What's cooking (2016/01 #03)
Junio C Hamano committed
Jan 13, 2016 at 14:21 UTC
f38ba4a335f098711f1ddc4adfdfb1b313722804
1 file changed
+344
-271
whats-cooking.txt
+344
-271
@@ -1,10 +1,10 @@
1
To: git@vger.kernel.org
2
Bcc: lwn@lwn.net
3
-Subject: What's cooking in git.git (Jan 2016, #02; Mon, 11)
4
-X-master-at: 754884255bb580df159e58defa81cdd30b5c430c
5
-X-next-at: ef7b32d4f2d070ec6e688815cb7d6b6c2bc843fb
3
+Subject: What's cooking in git.git (Jan 2016, #03; Wed, 13)
4
+X-master-at: fc10eb5b87a75af2cb93a3932897572f2c544915
5
+X-next-at: dc08a1951c8b3cd3e311ff0c2612db06ab28ae13
6
7
-What's cooking in git.git (Jan 2016, #02; Mon, 11)
7
+What's cooking in git.git (Jan 2016, #03; Wed, 13)
8
--------------------------------------------------
9
10
Here are the topics that have been cooking. Commits prefixed with
@@ -12,17 +12,9 @@ Here are the topics that have been cooking. Commits prefixed with
12
'+' are in 'next'. The ones marked with '.' do not appear in any of
13
the integration branches, but I am still holding onto them.
14
15
-It's been a week or so since Git 2.7 was released. I'll draw the
16
-projected timeline for the next cycle (which I expect to be a
17
-shorter and lightweight cycle), reorder 'next' and possibly eject a
18
-few topics from it to give them a fresh slate to rebuild on, and
19
-then start the 2.7.x maintenance track soonish, but none of these
20
-has happened yet; I'm feeling behind X-<.
21
-
22
-Big thanks to Stephen Smith for picking up a few topics from the
23
-Stalled pile and reviving them. And as always, thanks for the
24
-regular reviewers (you know who you are) for helping the topics
25
-move forward.
15
+The tip of 'next' has been rewound. A handful of topics have
16
+graduated to 'master'. A few topics that are fixes to 2.7.0 started
17
+cooking in 'next'.
18
19
You can find the changes described here in the integration branches of the
20
repositories listed at
@@ -30,77 +22,179 @@ repositories listed at
22
http://git-blame.blogspot.com/p/git-public-repositories.html
23
24
--------------------------------------------------
33
-[New Topics]
25
+[Graduated to "master"]
26
35
-* dw/signoff-doc (2016-01-05) 1 commit
36
- - Expand documentation describing --signoff
27
+* ea/blame-progress (2015-12-16) 1 commit
28
+ (merged to 'next' on 2015-12-22 at f8e8643)
29
+ + blame: add support for --[no-]progress option
30
38
- The documentation has been updated to hint the connection between
39
- the '--signoff' option and DCO.
31
+ Originally merged to 'next' on 2015-12-22
32
41
- Will merge to 'next'.
33
+ "git blame" learned to produce the progress eye-candy when it takes
34
+ too much time before emitting the first line of the result.
35
36
44
-* ew/for-each-ref-doc (2016-01-05) 1 commit
45
- - for-each-ref: document `creatordate` and `creator` fields
37
+* ep/make-phoney (2015-12-16) 1 commit
38
+ (merged to 'next' on 2015-12-22 at 27c7593)
39
+ + Makefile: add missing phony target
40
47
- Will merge to 'next'.
41
+ Originally merged to 'next' on 2015-12-22
42
43
+ A slight update to the Makefile.
44
50
-* sg/t6050-failing-editor-test-fix (2016-01-05) 1 commit
51
- - t6050-replace: make failing editor test more robust
45
53
- Will merge to 'next'.
46
+* nd/stop-setenv-work-tree (2015-12-22) 1 commit
47
+ (merged to 'next' on 2015-12-22 at 6d7bb0c)
48
+ + Revert "setup: set env $GIT_WORK_TREE when work tree is set, like $GIT_DIR"
49
+ (this branch is used by nd/clear-gitenv-upon-use-of-alias.)
50
51
+ Originally merged to 'next' on 2015-12-22
52
56
-* js/dirname-basename (2016-01-11) 4 commits
57
- - t0060: verify that basename() and dirname() work as expected
58
- - compat/basename.c: provide a dirname() compatibility function
59
- - compat/basename: make basename() conform to POSIX
60
- - Refactor skipping DOS drive prefixes
53
+ An earlier change in 2.5.x-era broke users' hooks and aliases by
54
+ exporting GIT_WORK_TREE to point at the root of the working tree,
55
+ interfering when they tried to use a different working tree without
56
+ setting GIT_WORK_TREE environment themselves.
57
62
- dirname() emulation has been added, as Msys2 lacks it.
58
64
- Will merge to 'next', 'master', and then to 'maint'.
59
+* ps/push-delete-option (2015-12-16) 2 commits
60
+ (merged to 'next' on 2015-12-22 at d83cc1d)
61
+ + push: add '-d' as shorthand for '--delete'
62
+ + push: add '--delete' flag to synopsis
63
66
- The third-patch has an optional suggestion to make it easier to
67
- follow; I'm slightly in favor but lack of a reroll for it is not a
68
- showstopper, either.
64
+ Originally merged to 'next' on 2015-12-22
65
66
+ "branch --delete" has "branch -d" but "push --delete" does not.
67
71
-* js/fopen-harder (2016-01-11) 2 commits
72
- - Handle more file writes correctly in shared repos
73
- - commit: allow editing the commit message even in shared repos
68
75
- Some codepaths used fopen(3) when opening a fixed path in $GIT_DIR
76
- (e.g. COMMIT_EDITMSG) that is meant to be left after the command is
77
- done. This however did not work well if the repository is set to
78
- be shared with core.sharedRepository and the umask of the previous
79
- user is tighter. Make them work better by calling unlink(2) and
80
- retrying after fopen(3) fails with EPERM.
69
+* sb/submodule-parallel-fetch (2015-12-16) 7 commits
70
+ (merged to 'next' on 2015-12-22 at 44e84ff)
71
+ + submodules: allow parallel fetching, add tests and documentation
72
+ + fetch_populated_submodules: use new parallel job processing
73
+ + run-command: add an asynchronous parallel child processor
74
+ + sigchain: add command to pop all common signals
75
+ + strbuf: add strbuf_read_once to read without blocking
76
+ + xread: poll on non blocking fds
77
+ + submodule.c: write "Fetching submodule <foo>" to stderr
78
+ (this branch is used by sb/submodule-parallel-update.)
79
+
80
+ Originally merged to 'next' on 2015-12-22
81
+
82
+ Add a framework to spawn a group of processes in parallel, and use
83
+ it to run "git fetch --recurse-submodules" in parallel.
84
+
85
+ Rerolled and this seems to be a lot cleaner. The merge of the
86
+ earlier one to 'next' has been reverted.
87
+
88
+
89
+* vl/grep-configurable-threads (2015-12-16) 3 commits
90
+ (merged to 'next' on 2015-12-22 at 8954705)
91
+ + grep: add --threads=<num> option and grep.threads configuration
92
+ + grep: slight refactoring to the code that disables threading
93
+ + grep: allow threading even on a single-core machine
94
+
95
+ Originally merged to 'next' on 2015-12-22
96
+
97
+ "git grep" can now be configured (or told from the command line)
98
+ how many threads to use when searching in the working tree files.
99
+
100
+--------------------------------------------------
101
+[New Topics]
102
+
103
+* ho/gitweb-squelch-undef-warning (2016-01-12) 1 commit
104
+ (merged to 'next' on 2016-01-12 at ef4fc5f)
105
+ + gitweb: squelch "uninitialized value" warning
106
+
107
+ Asking gitweb for a nonexistent commit left a warning in the server
108
+ log.
109
+
110
+ Somebody may want to follow this up with a new test, perhaps?
111
+ IIRC, we do test that no Perl warnings are given to the server log,
112
+ so this should have been caught if our test coverage were good.
113
+
114
+ Will merge to 'master'.
115
+
116
+
117
+* tg/grep-no-index-fallback (2016-01-12) 2 commits
118
+ (merged to 'next' on 2016-01-12 at 8960bdd)
119
+ + builtin/grep: add grep.fallbackToNoIndex config
120
+ + t7810: correct --no-index test
121
+
122
+ "git grep" by default does not fall back to its --no-index
123
+ behaviour outside a directory under Git's control (otherwise the
124
+ user may by mistake end up running a huge recursive search); with a
125
+ new configuration (set in $HOME/.gitconfig--by definition this
126
+ cannot be set in the config file per project), this safety can be
127
+ disabled.
128
+
129
+ Will merge to 'master'.
130
+
131
+
132
+* js/pull-rebase-i (2016-01-13) 3 commits
133
+ - completion: add missing branch.*.rebase values
134
+ - remote: handle the config setting branch.*.rebase=interactive
135
+ - pull: allow interactive rebase with --rebase=interactive
136
+
137
+ "git pull --rebase" has been extended to allow invoking
138
+ "rebase -i".
139
140
Will merge to 'next'.
141
142
85
-* nd/exclusion-regression-fix (2016-01-08) 1 commit
86
- - Revert "dir.c: don't exclude whole dir prematurely if neg pattern may match"
143
+* jk/ok-to-fail-gc-auto-in-rebase (2016-01-13) 1 commit
144
+ - rebase: ignore failures from "gc --auto"
145
88
- The ignore mechanism saw a few regressions around untracked file
89
- listing and sparse checkout selection areas in 2.7.0; the change
90
- that is responsible for the regression has been reverted.
146
+ "git rebase", unlike all other callers of "gc --auto", did not
147
+ ignore the exit code from "gc --auto".
148
+
149
+ Will merge to 'next'.
150
92
- Will merge to 'next' and then to 'master'.
151
152
+* js/close-packs-before-gc (2016-01-13) 4 commits
153
+ - receive-pack: release pack files before garbage-collecting
154
+ - merge: release pack files before garbage-collecting
155
+ - am: release pack files before garbage-collecting
156
+ - fetch: release pack files before garbage-collecting
157
95
-* ss/clone-depth-single-doc (2016-01-08) 3 commits
96
- - docs: clarify that --depth for git-fetch works with newly initialized repos
97
- - docs: say "commits" in the --depth option wording for git-clone
98
- - docs: clarify that passing --depth to git-clone implies --single-branch
158
+ Many codepaths that run "gc --auto" before exiting kept packfiles
159
+ mapped and left the file descriptors to them open, which was not
160
+ friendly to systems that cannot remove files that are open. They
161
+ now close the packs before doing so.
162
100
- Updates documents to clarify "git fetch --depth".
163
+ Will merge to 'next'.
164
+
165
+
166
+* js/msys2 (2016-01-13) 5 commits
167
+ - compat/winansi: support compiling with MSys2
168
+ - compat/mingw: support MSys2-based MinGW build
169
+ - nedmalloc: allow compiling with MSys2's compiler
170
+ - config.mak.uname: supporting 64-bit MSys2
171
+ - config.mak.uname: support MSys2
172
+
173
+ Beginning of the upstreaming process of Git for Windows effort.
174
+
175
+ Waiting for review.
176
+ ($gmane/283975).
177
+
178
+
179
+* rp/p4-filetype-change (2016-01-13) 1 commit
180
+ - git-p4.py: add support for filetype change
181
182
Will merge to 'next'.
183
184
+
185
+* tk/interpret-trailers-in-place (2016-01-13) 2 commits
186
+ - interpret-trailers: add option for in-place editing
187
+ - trailer: use fprintf instead of printf
188
+
189
+ "interpret-trailers" has been taught to optionally update a file in
190
+ place, instead of always writing the result to the standard output.
191
+
192
+ Test may need to be updated to ensure a failing rewrite does not
193
+ clobber the original input.
194
+
195
+ Waiting for review.
196
+ ($gmane/283959).
197
+
198
--------------------------------------------------
199
[Stalled]
200
@@ -152,22 +246,104 @@ repositories listed at
246
--------------------------------------------------
247
[Cooking]
248
155
-* wp/sha1-name-negative-match (2016-01-11) 2 commits
249
+* dw/signoff-doc (2016-01-05) 1 commit
250
+ (merged to 'next' on 2016-01-12 at 1b08b48)
251
+ + Expand documentation describing --signoff
252
+
253
+ The documentation has been updated to hint the connection between
254
+ the '--signoff' option and DCO.
255
+
256
+ Will merge to 'master'.
257
+
258
+
259
+* ew/for-each-ref-doc (2016-01-05) 1 commit
260
+ (merged to 'next' on 2016-01-12 at e5c4e75)
261
+ + for-each-ref: document `creatordate` and `creator` fields
262
+
263
+ Will merge to 'master'.
264
+
265
+
266
+* sg/t6050-failing-editor-test-fix (2016-01-05) 1 commit
267
+ (merged to 'next' on 2016-01-12 at dc08a19)
268
+ + t6050-replace: make failing editor test more robust
269
+
270
+ Will merge to 'master'.
271
+
272
+
273
+* js/dirname-basename (2016-01-12) 4 commits
274
+ (merged to 'next' on 2016-01-12 at c3c970a)
275
+ + t0060: verify that basename() and dirname() work as expected
276
+ + compat/basename.c: provide a dirname() compatibility function
277
+ + compat/basename: make basename() conform to POSIX
278
+ + Refactor skipping DOS drive prefixes
279
+
280
+ dirname() emulation has been added, as Msys2 lacks it.
281
+
282
+ The test needs to be loosened to allow implementation defined
283
+ behaviour; currently it fails on Macs.
284
+
285
+ Waiting for an update.
286
+ ($gmane/283968).
287
+
288
+
289
+* js/fopen-harder (2016-01-11) 2 commits
290
+ (merged to 'next' on 2016-01-12 at c6ef194)
291
+ + Handle more file writes correctly in shared repos
292
+ + commit: allow editing the commit message even in shared repos
293
+
294
+ Some codepaths used fopen(3) when opening a fixed path in $GIT_DIR
295
+ (e.g. COMMIT_EDITMSG) that is meant to be left after the command is
296
+ done. This however did not work well if the repository is set to
297
+ be shared with core.sharedRepository and the umask of the previous
298
+ user is tighter. Make them work better by calling unlink(2) and
299
+ retrying after fopen(3) fails with EPERM.
300
+
301
+ Will merge to 'master'.
302
+
303
+
304
+* nd/exclusion-regression-fix (2016-01-08) 1 commit
305
+ (merged to 'next' on 2016-01-12 at 0eb98a5)
306
+ + Revert "dir.c: don't exclude whole dir prematurely if neg pattern may match"
307
+
308
+ The ignore mechanism saw a few regressions around untracked file
309
+ listing and sparse checkout selection areas in 2.7.0; the change
310
+ that is responsible for the regression has been reverted.
311
+
312
+ Will merge to 'master'.
313
+
314
+
315
+* ss/clone-depth-single-doc (2016-01-08) 3 commits
316
+ (merged to 'next' on 2016-01-12 at 16ded8c)
317
+ + docs: clarify that --depth for git-fetch works with newly initialized repos
318
+ + docs: say "commits" in the --depth option wording for git-clone
319
+ + docs: clarify that passing --depth to git-clone implies --single-branch
320
+
321
+ Updates documents to clarify "git fetch --depth".
322
+
323
+ Will merge to 'master'.
324
+
325
+
326
+* wp/sha1-name-negative-match (2016-01-13) 2 commits
327
- object name: introduce '^{/!-<negative pattern>}' notation
328
- test for '!' handling in rev-parse's named commits
329
330
Introduce "<branch>^{/!-<pattern>}" notation to name a commit
331
reachable from <branch> that does not match the given <pattern>.
332
333
+ A questionable corner case where commit has no message remains.
334
+
335
+ Waiting for review.
336
+ ($gmane/283971)
337
+
338
163
-* ak/format-patch-odir-config (2016-01-11) 1 commit
164
- . format-patch: introduce format.outputDirectory configuration
339
+* ak/format-patch-odir-config (2016-01-13) 1 commit
340
+ - format-patch: introduce format.outputDirectory configuration
341
342
Allow "-o <dir>" option to be omitted on the command line of "git
343
format-patch" if you always use the same directory in your
344
workflow.
345
170
- Left out of 'pu' for now as t4014 seems to break with this topic.
346
+ Will merge to 'next'.
347
348
349
* jk/notes-merge-from-anywhere (2015-12-29) 1 commit
@@ -183,40 +359,33 @@ repositories listed at
359
360
361
* dk/reflog-walk-with-non-commit (2016-01-05) 1 commit
186
- - reflog-walk: don't segfault on non-commit sha1's in the reflog
362
+ (merged to 'next' on 2016-01-12 at 5f7b10e)
363
+ + reflog-walk: don't segfault on non-commit sha1's in the reflog
364
365
"git reflog" incorrectly assumed that all objects that used to be
366
at the tip of a ref must be commits, which caused it to segfault.
367
191
- Will merge to 'next'.
368
+ Will merge to 'master'.
369
370
371
* ew/send-email-mutt-alias-fix (2016-01-04) 1 commit
195
- - git-send-email: do not double-escape quotes from mutt
372
+ (merged to 'next' on 2016-01-12 at 84d1329)
373
+ + git-send-email: do not double-escape quotes from mutt
374
375
"git send-email" was confused by escaped quotes stored in the alias
376
files saved by "mutt".
377
200
- Will merge to 'next'.
378
+ Will merge to 'master'.
379
380
381
* jk/clang-pedantic (2016-01-04) 2 commits
204
- - bswap: add NO_UNALIGNED_LOADS define
205
- - avoid shifting signed integers 31 bits
382
+ (merged to 'next' on 2016-01-12 at b5be271)
383
+ + bswap: add NO_UNALIGNED_LOADS define
384
+ + avoid shifting signed integers 31 bits
385
386
A few unportable C construct have been spotted by clang compiler
387
and have been fixed.
388
210
- Will merge to 'next'.
211
-
212
-
213
-* ea/blame-progress (2015-12-16) 1 commit
214
- (merged to 'next' on 2015-12-22 at f8e8643)
215
- + blame: add support for --[no-]progress option
216
-
217
- "git blame" learned to produce the progress eye-candy when it takes
218
- too much time before emitting the first line of the result.
219
-
389
Will merge to 'master'.
390
391
@@ -227,14 +396,16 @@ repositories listed at
396
397
398
* jk/pack-revindex (2015-12-21) 2 commits
230
- - pack-revindex: store entries directly in packed_git
231
- - pack-revindex: drop hash table
399
+ (merged to 'next' on 2016-01-12 at 2e39a16)
400
+ + pack-revindex: store entries directly in packed_git
401
+ + pack-revindex: drop hash table
402
233
- Will merge to 'next'.
403
+ Will merge to 'master'.
404
405
236
-* jk/symbolic-ref (2016-01-11) 5 commits
406
+* jk/symbolic-ref (2016-01-13) 6 commits
407
- lock_ref_sha1_basic: handle REF_NODEREF with invalid refs
408
+ - lock_ref_sha1_basic: always fill old_oid while holding lock
409
- checkout,clone: check return value of create_symref
410
- create_symref: write reflog while holding lock
411
- create_symref: use existing ref-lock code
@@ -247,20 +418,29 @@ repositories listed at
418
Will merge to 'next'.
419
420
250
-* nd/stop-setenv-work-tree (2015-12-22) 1 commit
251
- (merged to 'next' on 2015-12-22 at 6d7bb0c)
252
- + Revert "setup: set env $GIT_WORK_TREE when work tree is set, like $GIT_DIR"
253
- (this branch is used by nd/clear-gitenv-upon-use-of-alias.)
254
-
255
- An earlier change in 2.5.x-era broke users' hooks and aliases by
256
- exporting GIT_WORK_TREE to point at the root of the working tree,
257
- interfering when they tried to use a different working tree without
258
- setting GIT_WORK_TREE environment themselves.
259
-
260
- Will merge to 'master'.
261
-
262
-
263
-* ep/update-command-substitution-style (2016-01-11) 70 commits
421
+* ep/shell-command-substitution-style (2016-01-12) 92 commits
422
+ - t9901-git-web--browse.sh: use the $( ... ) construct for command substitution
423
+ - t9501-gitweb-standalone-http-status.sh: use the $( ... ) construct for command substitution
424
+ - t9350-fast-export.sh: use the $( ... ) construct for command substitution
425
+ - t9300-fast-import.sh: use the $( ... ) construct for command substitution
426
+ - t9150-svk-mergetickets.sh: use the $( ... ) construct for command substitution
427
+ - t9145-git-svn-master-branch.sh: use the $( ... ) construct for command substitution
428
+ - t9138-git-svn-authors-prog.sh: use the $( ... ) construct for command substitution
429
+ - t9137-git-svn-dcommit-clobber-series.sh: use the $( ... ) construct for command substitution
430
+ - t9132-git-svn-broken-symlink.sh: use the $( ... ) construct for command substitution
431
+ - t9130-git-svn-authors-file.sh: use the $( ... ) construct for command substitution
432
+ - t9129-git-svn-i18n-commitencoding.sh: use the $( ... ) construct for command substitution
433
+ - t9119-git-svn-info.sh: use the $( ... ) construct for command substitution
434
+ - t9118-git-svn-funky-branch-names.sh: use the $( ... ) construct for command substitution
435
+ - t9114-git-svn-dcommit-merge.sh: use the $( ... ) construct for command substitution
436
+ - t9110-git-svn-use-svm-props.sh: use the $( ... ) construct for command substitution
437
+ - t9109-git-svn-multi-glob.sh: use the $( ... ) construct for command substitution
438
+ - t9108-git-svn-glob.sh: use the $( ... ) construct for command substitution
439
+ - t9107-git-svn-migrate.sh: use the $( ... ) construct for command substitution
440
+ - t9105-git-svn-commit-diff.sh: use the $( ... ) construct for command substitution
441
+ - t9104-git-svn-follow-parent.sh: use the $( ... ) construct for command substitution
442
+ - t9101-git-svn-props.sh: use the $( ... ) construct for command substitution
443
+ - t9100-git-svn-basic.sh: use the $( ... ) construct for command substitution
444
- t/t9001-send-email.sh: use the $( ... ) construct for command substitution
445
- t/t8003-blame-corner-cases.sh: use the $( ... ) construct for command substitution
446
- t/t7700-repack.sh: use the $( ... ) construct for command substitution
@@ -340,33 +520,36 @@ repositories listed at
520
521
522
* nd/dir-exclude-cleanup (2015-12-28) 1 commit
343
- - dir.c: clean the entire struct in clear_exclude_list()
523
+ (merged to 'next' on 2016-01-12 at e6584c9)
524
+ + dir.c: clean the entire struct in clear_exclude_list()
525
526
The "exclude_list" structure has the usual "alloc, nr" pair of
527
fields to be used by ALLOC_GROW(), but clear_exclude_list() forgot
528
to reset 'alloc' to 0 when it cleared 'nr' and discarded the
529
managed array.
530
350
- Will merge to 'next'.
531
+ Will merge to 'master'.
532
533
534
* ss/user-manual (2015-12-30) 4 commits
354
- - user-manual: add addition gitweb information
355
- - user-manual: add section documenting shallow clones
356
- - glossary: define the term shallow clone
357
- - user-manual: remove temporary branch entry from todo list
535
+ (merged to 'next' on 2016-01-12 at c7f0328)
536
+ + user-manual: add addition gitweb information
537
+ + user-manual: add section documenting shallow clones
538
+ + glossary: define the term shallow clone
539
+ + user-manual: remove temporary branch entry from todo list
540
541
Drop a few old "todo" items by deciding that the change one of them
542
suggests is not such a good idea, and doing the change the other
543
one suggested to do.
544
363
- Will merge to 'next'.
545
+ Will merge to 'master'.
546
547
548
* nd/ita-cleanup (2015-12-28) 3 commits
367
- - grep: make it clear i-t-a entries are ignored
368
- - add and use a convenience macro ce_intent_to_add()
369
- - blame: remove obsolete comment
549
+ (merged to 'next' on 2016-01-12 at 008a6e3)
550
+ + grep: make it clear i-t-a entries are ignored
551
+ + add and use a convenience macro ce_intent_to_add()
552
+ + blame: remove obsolete comment
553
554
Paths that have been told the index about with "add -N" are not
555
quite yet in the index, but a few commands behaved as if they
@@ -375,7 +558,7 @@ repositories listed at
558
Here are only the obviously correct bits; some other changes were
559
in the posted series, but not quite ready to be queued here.
560
378
- Will merge to 'next'.
561
+ Will merge to 'master'.
562
563
564
* cc/untracked (2015-12-29) 10 commits
@@ -397,59 +580,34 @@ repositories listed at
580
$gmane/283080
581
582
400
-* ep/make-phoney (2015-12-16) 1 commit
401
- (merged to 'next' on 2015-12-22 at 27c7593)
402
- + Makefile: add missing phony target
403
-
404
- A slight update to the Makefile.
405
-
406
- Will merge to 'master'.
407
-
408
-
409
-* vl/grep-configurable-threads (2015-12-16) 3 commits
410
- (merged to 'next' on 2015-12-22 at 8954705)
411
- + grep: add --threads=<num> option and grep.threads configuration
412
- + grep: slight refactoring to the code that disables threading
413
- + grep: allow threading even on a single-core machine
414
-
415
- "git grep" can now be configured (or told from the command line)
416
- how many threads to use when searching in the working tree files.
417
-
418
- Will merge to 'master'.
419
-
420
-
421
-* ps/push-delete-option (2015-12-16) 2 commits
422
- (merged to 'next' on 2015-12-22 at d83cc1d)
423
- + push: add '-d' as shorthand for '--delete'
424
- + push: add '--delete' flag to synopsis
425
-
426
- "branch --delete" has "branch -d" but "push --delete" does not.
427
-
428
- Will merge to 'master'.
429
-
430
-
431
-* dt/refs-backend-lmdb (2015-12-04) 16 commits
432
- - refs: tests for lmdb backend
433
- - refs: add LMDB refs backend
434
- - refs: allow ref backend to be set for clone
435
- - init: allow alternate backends to be set for new repos
436
- - refs: always handle non-normal refs in files backend
437
- - refs: move duplicate check to common code
438
- - refs: make lock generic
439
- - refs: add method to rename refs
440
- - refs: add methods to init refs backend and db
441
- - refs: add method for delete_refs
442
- - refs: add method for initial ref transaction commit
443
- - refs: add methods for reflog
444
- - refs: add do_for_each_per_worktree_ref
445
- - refs: add methods for the ref iterators
446
- - refs: add methods for misc ref operations
447
- - refs: add a backend method structure with transaction functions
583
+* dt/refs-backend-lmdb (2016-01-12) 22 commits
584
+ . DONTMERGE: compilation fix
585
+ . refs: tests for lmdb backend
586
+ . refs: add LMDB refs backend
587
+ . svn: learn ref-storage argument
588
+ . refs: allow ref backend to be set for clone
589
+ . clone: use child_process for recursive checkouts
590
+ . refs: check submodules ref storage config
591
+ . init: allow alternate backends to be set for new repos
592
+ . refs: always handle non-normal refs in files backend
593
+ . refs: resolve symbolic refs first
594
+ . refs: allow log-only updates
595
+ . refs: move duplicate check to common code
596
+ . refs: make lock generic
597
+ . refs: add method to rename refs
598
+ . refs: add methods to init refs db
599
+ . refs: add method for delete_refs
600
+ . refs: add method for initial ref transaction commit
601
+ . refs: add methods for reflog
602
+ . refs: add do_for_each_per_worktree_ref
603
+ . refs: add methods for the ref iterators
604
+ . refs: add methods for misc ref operations
605
+ . refs: add a backend method structure with transaction functions
606
607
Building on top of a few refs-backend preparatory series, LMDB
608
based refs backend has been plugged into the system.
609
452
- Still being discussed and worked on.
610
+ Rerolled, but left out of 'pu' for now due to conflicts.
611
612
613
* dw/subtree-split-do-not-drop-merge (2015-12-10) 1 commit
@@ -458,16 +616,17 @@ repositories listed at
616
The "split" subcommand of "git subtree" (in contrib/) incorrectly
617
skipped merges when it shouldn't, which was corrected.
618
461
- Waiting for review from 'subtree' folks.
619
+ Still being worked on.
620
+ ($gmane/283874)
621
622
623
* nd/clear-gitenv-upon-use-of-alias (2015-12-29) 5 commits
465
- - run-command: don't warn on SIGPIPE deaths
466
- - git.c: make sure we do not leak GIT_* to alias scripts
467
- - setup.c: re-fix d95138e (setup: set env $GIT_WORK_TREE when ..
468
- - git.c: make it clear save_env() is for alias handling only
469
- - Merge branch 'nd/stop-setenv-work-tree' into nd/clear-gitenv-upon-use-of-alias
470
- (this branch uses nd/stop-setenv-work-tree.)
624
+ (merged to 'next' on 2016-01-12 at 696b1f5)
625
+ + run-command: don't warn on SIGPIPE deaths
626
+ + git.c: make sure we do not leak GIT_* to alias scripts
627
+ + setup.c: re-fix d95138e (setup: set env $GIT_WORK_TREE when ..
628
+ + git.c: make it clear save_env() is for alias handling only
629
+ + Merge branch 'nd/stop-setenv-work-tree' into nd/clear-gitenv-upon-use-of-alias
630
631
d95138e6 (setup: set env $GIT_WORK_TREE when work tree is set, like
632
$GIT_DIR, 2015-06-26) attempted to work around a glitch in alias
@@ -477,7 +636,7 @@ repositories listed at
636
Try to address the same issue by always restoring the environment
637
and respawning the real underlying command when handling alias.
638
480
- Will merge to 'next'.
639
+ Will merge to 'master'.
640
641
642
* kn/ref-filter-atom-parsing (2016-01-05) 15 commits
@@ -515,9 +674,10 @@ repositories listed at
674
($gmane/281701)
675
676
518
-* dk/gc-more-wo-pack (2015-11-24) 3 commits
519
- - gc: Clean garbage .bitmap files from pack dir
520
- - t5304: Add test for .bitmap garbage files
677
+* dk/gc-more-wo-pack (2016-01-13) 4 commits
678
+ - gc: clean garbage .bitmap files from pack dir
679
+ - t5304: ensure non-garbage files are not deleted
680
+ - t5304: test .bitmap garbage files
681
- prepare_packed_git(): find more garbage
682
683
Follow-on to dk/gc-idx-wo-pack topic, to clean up stale
@@ -526,21 +686,16 @@ repositories listed at
686
Waiting for review.
687
688
529
-* ps/rebase-keep-empty (2015-11-24) 2 commits
530
- - rebase: fix preserving commits with --keep-empty
531
- - rebase: test broken behavior with --keep-empty
532
-
533
- Keep duplicate commits via rebase --keep-empty.
534
-
535
- Of dubious or negative value.
536
- Will discard.
537
- ($gmane/282107).
538
-
539
-
689
* rm/subtree-unwrap-tags (2015-11-24) 1 commit
690
- contrib/subtree: unwrap tag refs
691
543
- Waiting for review from subtree folks.
692
+ "git subtree" (in contrib/) records the tag object name in the
693
+ commit log message when a subtree is added using a tag, without
694
+ peeling it down to the underlying commit. The tag needs to be
695
+ peeled when "git subtree split" wants to work on the commit, but
696
+ the command forgot to do so.
697
+
698
+ Will merge to 'next'.
699
700
701
* sg/sh-require-clean-orphan (2015-11-24) 2 commits
@@ -561,7 +716,7 @@ repositories listed at
716
717
Add options to ls-files to help diagnose end-of-line problems.
718
564
- Needs review.
719
+ Will merge to 'next'.
720
721
722
* ec/annotate-deleted (2015-11-20) 1 commit
@@ -573,27 +728,7 @@ repositories listed at
728
Waiting for review.
729
730
576
-* sb/submodule-parallel-fetch (2015-12-16) 7 commits
577
- (merged to 'next' on 2015-12-22 at 44e84ff)
578
- + submodules: allow parallel fetching, add tests and documentation
579
- + fetch_populated_submodules: use new parallel job processing
580
- + run-command: add an asynchronous parallel child processor
581
- + sigchain: add command to pop all common signals
582
- + strbuf: add strbuf_read_once to read without blocking
583
- + xread: poll on non blocking fds
584
- + submodule.c: write "Fetching submodule <foo>" to stderr
585
- (this branch is used by sb/submodule-parallel-update.)
586
-
587
- Add a framework to spawn a group of processes in parallel, and use
588
- it to run "git fetch --recurse-submodules" in parallel.
589
-
590
- Rerolled and this seems to be a lot cleaner. The merge of the
591
- earlier one to 'next' has been reverted.
592
-
593
- Will merge to 'master'.
594
-
595
-
596
-* sb/submodule-parallel-update (2015-12-16) 8 commits
731
+* sb/submodule-parallel-update (2016-01-12) 8 commits
732
- clone: allow an explicit argument for parallel submodule clones
733
- submodule update: expose parallelism to the user
734
- git submodule update: have a dedicated helper for cloning
@@ -602,7 +737,6 @@ repositories listed at
737
- submodule-config: remove name_and_item_from_var
738
- submodule-config: drop check against NULL
739
- submodule-config: keep update strategy around
605
- (this branch uses sb/submodule-parallel-fetch.)
740
741
Builds on top of the "fetch --recurse-submodules" work to introduce
742
parallel downloading into multiple submodules for "submodule update".
@@ -635,16 +769,14 @@ repositories listed at
769
list of object names from the standard input instead of from the
770
command line, and does not involve files in the working tree.
771
638
- Rerolled.
639
- Needs review.
772
+ Will be rerolled.
773
774
642
-* mh/notes-allow-reading-treeish (2015-10-08) 3 commits
643
- (merged to 'next' on 2015-10-23 at 8a697f0)
775
+* mh/notes-allow-reading-treeish (2016-01-12) 1 commit
776
+ (merged to 'next' on 2016-01-12 at 7aa311f)
777
+ notes: allow treeish expressions as notes ref
645
- + Merge branch 'jk/notes-dwim-doc' into next
646
- + Merge branch 'jc/merge-drop-old-syntax' into next
647
- (this branch uses jc/merge-drop-old-syntax.)
778
+
779
+ Originally merged to 'next' on 2015-10-23
780
781
Some "git notes" operations, e.g. "git log --notes=<note>", should
782
be able to read notes from any tree-ish that is shaped like a notes
@@ -653,33 +785,12 @@ repositories listed at
785
when the operation would update the notes (in which case we must
786
have a place to store the updated notes tree, iow, a ref).
787
656
- As the patch was done on top of the 'drop old-syntax from merge',
657
- this has to wait until that other topic can graduate, unfortunately.
658
- It can be redone in a way that does not depend on that topic after
659
- this cycle, though.
660
-
661
- Will keep in 'next'.
662
-
663
-
664
-* jc/mailinfo (2015-10-21) 1 commit
665
- - mailinfo: ignore in-body header that we do not care about
666
-
667
- Some people write arbitrary garbage at the beginning of a piece of
668
- e-mail (or after -- >8 -- scissors -- >8 -- line) in the commit log
669
- message and expect them to be discarded, even though "From:" and
670
- "Subject:" are the only documented in-body headers that you are
671
- supposed to have there. Allow some garbage (specifically, what may
672
- look like RFC2822 headers like "MIME-Version: ...") to be there and
673
- ignore them.
674
-
675
- No comments after waiting for a long time.
676
- Will discard.
788
+ Will merge to 'master'.
789
790
791
* js/am-3-merge-recursive-direct (2015-10-12) 2 commits
680
- (merged to 'next' on 2015-10-23 at dc631e5)
681
- + am: make a direct call to merge_recursive
682
- + merge_recursive_options: introduce the "gently" flag
792
+ - am: make a direct call to merge_recursive
793
+ - merge_recursive_options: introduce the "gently" flag
794
795
The merge_recursive_generic() function has been made a bit safer to
796
call from inside a process. "git am -3" was taught to make a direct
@@ -697,38 +808,6 @@ repositories listed at
808
point of dying exactly because it knew it is going to exit, but now
809
they have to care, and they need to be audited.
810
700
- Will keep in 'next'.
701
-
702
-
703
-* sg/pretty-more-date-mode-format (2015-10-07) 1 commit
704
- - pretty: add format specifiers for short and raw date formats
705
-
706
- Introduce "%as" and "%aR" placeholders for "log --format" to show
707
- the author date in the short and raw formats.
708
-
709
- No comments after waiting for a long time.
710
- Will discard.
711
-
712
-
713
-* jk/graph-format-padding (2015-09-14) 1 commit
714
- - pretty: pass graph width to pretty formatting for use in '%>|(N)'
715
-
716
- Redefine the way '%>|(N)' padding and the "--graph" option
717
- interacts. It has been that the available columns to display the
718
- log message was measured from the edge of the area the graph ended,
719
- but with this it becomes the beginning of the entire output.
720
-
721
- I have a suspicion that 50% of the users would appreciate this
722
- change, and the remainder see this break their expectation. If
723
- that is the case, we might need to introduce a similar but
724
- different alignment operator so that this new behaviour is
725
- available to those who want to use it, without negatively affecting
726
- existing uses.
727
-
728
- No comments after waiting for a long time.
729
- Will discard.
730
- ($gmane/278326)
731
-
811
812
* ad/cygwin-wants-rename (2015-08-07) 1 commit
813
- config.mak.uname: Cygwin needs OBJECT_CREATION_USES_RENAMES
@@ -754,16 +833,10 @@ repositories listed at
833
834
835
* jc/merge-drop-old-syntax (2015-04-29) 1 commit
757
- (merged to 'next' on 2015-10-07 at 50fed71)
758
- + merge: drop 'git merge <message> HEAD <commit>' syntax
759
- (this branch is used by mh/notes-allow-reading-treeish.)
760
-
761
- Originally merged to 'next' on 2015-05-28
836
+ - merge: drop 'git merge <message> HEAD <commit>' syntax
837
838
Stop supporting "git merge <message> HEAD <commit>" syntax that has
839
been deprecated since October 2007. It has been reported that
840
git-gui still uses the deprecated syntax, which needs to be fixed
841
before this final step can proceed.
842
($gmane/282594)
768
-
769
- Will keep in 'next'.