wed Jun 4 17:57:42 EDT 2008
- Fixed: t/13_record_command.t relative path tests are incompatible
with Devel::Autoflush; removed the broken tests
1.15_50 Tue Jun 3 12:13:12 EDT 2008
- Added: on FAIL reports, now checks for missing configure_requires
and discards the report if configure_requires is not satisfied.
(Adds Parse::CPAN::Meta as dependency.)
- Fixed: some *.PL files prompt without autoflush, which works
normally, but appears to silently hang when piping to a tee.
CPAN::Reporter now uses Devel::Autoflush in PERL5OPT to always turn
on autoflush
- Fixed: distributions with odd names wouldn't be recorded in history;
now they are recorded as 'DISCARD'
1.15 Mon May 19 11:21:47 EDT 2008
- Changed: now requires Test::Reporter 1.40
- No other changes from 1.14_xx series
1.14_04 Thu Apr 17 06:37:26 EDT 2008
- Documented: added section on transport config option
1.14_03 Thu Apr 17 06:37:26 EDT 2008
- Uploaded without Changes file updated
1.14_02 Sun Apr 6 09:45:39 EDT 2008
- Changed: improved diagnostics messages for 'transport' failures
1.14_01 Sat Apr 5 18:30:44 EDT 2008
- Changed: if 'transport' configuration option is not valid according
to Test::Reporter, no report will be sent instead of falling back
to Net::SMTP
1.13 Mon Mar 24 09:22:44 EDT 2008
- Changed: Build.PL or Makefile.PL that returns without error but
does *not* create either 'Makefile' or 'Build' will be marked as
'DISCARD' for the PL phase instead of 'PASS'.
- Fixed: Prereq check would be confused if loading modules also
printed anything to STDOUT (somehow CPAN.pm is a culprit); added
"select DEVNULL" before requiring modules as a workaround
1.12 Mon Mar 10 08:55:48 EDT 2008
- Added: Win32::FsType() to special variables section of report
- Added: diagnostic warning if PrereqCheck output can't be parsed
1.11 Mon Feb 25 18:47:03 EST 2008
- Bug fix: successful PL's where prereqs happen not to be satisfied
were mistakenly being marked as 'DISCARD' for the PL phase. Fixed so
that PL's that PASS will stay that way regardless of prerequisites.
(This does not prevent a later make or test failure from being
downgraded,
- Bug fix: skipfiles patterns should match case-insensitive (Cantrell);
also, documented lack of starting anchor
- Bug fix: command_timeout option would not accept zero as valid; now
zero is a valid option and disables command_timeout
- more minor boilerplate tweaks
1.10 Mon Feb 25 10:07:28 EST 2008
I hereby dub 1.10 as the "I hate Windows" release
- Bug fix (sort of): On Win32, child processes were not timing out.
Changed from Win32::Process to Win32::Job to fix that. However,
discovered that process termination on Win32 can deadlock in some
cases, so timeout testing on Win32 will now skip unless
$ENV{PERL_AUTHOR_TESTING} is set. Added warning to documentation.
- Bug fix: On a Win32 virtual machine (VirtualBox), handles for output
capture were not inherited by child processes, which caused Test::Harness
to crash for some reason.
1.09 Mon Feb 18 22:10:55 EST 2008
- Simplified and shortened boilerplate
- Bug fix (minor): better detection of make vs Build for progress
messages
- Test fix: skip interrupt testing on MSWin32 if Win32::Process
is not installed
1.0801 Sun Feb 10 00:27:05 EST 2008
- Test fix: adjust timeout timings in t/14_command_timeout.t to try to
avoid failures on heavily loaded machines
1.08 Sat Jan 26 18:52:45 EST 2008
- No changes from 1.07_06 (seems to pass smoke tests)
- Summary of new features since 1.06:
- Added have_tested() function to CPAN::Reporter::History
- Added 'cc_skipfile' and 'send_skipfile' advanced config options
- Added 'command_timeout' advanced config option
- Added support for PERL_CPAN_REPORTER_DIR and
PERL_CPAN_REPORTER_CONFIG environment variables
1.07_06 Fri Jan 25 11:27:28 EST 2008
- Added 'command_timeout' config to halt commands after a period of time
- Added detection of commands killed with a signal; reports are discarded
(N.B. this may change in a future version)
- Bug fix: processes timing out could still hang with child processes
(killing 'make test' still leaves the harness running); fixed by making
'make test' its own process group and killing the process group; hoping
that's portable across non-Win32; Win32 timeouts use Win32::Process and
do the right thing already
- Downgraded lots of progress messages from CPAN's 'mywarn' to 'myprint'
(from red to blue if you use CPAN's colored output)
1.07_05 Thu Jan 24 08:27:48 EST 2008
- Added PERL_CPAN_REPORTER_DIR and PERL_CPAN_REPORTER_CONFIG environment
variables to specify alternate locations for default configuration
directory and file (RT#30314)
- Removed File::Temp as a full prerequisite; now only used in testing
(Requested by Adam Foxson to extract features to CPAN::Testers::*)
1.07_04 Sun Jan 20 22:06:56 EST 2008
- Test fix: stopping cc on blead broke lots of other tests. Fixed those.
Built my own blead perl so I can actually check these things in the
future instead of relying on Andreas' blead smoking
- Test fix: testing skipfile wasn't being created properly on Perl 5.005
1.07_03 Sat Jan 19 22:16:44 EST 2008
- Test fix: set sample history file writeable after copy
1.07_02 Sat Jan 19 20:33:55 EST 2008
- Added have_tested() function to CPAN::Reporter::History; provide ability
to search the CPAN::Reporter history file
- Bug fix: PL files that warned "OS Unsupported" and exited with 0 will
now be flagged as NA; this bends the CPAN Testers "rules" slightly, but
"OS Unsupported" is a hack anyway, so this should help things DWIM
- Test fix: more changes to timeout testing to avoid false failures
- Authors will never be CC'd on reports when running maint/blead perl
(anything with $Config{perl_patchlevel} set)
- Reports discarded due to missing prerequisites will now be recorded
in the history file with a 'grade' of DISCARD (unless a duplicate
DISCARD already exists)
- Changed "edit_report" prompt to clarify this can be used to review
reports in addition to just editing them
- Shortened duplicate report warning to fit on one line
1.07_01 Tue Jan 8 01:41:02 EST 2008
- Added "cc_skipfile" and "send_skipfile" advanced config options to
specify a text file of regex patterns; if a distribution ID
('AUTHOR/Dist-Name-0.01.tar.gz') matches any pattern in the file, the
author will not be copied on the report (cc_skipfile) or no reports
will be sent at all (skip_sendfile)
1.0602 Tue Dec 25 08:59:07 EST 2007
- Test fix: a timeout test would sometimes appear to fail on a highly
loaded system; changed the runtime and timeout timings to try to
prevent false failures; added better diagnostic output
1.0601 Tue Dec 11 12:19:25 EST 2007
- Test fix: Test::Harness 3.05 fixed a long standing bug; the fix
broke our tests
1.06 Thu Nov 15 14:17:04 EST 2007
- Bug fix: if older versions of a module are in multiple
locations in @INC, CPAN::Reporter could detect the older version
instead of the newer one; fixed in CPAN::Reporter::PrereqCheck
(RT#30345; additional investigation by Andy Armstrong)
- Bug fix: "our $VERSION" check still wasn't working on 5.005; now
fixed (reported by Slaven Rezic). Note, this situation only
applies to Module::Build <= 0.2808, where Build.PL dies before a
legitimate "require => { perl => 5.006 }" is processed. (Makefile.PL
does not die in this situation and Makefile.PL has no way of
signaling a required perl version other than "use 5.006" or equivalent
in the Makefile.PL)
- Bug fix: wasn't printing Test::Reporter error message if a report
couldn't be sent (RT#30730 by BLOONIX)
- Changed FAIL report boilerplate about CPAN::Testers wiki to be
more explicit that the wiki teaches how to deal with dependencies, OS,
etc. (suggested by Olly Betts)
1.05 Tue Nov 6 14:41:48 EST 2007
- Added checks for Makefile.PL/Build.PL dying early on Perl < 5.6 if
modules define their version using "our $VERSION"; such cases will
be graded NA (RT#30299; report and partial patch from Slaven Rezic)
- Added author's email address to the prompt question for whether to
CC the author. Also fixed typo in the prompt.
(suggested by Slaven Rezic)
- Added "CPAN::Reporter" consistently as a prefix to all console
messages to distinguish them from messages from CPAN.pm. Noisy, but
helps track errors/complaints to the right source.
- Bug fix: on first installation, "reload cpan" might load CPAN::Reporter
when an old version File::Temp is already loaded; added an explicit
version check on File::Temp (reported by Andreas Koenig)
1.04 Fri Oct 26 06:56:32 EDT 2007
- Bug fix: temp files were being created in the current directory,
not the File::Spec->tmpdir() directory; this broke distributions
that used their own distribution directory for signature or
MANIFEST checks
1.03 Wed Oct 24 17:13:59 EDT 2007
- Added a check for prerequisites that are present but fail to load;
in such a case the prereq "have" field will show "broken" (e.g a
dependency of that module is missing)
- Refactored prerequisite checking code into a separate modulino
CPAN::Reporter::PrereqCheck to cut down on excessive tempfile use
and simplify testing
- Added name templates for all tempfiles to help identify leakage
or left-over files if tests are interrupted. All names are like
CR-ZZ-XXXXXXXX, where ZZ is unique to each File::Temp call and X's
are a random suffix
- Testing bug fix: in the rare case where Module::Build is present but
broken, we would only skip Build.PL tests once instead of always
1.02 Sun Oct 14 08:57:38 EDT 2007
- Bug fix: shell quotes in command passed to record_command() were not
being properly escaped in creation of the wrapper script
(found by Andreas Koenig)
- Bug fix: wasn't detecting Perl version too low during make/Build
phase. Reports were improperly graded as 'FAIL' instead of 'NA'
(found by Slaven Rezic and David Cantrell)
- Bug fix: wasn't detecting missing prerequisites during "make" phase
and FAIL reports were being sent that should have been discarded
(found by Michel Rodriguez and David Cantrell)
1.01 Sat Oct 13 03:21:40 EDT 2007
- Bug fix: no prompt for sending reports was being shown if
advanced send_*_report options were set. (Found by David Cantrell)
- Bug fix: was not detecting Module::Install's perl version warning
(found by Slaven Rezic)
1.00 Wed Oct 10 20:39:42 EDT 2007
*** New major release -- API changes and other enhancements ***
Summary of major changes from 0.XX series:
* Added support for reporting for *.PL and make/Build stages; bumped
CPAN.pm prerequisite to 1.9203 to take advantage of this support
* Added support for the forthcoming Test::Harness 3.00
* Changed the name and format of the history file of sent reports to
track history by PL/make/test phase. Old history.db will be
automatically upgraded to new reports-sent.db.
* Moved 'cc_author' and 'send_duplicates' options from interactive
to advanced (manual) configuration; defaults are strongly recommended
* Bumped Test::Reporter prereq to 1.34 for transport() method and set
default transport to Net::SMTP on all platforms
Additional minor new change in 1.00:
- Added CPAN Testers wiki URL to explanation paragraph for FAIL reports
(request by Slavan Rezic)
0.99_15 Mon Oct 8 08:33:56 EDT 2007
This release is dedicated to Slaven Rezic for boldly smoking with
the pre-release of Perl 5.005_05.
- Work around missing 5.005 Fcntl constants (noted by Slaven Rezic)
- Fix bug detecting perl version failure in test output
(found by Slaven Rezic while testing Sub::Uplevel on 5.00505)
- Fix bug detecting OS unsupported in test.pl and make
- Test output truncated at 50K to ensure reports will be accepted
by perl.org's MX and to avoid excessive output in general
(MX limitation noted by Slaven Rezic)
- Added documentation note in History.pm that all methods are
currently private
- Removed API.pm and FAQ.pm from MANIFEST and tarball -- these are
just source files for API.pod and FAQ.pod generated with Pod::WikiDoc
0.99_14 Tue Oct 2 14:49:40 EDT 2007
* Bumped CPAN.pm prereq to 1.9203 -- fixes a bug where make/build
continues even if Makefile/Build is missing, causing FAIL
reports for modules that exit(0) without generating a Makefile.
- Set transport() default to Net::SMTP; ensures CPAN::Reporter can
work out-of-the-box on systems that happen to have Mail::Send and
a broken sendmail installation
- eliminated several Perl::Critic level 4-5 complaints
- added some entries to the FAQ
0.99_13 Thu Sep 27 06:24:00 EDT 2007
* Bump Test::Reporter prereq to 1.34 for transport() method. Added
"transport" advanced configuration option.
(Requested by Andreas Koenig)
- Improved robustness of conversion of old history files on Win32 for
really old CPAN::Reporter history formats
- Minor report format tweaks
0.99_12 Sun Sep 23 13:14:08 EDT 2007
* Bumped CPAN.pm prereq to 1.92 (full support for PL/make)
- Reordered report sections to move program output to immediately
after tester comments; this keeps most relevant information at the
top and prereqs and other context follow
- Added perl version to the report intro paragraph since otherwise it
doesn't appear until the "perl -V" output at the end of the report
- Compressed UID/EUID variables to a single line in the report
- Made report language less 'test' specific, since we can also report
on PL or make failures
- Changed the way perl version is stringified for history file entries to
workaround an old bleadperl bug; (a "better safe than sorry" fix)
(Reported by Andreas Koenig)
- Cleans up temp directories in testing more frequently
0.99_11 Fri Sep 21 17:02:18 EDT 2007
- Fallback to "make test" or "Build test" exit value when output parsing
fails. This supports custom testing in Makefile.PL or Build.PL
(Reported on Prima by Dmitry Karasik)
0.99_10 Tue Sep 11 17:29:18 EDT 2007
- Added advanced options "send_PL_report", "send_make_report" and
"send_test_report" to override "send_report" for individual phases
(requested by David Cantrell)
0.99_09 Mon Sep 10 09:50:57 EDT 2007
- Fix test failure on bleadperl
0.99_08 Sat Sep 8 18:07:08 EDT 2007
* History file of sent reports has a new name ('reports-sent.db') and a
new format with phase of testing for report ('PL', 'make', 'test')
- More robust check for Test::Harness 2.99_01 (works even with out-of-date
releases of version.pm)
- More refactoring of config/history functions into separate modules
0.99_07 Wed Sep 5 18:20:37 EDT 2007
- Skips inactivity timeout on Win32 without Win32::Process >= 0.10
(bug found by Corion)
- Bails out on Test::Harness 2.99_01 (not supported)
0.99_06 Wed Sep 5 16:56:23 EDT 2007
- Added support for new message format from Test::Harness alpha (2.99_02)
- Changed grading for test files with no output from FAIL to UNKNOWN to
harmonize with new Test::Harness approach
- Falls back to overall 'make test' exit code for success or failure if
recursive Makefile.PL's are detected; File::Find added to prereqs;
Bumped EU::MM prereq to 6.36 to get bug fixes for recursion and more
- record_command() wrappers do better error checking for exec failures
- Bumped File::Copy::Recursive dependency to 0.35 (fixes Win32 bug)
- More defensive test coding
- changed to the Apache License, version 2.0; (it's clearer, relicensable,
and is explicit about contributions)
0.99_05 Mon Aug 27 16:16:50 EDT 2007
- Fixed copy/paste typo that caused testing to fail; (serves me right
for rushing out 0.99_04 without using my normal release-testing script)
0.99_04 Mon Aug 27 07:40:59 EDT 2007
- Merged changes from 0.4801
0.4801 Mon Aug 27 07:30:24 EDT 2007
- File::Copy::Recursive 0.34 is badly broken on Windows, causing our
tests to fail; added an "xcopy" workaround
0.99_03 Fri Aug 10 12:03:07 EDT 2007
- Fixed broken PL tests when Module::Build is not available
- Tweaked grade result message printed to screen
- Noted in Pod that CPAN 1.91_53 is required for full support
- Reordered Changes to be chronological
0.48 Tue Aug 7 16:28:42 EDT 2007
- Bumped Test::More prereq to 0.62 (from 0.99_02)
- More portable directory checks in test helper module
- Cleaned up DOS/unix line endings in many files
0.99_02 Tue Aug 7 06:54:59 EDT 2007
- Author tests (e.g. pod/pod-coverage) moved to 'author_t' directory
- Module::Build subclass used for development moved to inc
- Pulled forward maintenance changes from 0.47_01
- Bumped Test::More prereq to 0.62
0.47_01 Sat Aug 4 00:09:54 EDT 2007
- Added TEMP and TMPDIR to environment variables included in report
- Changed inclusion of AUTHOR_TESTING environment variable
to any matching /AUTHOR_TEST/
- Changed how "Build.PL" is executed in tests to diagnose odd Cygwin
failure; increased test diagnostics
0.99_01 Tue Jul 31 22:28:25 EDT 2007
* Major API additions and configuration changes in preparation for
adding Makefile/Build.PL and make/Build support into CPAN.pm
* Deprecated test() function; test() separated into record_command()
and grade_test() functions to support CPAN.pm sending
reports from output generated indpendently from CPAN::Reporter,
e.g. CPAN.pm using Expect with distroprefs
* Added record_command() to wrap and tee a system() command and
return results for further evaluation in grade_*() functions;
wrapping used to capture exit values that would otherwise be lost
from pipe to tee; will support CPAN inactivity timouts on both *nix
and Win32
* Added grade_PL() function to evaluate results of 'perl Makefile.PL'
or 'perl Build.PL'; recognizes 'require 5.xxxxx' and unsupported
OS error messages to be 'NA' as well as ordinary pass/fail outcomes
* Added grade_make() function to evaluate results of 'make' or