What's cooking (2012/09 #01)
Junio C Hamano committed
Sep 4, 2012 at 16:15 UTC
e911aec440901521aff0ddc6cb853e2c3a387cf7
1 file changed
+229
-170
whats-cooking.txt
+229
-170
@@ -1,16 +1,16 @@
1
To: git@vger.kernel.org
2
Bcc: lwn@lwn.net
3
-Subject: What's cooking in git.git (Aug 2012, #10; Fri, 31)
4
-X-master-at: 16d26b168b371b2f4f86b1adb61470c6b08b27b9
5
-X-next-at: d550ecdc42bd9a5e364767da2177640724535e93
3
+Subject: What's cooking in git.git (Sep 2012, #01; Tue, 4)
4
+X-master-at: 3fc0e4ca00330ce1dfd92baeda04f0aa2fbd7e68
5
+X-next-at: cfd4979c8f619202e04abe06b9cd742131d3422b
6
7
-What's cooking in git.git (Aug 2012, #10; Fri, 31)
7
+What's cooking in git.git (Sep 2012, #01; Tue, 4)
8
--------------------------------------------------
9
10
Here are the topics that have been cooking. Commits prefixed with '-' are
11
only in 'pu' (proposed updates) while commits prefixed with '+' are in 'next'.
12
13
-The latter half of the second batch of topics are now in 'master'.
13
+The third batch of topics have graduated to 'master'.
14
15
I'm planning to keep this cycle reasonably short and aim for tagging
16
the result as 1.8.0 at the end of 9th week, on October 21st, after
@@ -22,46 +22,180 @@ repositories listed at
22
23
http://git-blame.blogspot.com/p/git-public-repositories.html
24
25
+--------------------------------------------------
26
+[Graduated to "master"]
27
+
28
+* cw/cherry-pick-allow-empty-message (2012-08-06) 1 commit
29
+ (merged to 'next' on 2012-08-20 at 0dc0590)
30
+ + cherry-pick: add --allow-empty-message option
31
+
32
+(Originally merged to 'next' on 2012-08-09)
33
+
34
+"git cherry-pick" by default stops when it sees a commit without any
35
+log message. The "--allow-empty-message" option can be used to
36
+silently proceed.
37
+
38
+* jc/daemon-access-hook (2012-08-15) 1 commit
39
+ (merged to 'next' on 2012-08-20 at c8f5b2e)
40
+ + daemon: --access-hook option
41
+
42
+Allow an external command to tell git-daemon to decline service
43
+based on the client address, repository path, etc.
44
+
45
+* jc/maint-config-exit-status (2012-07-30) 1 commit
46
+ (merged to 'next' on 2012-08-20 at e18ddd8)
47
+ + config: "git config baa" should exit with status 1
48
+
49
+(Originally merged to 'next' on 2012-07-30)
50
+
51
+The exit status code from "git config" was way overspecified while
52
+being incorrect. Update the implementation to give the documented
53
+status for a case that was documented, and introduce a new code for
54
+"all other errors".
55
+
56
+* jc/maint-mergetool-style-fix (2012-08-23) 1 commit
57
+ (merged to 'next' on 2012-08-28 at ced5868)
58
+ + mergetool: style fixes
59
+
60
+* jc/maint-sane-execvp-notdir (2012-07-31) 1 commit
61
+ (merged to 'next' on 2012-08-20 at 0d76cf2)
62
+ + sane_execvp(): ignore non-directory on $PATH
63
+
64
+(Originally merged to 'next' on 2012-08-09)
65
+
66
+"git foo" errored out with "Not a directory" when the user had a non
67
+directory on $PATH, and worse yet it masked an alias "foo" to run.
68
+
69
+* jc/send-email-reconfirm (2012-08-14) 1 commit
70
+ (merged to 'next' on 2012-08-20 at dffdef6)
71
+ + send-email: validate & reconfirm interactive responses
72
+
73
+Validate interactive input to "git send-email" to avoid common
74
+mistakes such as saying "y<RETURN>" to sender mail address whose
75
+prompt is given with a correctly guessed default.
76
+
77
+* js/compat-mkdir (2012-08-24) 1 commit
78
+ (merged to 'next' on 2012-08-28 at e92bf74)
79
+ + compat: some mkdir() do not like a slash at the end
80
+
81
+Some mkdir(2) implementations do not want to see trailing slash in
82
+its parameter.
83
+
84
+* js/no-curl-easy-strerror-on-old-curl (2012-08-23) 1 commit
85
+ (merged to 'next' on 2012-08-28 at 25e2e7a)
86
+ + http.c: don't use curl_easy_strerror prior to curl-7.12.0
87
+
88
+Done to support compilation on __TANDEM, but is independently useful
89
+for people with older version of libcURL.
90
+
91
+* js/use-sc-open-max (2012-08-24) 1 commit
92
+ (merged to 'next' on 2012-08-28 at 8b9ac7c)
93
+ + sha1_file.c: introduce get_max_fd_limit() helper
94
+
95
+Introduce get_max_fd_limit() to absorb platforms that do not have
96
+getrlimit(RLIMIT_NOFILE) and/or sysconf(_SC_OPEN_MAX).
97
+
98
+* ph/stash-rerere (2012-08-17) 2 commits
99
+ (merged to 'next' on 2012-08-20 at ae37111)
100
+ + stash: invoke rerere in case of conflict
101
+ + test: git-stash conflict sets up rerere
102
+
103
+"git stash" internally used "git merge-recursive" backend, which did
104
+not trigger "rerere" upon conflicts unlike other mergy operations.
105
+
106
+* sz/submodule-force-update (2012-08-24) 1 commit
107
+ (merged to 'next' on 2012-08-24 at f82b466)
108
+ + Make 'git submodule update --force' always check out submodules.
109
+
110
+"git submodule update --force" used to leave the working tree of the
111
+submodule intact when there were local changes. It is more intiutive
112
+to make "--force" a sign to run "checkout -f" to overwrite them.
113
+
114
--------------------------------------------------
115
[New Topics]
116
28
-* mz/cherry-pick-cmdline-order (2012-08-30) 3 commits
29
- (merged to 'next' on 2012-08-31 at fc8eec4)
30
- + cherry-pick/revert: respect order of revisions to pick
31
- + demonstrate broken 'git cherry-pick three one two'
32
- + teach log --no-walk=unsorted, which avoids sorting
117
+* as/check-ignore (2012-09-02) 10 commits
118
+ . fixup: decl-after-stmt etc.
119
+ . Add git-check-ignore
120
+ . Provide free_directory() for reclaiming dir_struct memory
121
+ . Extract some useful pathspec handling code from builtin/add.c into a library
122
+ . For each exclude pattern, store information about where it came from
123
+ . dir.c: refactor excluded() and path_excluded()
124
+ . dir.c: refactor excluded_from_list()
125
+ . dir.c: rename cryptic 'which' variable to more consistent name
126
+ . Improve documentation and comments regarding directory traversal API
127
+ . Update directory listing API doc to match code
128
34
-"git cherry-pick A C B" used to replay changes in A and then B and
35
-then C if these three commits had committer timestamps in that
36
-order, which is not what the user who said "A C B" naturally expects.
129
+Queued, but I think the design of the end-user experience needs to
130
+be discussed further to polish this topic.
131
38
-Will merge to 'master' as part of the fifth batch.
132
+* jc/maint-ident-missing-human-name (2012-08-31) 1 commit
133
+ - split_ident_line(): make best effort when parsing author/committer line
134
40
-* ph/credential-gnome-keyring (2012-08-24) 1 commit
41
- (merged to 'next' on 2012-08-31 at 6f3b1de)
42
- + contrib: add credential helper for GnomeKeyring
43
- (this branch is used by ph/credential-refactor.)
135
+"git show --format='%ci'" did not give timestamp correctly for
136
+commits created without human readable name on "committer" line.
137
45
-The later refactoring of the shared code in the original series may
46
-not be worth the trouble, so it is split into a separate topic that
47
-builds on top of this one, which independently should be useful.
138
+Will merge to 'next' after eyeballing one more time.
139
49
-Will merge to 'master' as part of the fifth batch.
140
+* jk/argv-array (2012-09-02) 4 commits
141
+ - submodule: use argv_array instead of hand-building arrays
142
+ - fetch: use argv_array instead of hand-building arrays
143
+ - argv-array: fix bogus cast when freeing array
144
+ - argv-array: add pop function
145
51
-* ms/contrib-thunderbird-updates (2012-08-31) 2 commits
52
- - [SQUASH] minimum fixup
53
- - Thunderbird: fix appp.sh format problems
146
+Use argv-array API in "git fetch" implementation.
147
+Will merge to 'next' after eyeballing one more time.
148
55
-Update helper to send out format-patch output using Thunderbird.
149
+* rj/tap-fix (2012-09-02) 6 commits
150
+ - test-lib.sh: Suppress the "passed all ..." message if no tests run
151
+ - test-lib.sh: Add check for invalid use of 'skip_all' facility
152
+ - test-lib.sh: Fix some shell coding style violations
153
+ - t4016-*.sh: Skip all tests rather than each test
154
+ - t3902-*.sh: Skip all tests rather than each test
155
+ - t3300-*.sh: Fix a TAP parse error
156
57
-* nd/i18n-index-pack (2012-08-31) 1 commit
58
- - i18n: mark more index-pack strings for translation
157
+Will merge to 'next' after eyeballing one more time.
158
60
-Will merge to 'next'.
159
+* rj/test-regex (2012-09-02) 1 commit
160
+ - test-regex: Add a test to check for a bug in the regex routines
161
+
162
+Git ships with a fall-back regexp implementation for platforms with
163
+buggy regexp library; give people a tool to see if they should be
164
+using it on their platform.
165
+
166
+Will merge to 'next' after eyeballing one more time.
167
+
168
+* jc/maint-checkout-fileglob-doc (2012-09-04) 1 commit
169
+ - Document file-glob for "git checkout -- '*.c'"
170
+
171
+Will merge to 'next' after eyeballing one more time.
172
+
173
+* jc/xprm-generation (2012-09-04) 1 commit
174
+ - test-generation: compute generation numbers and clock skews
175
+
176
+* rj/path-cleanup (2012-09-04) 5 commits
177
+ - Call mkpathdup() rather than xstrdup(mkpath(...))
178
+ - Call git_pathdup() rather than xstrdup(git_path("..."))
179
+ - path.c: Use vsnpath() in the implementation of git_path()
180
+ - path.c: Don't discard the return value of vsnpath()
181
+ - path.c: Remove the 'git_' prefix from a file scope function
182
+
183
+* rs/archive-zip-utf8 (2012-09-04) 1 commit
184
+ - archive-zip: support UTF-8 paths
185
186
--------------------------------------------------
187
[Stalled]
188
189
+* jx/test-real-path (2012-08-27) 1 commit
190
+ - test: set the realpath of CWD as TRASH_DIRECTORY
191
+
192
+Running tests with the "trash" directory elsewhere with the "--root"
193
+option did not work well if the directory was specified by a symbolic
194
+link pointing at it.
195
+
196
+Seems broken as it makes $(pwd) and TRASH_DIRECTORY inconsistent.
197
+Needs rerolling.
198
+
199
* jc/maint-push-refs-all (2012-08-27) 2 commits
200
- get_fetch_map(): tighten checks on dest refs
201
- [BROKEN] fetch/push: allow refs/*:refs/*
@@ -93,6 +227,25 @@ somewhat more maintainable form, though it breaks existing i18n.
227
228
Anybody interested in fixing it up? Otherwise may discard.
229
230
+* tg/index-v5 (2012-08-17) 13 commits
231
+ . p0002-index.sh: add perf test for the index formats
232
+ . update-index.c: rewrite index when index-version is given
233
+ . Write resolve-undo data for index-v5
234
+ . Write index-v5 cache-tree data
235
+ . Write index-v5
236
+ . Read cache-tree in index-v5
237
+ . Read resolve-undo data
238
+ . Read index-v5
239
+ . Make in-memory format aware of stat_crc
240
+ . Add documentation of the index-v5 file format
241
+ . t2104: Don't fail for index versions other than [23]
242
+ . read-cache.c: Re-read index if index file changed
243
+ . Move index v2 specific functions to their own file
244
+
245
+A GSoC project. Was waiting for comments from mentors and
246
+stakeholders, but nothing seems to be happening, other than breakage
247
+fixes on Cygwin. May discard.
248
+
249
* pw/p4-submit-conflicts (2012-08-16) 12 commits
250
- git p4: add submit --prepare-p4-only option
251
- git p4: add submit --dry-run option
@@ -173,6 +326,41 @@ working :-<.
326
--------------------------------------------------
327
[Cooking]
328
329
+* mz/cherry-pick-cmdline-order (2012-08-30) 3 commits
330
+ (merged to 'next' on 2012-08-31 at fc8eec4)
331
+ + cherry-pick/revert: respect order of revisions to pick
332
+ + demonstrate broken 'git cherry-pick three one two'
333
+ + teach log --no-walk=unsorted, which avoids sorting
334
+
335
+"git cherry-pick A C B" used to replay changes in A and then B and
336
+then C if these three commits had committer timestamps in that
337
+order, which is not what the user who said "A C B" naturally expects.
338
+
339
+Will merge to 'master' as part of the fifth batch.
340
+
341
+* ph/credential-gnome-keyring (2012-08-24) 1 commit
342
+ (merged to 'next' on 2012-08-31 at 6f3b1de)
343
+ + contrib: add credential helper for GnomeKeyring
344
+ (this branch is used by ph/credential-refactor.)
345
+
346
+The later refactoring of the shared code in the original series may
347
+not be worth the trouble, so it is split into a separate topic that
348
+builds on top of this one, which independently should be useful.
349
+
350
+Will merge to 'master' as part of the fifth batch.
351
+
352
+* ms/contrib-thunderbird-updates (2012-08-31) 2 commits
353
+ - [SQUASH] minimum fixup
354
+ - Thunderbird: fix appp.sh format problems
355
+
356
+Update helper to send out format-patch output using Thunderbird.
357
+Seems to have design regression for silent users.
358
+
359
+* nd/i18n-index-pack (2012-08-31) 1 commit
360
+ - i18n: mark more index-pack strings for translation
361
+
362
+Will merge to 'next'.
363
+
364
* jk/maint-quiet-is-synonym-to-s-in-log (2012-08-28) 1 commit
365
(merged to 'next' on 2012-08-31 at 06f6953)
366
+ log: fix --quiet synonym for -s
@@ -195,10 +383,13 @@ Reorganize the code and allow giving a proper diagnosis when the
383
user says "git checkout -b -t foo bar" (e.g. "-t" is not a good name
384
for a branch).
385
198
-* jc/maint-name-rev (2012-08-29) 6 commits
386
+Will merge to 'next' after eyeballing one more time.
387
+
388
+* jc/maint-name-rev (2012-09-04) 7 commits
389
- describe --contains: use "name-rev --weight"
390
- name-rev --weight: tests and documentation
391
- name-rev --weight: cache the computed weight in notes
392
+ - name-rev --weight: trivial optimization
393
- name-rev: --weight option
394
- name_rev: clarify the logic to assign a new tip-name to a commit
395
- name-rev: lose unnecessary typedef
@@ -215,6 +406,10 @@ slower. Also the command will be unusably slow for a user who does
406
not have a write access (hence unable to create or update the
407
notes-cache).
408
409
+Needs another round to at least find a better name for the option,
410
+and possibly a cheaper but still better than the current "close to
411
+the tip" heuristics.
412
+
413
* cn/branch-set-upstream-to (2012-08-30) 3 commits
414
(merged to 'next' on 2012-08-31 at d550ecd)
415
+ branch: deprecate --set-upstream and show help if we detect possible mistaken use
@@ -240,46 +435,13 @@ command line parser to interpret ".." as a path in such a case.
435
436
Will merge to 'master' as part of the fourth batch.
437
243
-* jc/maint-mergetool-style-fix (2012-08-23) 1 commit
244
- (merged to 'next' on 2012-08-28 at ced5868)
245
- + mergetool: style fixes
246
-
247
-Will merge to 'master' as part of the third batch.
248
-
438
* js/compat-itimer (2012-08-24) 1 commit
439
- Support for setitimer() on platforms lacking it
440
441
Pieces to support compilation on __TANDEM.
442
254
-* js/compat-mkdir (2012-08-24) 1 commit
255
- (merged to 'next' on 2012-08-28 at e92bf74)
256
- + compat: some mkdir() do not like a slash at the end
257
-
258
-Some mkdir(2) implementations do not want to see trailing slash in
259
-its parameter.
260
-
261
-Will merge to 'master' as part of the third batch.
262
-
263
-* js/no-curl-easy-strerror-on-old-curl (2012-08-23) 1 commit
264
- (merged to 'next' on 2012-08-28 at 25e2e7a)
265
- + http.c: don't use curl_easy_strerror prior to curl-7.12.0
266
-
267
-Done to support compilation on __TANDEM, but is independently useful
268
-for people with older version of libcURL.
269
-
270
-Will merge to 'master' as part of the third batch.
271
-
272
-* js/use-sc-open-max (2012-08-24) 1 commit
273
- (merged to 'next' on 2012-08-28 at 8b9ac7c)
274
- + sha1_file.c: introduce get_max_fd_limit() helper
275
-
276
-Introduce get_max_fd_limit() to absorb platforms that do not have
277
-getrlimit(RLIMIT_NOFILE) and/or sysconf(_SC_OPEN_MAX).
278
-
279
-Will merge to 'master' as part of the third batch.
280
-
281
-* ph/credential-refactor (2012-08-27) 5 commits
282
- - wincred: port to generic credential helper (UNTESTED)
443
+* ph/credential-refactor (2012-09-02) 5 commits
444
+ - wincred: port to generic credential helper
445
- Merge branch 'ef/win32-cred-helper' into ph/credential-refactor
446
- osxkeychain: port to generic credential helper implementation
447
- gnome-keyring: port to generic helper implementation
@@ -333,16 +495,6 @@ for POST.
495
496
Will merge to 'master' as part of the fourth batch.
497
336
-* jx/test-real-path (2012-08-27) 1 commit
337
- - test: set the realpath of CWD as TRASH_DIRECTORY
338
-
339
-Running tests with the "trash" directory elsewhere with the "--root"
340
-option did not work well if the directory was specified by a symbolic
341
-link pointing at it.
342
-
343
-Seems broken as it makes $(pwd) and TRASH_DIRECTORY inconsistent.
344
-Needs rerolling.
345
-
498
* mh/fetch-filter-refs (2012-08-26) 17 commits
499
- filter_refs(): simplify logic
500
- fetch_pack(): free matching heads
@@ -451,7 +603,7 @@ Will merge to 'master' as part of the fourth batch.
603
- Implement a remote helper for svn in C
604
(this branch is used by fa/vcs-svn.)
605
454
-A GSoC project.
606
+A GSoC project. Looked promising.
607
Waiting for comments from mentors and stakeholders.
608
609
* fa/vcs-svn (2012-08-28) 4 commits
@@ -461,7 +613,7 @@ Waiting for comments from mentors and stakeholders.
613
- svndump: move struct definitions to .h
614
(this branch uses fa/remote-svn.)
615
464
-A GSoC project.
616
+A GSoC project. Looked promising.
617
Waiting for comments from mentors and stakeholders.
618
619
* nd/i18n-parseopt-help (2012-08-22) 66 commits
@@ -574,35 +726,6 @@ Queued with trivial fix-ups.
726
727
Will merge to 'master' as part of the third batch.
728
577
-* jc/daemon-access-hook (2012-08-15) 1 commit
578
- (merged to 'next' on 2012-08-20 at c8f5b2e)
579
- + daemon: --access-hook option
580
-
581
-Allow an external command to tell git-daemon to decline service
582
-based on the client address, repository path, etc.
583
-
584
-Will merge to 'master' as part of the third batch.
585
-
586
-* jc/send-email-reconfirm (2012-08-14) 1 commit
587
- (merged to 'next' on 2012-08-20 at dffdef6)
588
- + send-email: validate & reconfirm interactive responses
589
-
590
-Validate interactive input to "git send-email" to avoid common
591
-mistakes such as saying "y<RETURN>" to sender mail address whose
592
-prompt is given with a correctly guessed default.
593
-
594
-Will merge to 'master' as part of the third batch.
595
-
596
-* ph/stash-rerere (2012-08-17) 2 commits
597
- (merged to 'next' on 2012-08-20 at ae37111)
598
- + stash: invoke rerere in case of conflict
599
- + test: git-stash conflict sets up rerere
600
-
601
-"git stash" internally used "git merge-recursive" backend, which did
602
-not trigger "rerere" upon conflicts unlike other mergy operations.
603
-
604
-Will merge to 'master' as part of the third batch.
605
-
729
* pw/p4-use-client-spec-branch-detection (2012-08-20) 5 commits
730
(merged to 'next' on 2012-08-24 at 08abc61)
731
+ git p4: make branch detection work with --use-client-spec
@@ -616,70 +739,6 @@ used together (the command used to misdetect branches).
739
740
Will merge to 'master' as part of the fourth batch.
741
619
-* tg/index-v5 (2012-08-17) 13 commits
620
- - p0002-index.sh: add perf test for the index formats
621
- - update-index.c: rewrite index when index-version is given
622
- - Write resolve-undo data for index-v5
623
- - Write index-v5 cache-tree data
624
- - Write index-v5
625
- - Read cache-tree in index-v5
626
- - Read resolve-undo data
627
- - Read index-v5
628
- - Make in-memory format aware of stat_crc
629
- - Add documentation of the index-v5 file format
630
- - t2104: Don't fail for index versions other than [23]
631
- - read-cache.c: Re-read index if index file changed
632
- - Move index v2 specific functions to their own file
633
-
634
-A GSoC project.
635
-Waiting for comments from mentors and stakeholders.
636
-
637
-* jc/maint-sane-execvp-notdir (2012-07-31) 1 commit
638
- (merged to 'next' on 2012-08-20 at 0d76cf2)
639
- + sane_execvp(): ignore non-directory on $PATH
640
-
641
-(Originally merged to 'next' on 2012-08-09)
642
-
643
-"git foo" errored out with "Not a directory" when the user had a non
644
-directory on $PATH, and worse yet it masked an alias "foo" to run.
645
-
646
-Will merge to 'master' as part of the third batch.
647
-
648
-* cw/cherry-pick-allow-empty-message (2012-08-06) 1 commit
649
- (merged to 'next' on 2012-08-20 at 0dc0590)
650
- + cherry-pick: add --allow-empty-message option
651
-
652
-(Originally merged to 'next' on 2012-08-09)
653
-
654
-"git cherry-pick" by default stops when it sees a commit without any
655
-log message. The "--allow-empty-message" option can be used to
656
-silently proceed.
657
-
658
-Will merge to 'master' as part of the third batch.
659
-
660
-* jc/maint-config-exit-status (2012-07-30) 1 commit
661
- (merged to 'next' on 2012-08-20 at e18ddd8)
662
- + config: "git config baa" should exit with status 1
663
-
664
-(Originally merged to 'next' on 2012-07-30)
665
-
666
-The exit status code from "git config" was way overspecified while
667
-being incorrect. Update the implementation to give the documented
668
-status for a case that was documented, and introduce a new code for
669
-"all other errors".
670
-
671
-Will merge to 'master' as part of the third batch.
672
-
673
-* sz/submodule-force-update (2012-08-24) 1 commit
674
- (merged to 'next' on 2012-08-24 at f82b466)
675
- + Make 'git submodule update --force' always check out submodules.
676
-
677
-"git submodule update --force" used to leave the working tree of the
678
-submodule intact when there were local changes. It is more intiutive
679
-to make "--force" a sign to run "checkout -f" to overwrite them.
680
-
681
-Will merge to 'master' as part of the third batch.
682
-
742
* jk/no-more-pre-exec-callback (2012-06-05) 1 commit
743
- pager: drop "wait for output to run less" hack
744