What's cooking (2016/06 #09)
Junio C Hamano committed
Jun 27, 2016 at 16:20 UTC
2b74c924987b025d1312833f302c691a62445be2
1 file changed
+436
-362
whats-cooking.txt
+436
-362
@@ -1,10 +1,10 @@
1
To: git@vger.kernel.org
2
Bcc: lwn@lwn.net
3
-Subject: What's cooking in git.git (Jun 2016, #08; Thu, 23)
4
-X-master-at: ab7797dbe95fff38d9265869ea367020046db118
5
-X-next-at: 44ae68fed1736e48fe3c8224731bdda76bb82e75
3
+Subject: What's cooking in git.git (Jun 2016, #09; Mon, 27)
4
+X-master-at: cf4c2cfe52be5bd973a4838f73a35d3959ce2f43
5
+X-next-at: 760e8615f44f1ce38bfde1e76620c8f8fb1bd5c2
6
7
-What's cooking in git.git (Jun 2016, #08; Thu, 23)
7
+What's cooking in git.git (Jun 2016, #09; Mon, 27)
8
--------------------------------------------------
9
10
Here are the topics that have been cooking. Commits prefixed with
@@ -22,110 +22,309 @@ of the repositories listed at
22
http://git-blame.blogspot.com/p/git-public-repositories.html
23
24
--------------------------------------------------
25
-[New Topics]
25
+[Graduated to "master"]
26
27
-* ak/t7800-wo-readlink (2016-06-21) 1 commit
28
- - t7800: readlink may not be available
27
+* ap/git-svn-propset-doc (2016-06-15) 1 commit
28
+ (merged to 'next' on 2016-06-22 at 5a34f7d)
29
+ + git-svn: document the 'git svn propset' command
30
30
- One among four invocations of readlink(1) in our test suite has
31
- been rewritten so that the test can run on systems without the
32
- command (others are in valgrind test framework and t9802).
31
+ "git svn propset" subcommand that was added in 2.3 days is
32
+ documented now.
33
34
- Will merge to 'next'.
34
35
+* cc/apply-introduce-state (2016-06-06) 50 commits
36
+ (merged to 'next' on 2016-06-20 at 4f205b8)
37
+ + builtin/apply: remove misleading comment on lock_file field
38
+ + builtin/apply: move 'newfd' global into 'struct apply_state'
39
+ + builtin/apply: add 'lock_file' pointer into 'struct apply_state'
40
+ + builtin/apply: move applying patches into apply_all_patches()
41
+ + builtin/apply: move 'state' check into check_apply_state()
42
+ + builtin/apply: move 'symlink_changes' global into 'struct apply_state'
43
+ + builtin/apply: move 'fn_table' global into 'struct apply_state'
44
+ + builtin/apply: move 'state_linenr' global into 'struct apply_state'
45
+ + builtin/apply: move 'max_change' and 'max_len' into 'struct apply_state'
46
+ + builtin/apply: move 'ws_ignore_action' into 'struct apply_state'
47
+ + builtin/apply: move 'ws_error_action' into 'struct apply_state'
48
+ + builtin/apply: move 'applied_after_fixing_ws' into 'struct apply_state'
49
+ + builtin/apply: move 'squelch_whitespace_errors' into 'struct apply_state'
50
+ + builtin/apply: remove whitespace_option arg from set_default_whitespace_mode()
51
+ + builtin/apply: move 'whitespace_option' into 'struct apply_state'
52
+ + builtin/apply: move 'whitespace_error' global into 'struct apply_state'
53
+ + builtin/apply: move 'root' global into 'struct apply_state'
54
+ + builtin/apply: move 'p_value_known' global into 'struct apply_state'
55
+ + builtin/apply: move 'p_value' global into 'struct apply_state'
56
+ + builtin/apply: move 'has_include' global into 'struct apply_state'
57
+ + builtin/apply: move 'limit_by_name' global into 'struct apply_state'
58
+ + builtin/apply: move 'patch_input_file' global into 'struct apply_state'
59
+ + builtin/apply: move 'apply' global into 'struct apply_state'
60
+ + builtin/apply: move 'p_context' global into 'struct apply_state'
61
+ + builtin/apply: move 'fake_ancestor' global into 'struct apply_state'
62
+ + builtin/apply: move 'line_termination' global into 'struct apply_state'
63
+ + builtin/apply: move 'unsafe_paths' global into 'struct apply_state'
64
+ + builtin/apply: move 'no_add' global into 'struct apply_state'
65
+ + builtin/apply: move 'threeway' global into 'struct apply_state'
66
+ + builtin/apply: move 'summary' global into 'struct apply_state'
67
+ + builtin/apply: move 'numstat' global into 'struct apply_state'
68
+ + builtin/apply: move 'diffstat' global into 'struct apply_state'
69
+ + builtin/apply: move 'cached' global into 'struct apply_state'
70
+ + builtin/apply: move 'allow_overlap' global into 'struct apply_state'
71
+ + builtin/apply: move 'update_index' global into 'struct apply_state'
72
+ + builtin/apply: move 'apply_verbosely' global into 'struct apply_state'
73
+ + builtin/apply: move 'apply_with_reject' global into 'struct apply_state'
74
+ + builtin/apply: move 'apply_in_reverse' global into 'struct apply_state'
75
+ + builtin/apply: move 'check_index' global into 'struct apply_state'
76
+ + builtin/apply: move 'check' global into 'struct apply_state'
77
+ + builtin/apply: move 'unidiff_zero' global into 'struct apply_state'
78
+ + builtin/apply: move 'state' init into init_apply_state()
79
+ + builtin/apply: introduce 'struct apply_state' to start libifying
80
+ + builtin/apply: move 'read_stdin' global into cmd_apply()
81
+ + builtin/apply: move 'options' variable into cmd_apply()
82
+ + builtin/apply: extract line_by_line_fuzzy_match() from match_fragment()
83
+ + builtin/apply: avoid local variable shadowing 'len' parameter
84
+ + builtin/apply: avoid parameter shadowing 'linenr' global
85
+ + builtin/apply: avoid parameter shadowing 'p_value' global
86
+ + builtin/apply: make gitdiff_verify_name() return void
87
37
-* js/perf-on-apple (2016-06-21) 1 commit
38
- - perf: accommodate for MacOSX
88
+ Originally merged to 'next' on 2016-06-06
89
40
- t/perf needs /usr/bin/time with GNU extension; the invocation of it
41
- is updated to "gtime" on Darwin.
90
+ The "git apply" standalone program is being libified; this is the
91
+ first step to move many state variables into a structure that can
92
+ be explicitly (re)initialized to make the machinery callable more
93
+ than once.
94
43
- Will merge to 'next'.
95
+ The next step that moves some remaining state variables into the
96
+ structure and turns die()s into an error return that propagates up
97
+ to the caller is not queued yet but in flight. It would be good to
98
+ review the above first and give the remainder of the series a solid
99
+ base to build on.
100
101
46
-* sb/t5614-modernize (2016-06-21) 1 commit
47
- - t5614: don't use subshells
48
- (this branch uses sb/clone-shallow-passthru.)
102
+* dn/gpg-doc (2016-06-16) 1 commit
103
+ (merged to 'next' on 2016-06-22 at f467355)
104
+ + Documentation: GPG capitalization
105
50
- Test clean-up.
106
+ The documentation tries to consistently spell "GPG"; when
107
+ referring to the specific program name, "gpg" is used.
108
52
- Will merge to 'next'.
109
110
+* em/newer-freebsd-shells-are-fine-with-returns (2016-06-17) 1 commit
111
+ (merged to 'next' on 2016-06-22 at dbee33c)
112
+ + rebase: update comment about FreeBSD /bin/sh
113
55
-* ao/p4-has-branch-prefix-fix (2016-06-22) 1 commit
56
- - git-p4: correct hasBranchPrefix verbose output
114
+ Comments about misbehaving FreeBSD shells have been clarified with
115
+ the version number (9.x and before are broken, newer ones are OK).
116
58
- A bug, which caused "git p4" while running under verbose mode to
59
- report paths that are omitted due to branch prefix incorrectly, has
60
- been fixed; the command said "Ignoring file outside of prefix" for
61
- paths that are _inside_.
117
63
- Will merge to 'next'.
118
+* et/add-chmod-x (2016-06-07) 1 commit
119
+ (merged to 'next' on 2016-06-22 at 71d65a0)
120
+ + add: add --chmod=+x / --chmod=-x options
121
122
+ "git update-index --add --chmod=+x file" may be usable as an escape
123
+ hatch, but not a friendly thing to force for people who do need to
124
+ use it regularly. "git add --chmod=+x file" can be used instead.
125
66
-* cb/t7810-test-label-fix (2016-06-21) 1 commit
67
- - t7810: fix duplicated test title
126
69
- Test clean-up.
127
+* jc/deref-tag (2016-06-14) 1 commit
128
+ (merged to 'next' on 2016-06-22 at 1075713)
129
+ + blame, line-log: do not loop around deref_tag()
130
71
- Will merge to 'next'.
131
+ Code clean-up.
132
133
74
-* jc/t2300-setup (2016-06-22) 1 commit
75
- - t2300: "git --exec-path" is not usable in $PATH on Windows as-is
134
+* jk/avoid-unbounded-alloca (2016-06-07) 1 commit
135
+ (merged to 'next' on 2016-06-22 at 93feb23)
136
+ + tree-diff: avoid alloca for large allocations
137
77
- Portability fix for Windows.
138
+ A codepath that used alloca(3) to place an unbounded amount of data
139
+ on the stack has been updated to avoid doing so.
140
79
- Will merge to 'next'.
141
142
+* jk/fetch-prune-doc (2016-06-14) 1 commit
143
+ (merged to 'next' on 2016-06-22 at 6563376)
144
+ + fetch: document that pruning happens before fetching
145
+
146
+ Minor doc update.
147
82
-* jk/perf-any-version (2016-06-22) 2 commits
83
- - p4211: explicitly disable renames in no-rename test
84
- - t/perf: fix regression in testing older versions of git
148
86
- Allow t/perf framework to use the features from the most recent
87
- version of Git even when testing an older installed version.
149
+* lf/receive-pack-auto-gc-to-client (2016-06-06) 1 commit
150
+ (merged to 'next' on 2016-06-22 at 92162f5)
151
+ + receive-pack: send auto-gc output over sideband 2
152
+
153
+ Messages that are generated by auto gc during "git push" on the
154
+ receiving end are now passed back to the sending end in such a way
155
+ that they are shown with "remote: " prefix to avoid confusing the
156
+ users.
157
+
158
+
159
+* lv/status-say-working-tree-not-directory (2016-06-09) 1 commit
160
+ (merged to 'next' on 2016-06-22 at c65c7c1)
161
+ + Use "working tree" instead of "working directory" for git status
162
+
163
+ "git status" used to say "working directory" when it meant "working
164
+ tree".
165
+
166
+
167
+* mg/cherry-pick-multi-on-unborn (2016-06-06) 1 commit
168
+ (merged to 'next' on 2016-06-22 at 183295b)
169
+ + cherry-pick: allow to pick to unborn branches
170
+
171
+ "git cherry-pick A" worked on an unborn branch, but "git
172
+ cherry-pick A..B" didn't.
173
+
174
+
175
+* nb/gnome-keyring-build (2016-06-14) 1 commit
176
+ (merged to 'next' on 2016-06-22 at 0dfb90c)
177
+ + gnome-keyring: Don't hard-code pkg-config executable
178
+
179
+ Build improvements for gnome-keyring (in contrib/)
180
+
181
+
182
+* pb/strbuf-read-file-doc (2016-06-14) 1 commit
183
+ (merged to 'next' on 2016-06-22 at 10e4b4f)
184
+ + strbuf: describe the return value of strbuf_read_file
185
+
186
+ Minor doc update.
187
+
188
+
189
+* pc/occurred (2016-06-10) 2 commits
190
+ (merged to 'next' on 2016-06-22 at ce0b944)
191
+ + config.c: fix misspelt "occurred" in an error message
192
+ + refs.h: fix misspelt "occurred" in a comment
193
+
194
+ Typofix.
195
+
196
+
197
+* rj/compat-regex-size-max-fix (2016-06-06) 1 commit
198
+ (merged to 'next' on 2016-06-22 at 376c5b1)
199
+ + regex: fix a SIZE_MAX macro redefinition warning
200
+
201
+ A compilation fix.
202
+
203
+
204
+* sg/reflog-past-root (2016-06-06) 1 commit
205
+ (merged to 'next' on 2016-06-22 at c5d4e29)
206
+ + reflog: continue walking the reflog past root commits
207
+
208
+ "git reflog" stopped upon seeing an entry that denotes a branch
209
+ creation event (aka "unborn"), which made it appear as if the
210
+ reflog was truncated.
211
+
212
+
213
+* tb/complete-status (2016-06-10) 3 commits
214
+ (merged to 'next' on 2016-06-22 at 44ae68f)
215
+ + completion: add git status
216
+ + completion: add __git_get_option_value helper
217
+ + completion: factor out untracked file modes into a variable
218
+
219
+ The completion script (in contrib/) learned to complete "git
220
+ status" options.
221
+
222
+
223
+* tr/doc-tt (2016-06-08) 4 commits
224
+ (merged to 'next' on 2016-06-22 at dc6df3b)
225
+ + doc: change configuration variables format
226
+ + doc: more consistency in environment variables format
227
+ + doc: change environment variables format
228
+ + doc: clearer rule about formatting literals
229
+
230
+ The documentation set has been updated so that literal commands,
231
+ configuration variables and environment variables are consistently
232
+ typeset in fixed-width font and bold in manpages.
233
+
234
+
235
+* vs/prompt-avoid-unset-variable (2016-06-06) 1 commit
236
+ (merged to 'next' on 2016-06-22 at 8bf21d3)
237
+ + git-prompt.sh: Don't error on null ${ZSH,BASH}_VERSION, $short_sha
238
+
239
+ The git-prompt scriptlet (in contrib/) was not friendly with those
240
+ who uses "set -u", which has been fixed.
241
+
242
+--------------------------------------------------
243
+[New Topics]
244
+
245
+* jk/test-match-signal (2016-06-24) 4 commits
246
+ - t/lib-git-daemon: use test_match_signal
247
+ - test_must_fail: use test_match_signal
248
+ - t0005: use test_match_signal as appropriate
249
+ - tests: factor portable signal check out of t0005
250
+
251
+ The test framework learned a new helper test_match_signal to check
252
+ an exit code from getting killed by an expected signal.
253
254
Will merge to 'next'.
255
256
92
-* jn/preformatted-doc-url (2016-06-22) 1 commit
93
- - doc: git-htmldocs.googlecode.com is no more
257
+* ah/unpack-trees-advice-messages (2016-06-27) 1 commit
258
+ - unpack-trees: fix English grammar in do-this-before-that messages
259
260
Will merge to 'next'.
261
262
98
-* ex/deprecate-empty-pathspec-as-match-all (2016-06-22) 1 commit
99
- - pathspec: warn on empty strings as pathspec
263
+* ew/gc-auto-pack-limit-fix (2016-06-27) 1 commit
264
+ - gc: fix off-by-one error with gc.autoPackLimit
265
101
- An empty string used as a pathspec element has always meant
102
- 'everything matches', but it is too easy to write a script that
103
- finds a path to remove in $path and run 'git rm "$paht"', which
104
- ends up removing everything. Start warning about this use of an
105
- empty string used for 'everything matches' and ask users to use a
106
- more explicit '.' for that instead.
266
+ "gc.autoPackLimit" when set to 1 should not trigger a repacking
267
+ when there is only one pack, but the code counted poorly and did
268
+ so.
269
108
- The hope is that existing users will not mind this change, and
109
- eventually the warning can be turned into a hard error, upgrading
110
- the deprecation into removal of this (mis)feature.
270
+ Will merge to 'next'.
271
112
- Will wait for further comments for a bit before merging to 'next'.
272
273
+* nd/connect-ssh-command-config (2016-06-27) 1 commit
274
+ - connect: read $GIT_SSH_COMMAND from config file
275
115
-* jk/ansi-color (2016-06-23) 7 commits
116
- - color: support strike-through attribute
117
- - color: support "italic" attribute
118
- - color: allow "no-" for negating attributes
119
- - color: refactor parse_attr
120
- - add skip_prefix_mem helper
121
- - doc: refactor description of color format
122
- - color: fix max-size comment
276
+ A new configuration variable core.sshCommand to specify what value
277
+ for GIT_SSH_COMMAND to use per repository.
278
124
- The output coloring scheme learned two new attributes, italic and
125
- strike, in addition to existing bold, reverse, etc.
279
+
280
+* nd/doc-new-command (2016-06-27) 1 commit
281
+ - new-command.txt: correct the command description file
282
+
283
+ Will merge to 'next'.
284
+
285
+
286
+* po/range-doc (2016-06-27) 3 commits
287
+ - doc: give headings for the two and three dot notations
288
+ - doc: show the actual left, right, and boundary marks
289
+ - doc: use 'symmetric difference' consistently
290
+
291
+
292
+* sb/submodule-parallel-fetch (2016-06-27) 2 commits
293
+ - xwrite: poll on non-blocking FDs
294
+ - xread: retry after poll on EAGAIN/EWOULDBLOCK
295
+
296
+ Fix a recently introduced codepaths that are involved in parallel
297
+ submodule operations, which gave up on reading too early, and
298
+ could have wasted CPU while attempting to write under a corner case
299
+ condition.
300
301
Will merge to 'next'.
302
303
+
304
+* mm/doc-tt (2016-06-27) 6 commits
305
+ - doc: typeset HEAD and variants as litteral
306
+ - CodingGuidelines: formatting HEAD in documentation
307
+ - doc: typeset long options with argument as litteral
308
+ - doc: typeset -- as litteral
309
+ - doc: typeset long command-line options as literal
310
+ - doc: typeset short command-line options as literal
311
+
312
+ More mark-up updates to typeset strings that are expected to
313
+ literally typed by the end user in fixed-width font.
314
+
315
+
316
+* nd/fetch-ref-summary (2016-06-27) 5 commits
317
+ - fetch: reduce duplicate in ref update status lines with placeholder
318
+ - fetch: align all "remote -> local" output
319
+ - fetch: change flag code for displaying tag update and deleted ref
320
+ - fetch: refactor ref update status formatting code
321
+ - git-fetch.txt: document fetch output
322
+
323
+ Improve the look of the way "git fetch" reports what happened to
324
+ each ref that was fetched.
325
+
326
+ Still being discussed.
327
+
328
--------------------------------------------------
329
[Stalled]
330
@@ -244,13 +443,118 @@ of the repositories listed at
443
--------------------------------------------------
444
[Cooking]
445
247
-* nd/icase (2016-06-23) 11 commits
446
+* ak/t7800-wo-readlink (2016-06-21) 1 commit
447
+ - t7800: readlink may not be available
448
+
449
+ One among four invocations of readlink(1) in our test suite has
450
+ been rewritten so that the test can run on systems without the
451
+ command (others are in valgrind test framework and t9802).
452
+
453
+ Will merge to 'next'.
454
+
455
+
456
+* js/perf-on-apple (2016-06-21) 1 commit
457
+ - perf: accommodate for MacOSX
458
+
459
+ t/perf needs /usr/bin/time with GNU extension; the invocation of it
460
+ is updated to "gtime" on Darwin.
461
+
462
+ Will merge to 'next'.
463
+
464
+
465
+* sb/t5614-modernize (2016-06-21) 1 commit
466
+ - t5614: don't use subshells
467
+ (this branch uses sb/clone-shallow-passthru.)
468
+
469
+ Test clean-up.
470
+
471
+ Will merge to 'next'.
472
+
473
+
474
+* ao/p4-has-branch-prefix-fix (2016-06-22) 1 commit
475
+ - git-p4: correct hasBranchPrefix verbose output
476
+
477
+ A bug, which caused "git p4" while running under verbose mode to
478
+ report paths that are omitted due to branch prefix incorrectly, has
479
+ been fixed; the command said "Ignoring file outside of prefix" for
480
+ paths that are _inside_.
481
+
482
+ Will merge to 'next'.
483
+
484
+
485
+* cb/t7810-test-label-fix (2016-06-21) 1 commit
486
+ - t7810: fix duplicated test title
487
+
488
+ Test clean-up.
489
+
490
+ Will merge to 'next'.
491
+
492
+
493
+* jc/t2300-setup (2016-06-22) 1 commit
494
+ - t2300: "git --exec-path" is not usable in $PATH on Windows as-is
495
+
496
+ Portability fix for Windows.
497
+
498
+ Will merge to 'next'.
499
+
500
+
501
+* jk/perf-any-version (2016-06-22) 2 commits
502
+ - p4211: explicitly disable renames in no-rename test
503
+ - t/perf: fix regression in testing older versions of git
504
+
505
+ Allow t/perf framework to use the features from the most recent
506
+ version of Git even when testing an older installed version.
507
+
508
+ Will merge to 'next'.
509
+
510
+
511
+* jn/preformatted-doc-url (2016-06-22) 1 commit
512
+ - doc: git-htmldocs.googlecode.com is no more
513
+
514
+ Will merge to 'next'.
515
+
516
+
517
+* ex/deprecate-empty-pathspec-as-match-all (2016-06-22) 1 commit
518
+ - pathspec: warn on empty strings as pathspec
519
+
520
+ An empty string used as a pathspec element has always meant
521
+ 'everything matches', but it is too easy to write a script that
522
+ finds a path to remove in $path and run 'git rm "$paht"', which
523
+ ends up removing everything. Start warning about this use of an
524
+ empty string used for 'everything matches' and ask users to use a
525
+ more explicit '.' for that instead.
526
+
527
+ The hope is that existing users will not mind this change, and
528
+ eventually the warning can be turned into a hard error, upgrading
529
+ the deprecation into removal of this (mis)feature.
530
+
531
+ Will wait for further comments for a bit before merging to 'next'.
532
+
533
+
534
+* jk/ansi-color (2016-06-23) 7 commits
535
+ - color: support strike-through attribute
536
+ - color: support "italic" attribute
537
+ - color: allow "no-" for negating attributes
538
+ - color: refactor parse_attr
539
+ - add skip_prefix_mem helper
540
+ - doc: refactor description of color format
541
+ - color: fix max-size comment
542
+
543
+ The output coloring scheme learned two new attributes, italic and
544
+ strike, in addition to existing bold, reverse, etc.
545
+
546
+ Will merge to 'next'.
547
+
548
+
549
+* nd/icase (2016-06-27) 13 commits
550
+ - SQUASH???
551
- grep.c: reuse "icase" variable
552
- diffcore-pickaxe: support case insensitive match on non-ascii
250
- - diffcore-pickaxe: "share" regex error handling code
553
+ - diffcore-pickaxe: Add regcomp_or_die()
554
- grep/pcre: support utf-8
555
- gettext: add is_utf8_locale()
556
- grep/pcre: prepare locale-dependent tables for icase matching
557
+ - grep: rewrite an if/else condition to avoid duplicate expression
558
- grep/icase: avoid kwsset when -F is specified
559
- grep/icase: avoid kwsset on literal non-ascii strings
560
- test-regex: expose full regcomp() to the command line
@@ -263,25 +567,15 @@ of the repositories listed at
567
Modulo minor possible nits, this round looked mostly sensible.
568
569
266
-* mj/log-show-signature-conf (2016-06-22) 3 commits
267
- - log: add log.showSignature configuration variable
268
- - log: add "--no-show-signature" command line option
269
- - t4202: refactor test
270
-
271
- "git log" learns log.showSignature configuration variable, and a
272
- command line option "--no-show-signature" to countermand it.
273
-
274
- The value of the first step is unclear. The top two looked OK.
275
-
276
-
277
-* em/newer-freebsd-shells-are-fine-with-returns (2016-06-17) 1 commit
278
- (merged to 'next' on 2016-06-22 at dbee33c)
279
- + rebase: update comment about FreeBSD /bin/sh
570
+* mj/log-show-signature-conf (2016-06-24) 3 commits
571
+ - log: add log.showSignature configuration variable
572
+ - log: add "--no-show-signature" command line option
573
+ - t4202: refactor test
574
281
- Comments about misbehaving FreeBSD shells have been clarified with
282
- the version number (9.x and before are broken, newer ones are OK).
575
+ "git log" learns log.showSignature configuration variable, and a
576
+ command line option "--no-show-signature" to countermand it.
577
284
- Will merge to 'master'.
578
+ Will merge to 'next'.
579
580
581
* mg/signature-doc (2016-06-17) 4 commits
@@ -297,8 +591,9 @@ of the repositories listed at
591
592
593
* jk/string-list-static-init (2016-06-13) 2 commits
300
- - use string_list initializer consistently
301
- - Merge branch 'jk/parseopt-string-list' into jk/string-list-static-init
594
+ (merged to 'next' on 2016-06-27 at 3d4b2fa)
595
+ + use string_list initializer consistently
596
+ + Merge branch 'jk/parseopt-string-list' into jk/string-list-static-init
597
(this branch uses jk/parseopt-string-list.)
598
599
Instead of taking advantage of a struct string_list that is
@@ -306,7 +601,7 @@ of the repositories listed at
601
initialize them explicitly as such, to document their behaviour
602
better.
603
309
- Will merge to 'next'.
604
+ Will merge to 'master'.
605
606
607
* mh/ref-store (2016-06-20) 38 commits
@@ -424,7 +719,9 @@ of the repositories listed at
719
Will merge to 'next'.
720
721
427
-* js/log-to-diffopt-file (2016-06-21) 9 commits
722
+* js/log-to-diffopt-file (2016-06-24) 10 commits
723
+ - t4211: ensure that log respects --output=<file>
724
+ - shortlog: respect the --output=<file> setting
725
- format-patch: use stdout directly
726
- format-patch: avoid freopen()
727
- format-patch: explicitly switch off color when writing to files
@@ -432,15 +729,14 @@ of the repositories listed at
729
- graph: respect the diffopt.file setting
730
- line-log: respect diffopt's configured output file stream
731
- log-tree: respect diffopt's configured output file stream
435
- - Disallow diffopt.close_file when using the log_tree machinery
436
- - am: stop ignoring errors reported by log_tree_diff()
732
+ - log: prepare log/log-tree to reuse the diffopt.close_file attribute
733
438
- The commands in the "log/diff" family had an optional FILE* pointer
439
- in the data structure they pass around for a long time, but some
440
- codepaths used to always write to the standard output. As a
441
- preparatory step to make "git format-patch" available to the
442
- internal callers, these codepaths have been updated to consistently
443
- write into that FILE* instead.
734
+ The commands in the "log/diff" family had an FILE* pointer in the
735
+ data structure they pass around for a long time, but some codepaths
736
+ used to always write to the standard output. As a preparatory step
737
+ to make "git format-patch" available to the internal callers, these
738
+ codepaths have been updated to consistently write into that FILE*
739
+ instead.
740
741
742
* js/mingw-parameter-less-c-functions (2016-06-20) 1 commit
@@ -481,24 +777,15 @@ of the repositories listed at
777
Will merge to 'next'.
778
779
484
-* ap/git-svn-propset-doc (2016-06-15) 1 commit
485
- (merged to 'next' on 2016-06-22 at 5a34f7d)
486
- + git-svn: document the 'git svn propset' command
487
-
488
- "git svn propset" subcommand that was added in 2.3 days is
489
- documented now.
490
-
491
- Will merge to 'master'.
492
-
493
-
780
* jk/add-i-diff-compact-heuristics (2016-06-16) 1 commit
495
- - add--interactive: respect diff.compactionHeuristic
781
+ (merged to 'next' on 2016-06-27 at 568f892)
782
+ + add--interactive: respect diff.compactionHeuristic
783
784
"git add -i/-p" learned to honor diff.compactionHeuristic
785
experimental knob, so that the user can work on the same hunk split
786
as "git diff" output.
787
501
- Will merge to 'next'.
788
+ Will merge to 'master'.
789
790
791
* jk/big-and-future-archive-tar (2016-06-21) 2 commits
@@ -510,6 +797,7 @@ of the repositories listed at
797
format.
798
799
Expecting a reroll.
800
+ ($gmane/298119)
801
802
803
* jk/gpg-interface-cleanup (2016-06-17) 7 commits
@@ -534,12 +822,13 @@ of the repositories listed at
822
823
824
* lf/sideband-returns-void (2016-06-16) 2 commits
537
- - upload-pack.c: make send_client_data() return void
538
- - sideband.c: make send_sideband() return void
825
+ (merged to 'next' on 2016-06-27 at 558c781)
826
+ + upload-pack.c: make send_client_data() return void
827
+ + sideband.c: make send_sideband() return void
828
829
A small internal API cleanup.
830
542
- Will merge to 'next'.
831
+ Will merge to 'master'.
832
833
834
* nd/graph-width-padded (2016-06-16) 2 commits
@@ -555,46 +844,28 @@ of the repositories listed at
844
Will merge to 'next'.
845
846
558
-* dn/gpg-doc (2016-06-16) 1 commit
559
- (merged to 'next' on 2016-06-22 at f467355)
560
- + Documentation: GPG capitalization
561
-
562
- The documentation tries to consistently spell "GPG"; when
563
- referring to the specific program name, "gpg" is used.
564
-
565
- Will merge to 'master'.
566
-
567
-
847
* jk/bisect-show-tree (2016-06-16) 1 commit
569
- - bisect: always call setup_revisions after init_revisions
848
+ (merged to 'next' on 2016-06-27 at 6970f87e)
849
+ + bisect: always call setup_revisions after init_revisions
850
851
"git bisect" makes an internal call to "git diff-tree" when
852
bisection finds the culprit, but this call did not initialize the
853
data structure to pass to the diff-tree API correctly.
854
575
- Will merge to 'next'.
576
-
577
-
578
-* lv/status-say-working-tree-not-directory (2016-06-09) 1 commit
579
- (merged to 'next' on 2016-06-22 at c65c7c1)
580
- + Use "working tree" instead of "working directory" for git status
581
-
582
- "git status" used to say "working directory" when it meant "working
583
- tree".
584
-
855
Will merge to 'master'.
856
857
858
* jk/parseopt-string-list (2016-06-13) 3 commits
589
- - blame,shortlog: don't make local option variables static
590
- - interpret-trailers: don't duplicate option strings
591
- - parse_opt_string_list: stop allocating new strings
859
+ (merged to 'next' on 2016-06-27 at 27462e6)
860
+ + blame,shortlog: don't make local option variables static
861
+ + interpret-trailers: don't duplicate option strings
862
+ + parse_opt_string_list: stop allocating new strings
863
(this branch is used by jk/string-list-static-init.)
864
865
The command line argument parsing that uses OPT_STRING_LIST() often
866
made a copy of the argv[] element, which was unnecessary.
867
597
- Will merge to 'next'.
868
+ Will merge to 'master'.
869
870
871
* jk/repack-keep-unreachable (2016-06-14) 3 commits
@@ -621,18 +892,11 @@ of the repositories listed at
892
893
894
* nd/test-lib-httpd-show-error-log-in-verbose (2016-06-13) 1 commit
624
- - lib-httpd.sh: print error.log on error
895
+ (merged to 'next' on 2016-06-27 at 9793d81)
896
+ + lib-httpd.sh: print error.log on error
897
898
Debugging aid.
899
628
- Will merge to 'next'.
629
-
630
-
631
-* pc/occurred (2016-06-10) 2 commits
632
- (merged to 'next' on 2016-06-22 at ce0b944)
633
- + config.c: fix misspelt "occurred" in an error message
634
- + refs.h: fix misspelt "occurred" in a comment
635
-
900
Will merge to 'master'.
901
902
@@ -659,40 +923,12 @@ of the repositories listed at
923
Any supporters? Otherwise will drop.
924
925
662
-* jc/deref-tag (2016-06-14) 1 commit
663
- (merged to 'next' on 2016-06-22 at 1075713)
664
- + blame, line-log: do not loop around deref_tag()
665
-
666
- Code clean-up.
667
-
668
- Will merge to 'master'.
669
-
670
-
671
-* jk/fetch-prune-doc (2016-06-14) 1 commit
672
- (merged to 'next' on 2016-06-22 at 6563376)
673
- + fetch: document that pruning happens before fetching
674
-
675
- Will merge to 'master'.
676
-
677
-
926
* km/fetch-do-not-free-remote-name (2016-06-14) 1 commit
679
- - builtin/fetch.c: don't free remote->name after fetch
680
-
681
- Will merge to 'next'.
682
-
683
-
684
-* nb/gnome-keyring-build (2016-06-14) 1 commit
685
- (merged to 'next' on 2016-06-22 at 0dfb90c)
686
- + gnome-keyring: Don't hard-code pkg-config executable
687
-
688
- Build improvements for gnome-keyring (in contrib/)
927
+ (merged to 'next' on 2016-06-27 at 4bc34c4)
928
+ + builtin/fetch.c: don't free remote->name after fetch
929
690
- Will merge to 'master'.
691
-
692
-
693
-* pb/strbuf-read-file-doc (2016-06-14) 1 commit
694
- (merged to 'next' on 2016-06-22 at 10e4b4f)
695
- + strbuf: describe the return value of strbuf_read_file
930
+ The ownership rule for the piece of memory that hold references to
931
+ be fetched in "git fetch" was screwy, which has been cleaned up.
932
933
Will merge to 'master'.
934
@@ -739,28 +975,23 @@ of the repositories listed at
975
would think.
976
977
742
-* jk/avoid-unbounded-alloca (2016-06-07) 1 commit
743
- (merged to 'next' on 2016-06-22 at 93feb23)
744
- + tree-diff: avoid alloca for large allocations
745
-
746
- Will merge to 'master'.
747
-
748
-
978
* jk/send-pack-stdio (2016-06-10) 2 commits
750
- - write_or_die: remove the unused write_or_whine() function
751
- - send-pack: use buffered I/O to talk to pack-objects
979
+ (merged to 'next' on 2016-06-27 at 2cfb0ff)
980
+ + write_or_die: remove the unused write_or_whine() function
981
+ + send-pack: use buffered I/O to talk to pack-objects
982
983
Code clean-up.
984
755
- Will merge to 'next'.
985
+ Will merge to 'master'.
986
987
988
* pb/commit-editmsg-path (2016-06-09) 1 commit
759
- - builtin/commit.c: memoize git-path for COMMIT_EDITMSG
989
+ (merged to 'next' on 2016-06-27 at 0f01ce1)
990
+ + builtin/commit.c: memoize git-path for COMMIT_EDITMSG
991
992
Code clean-up.
993
763
- Will merge to 'next'.
994
+ Will merge to 'master'.
995
996
997
* jc/attr-more (2016-06-09) 8 commits
@@ -814,95 +1045,6 @@ of the repositories listed at
1045
Will merge to 'next'.
1046
1047
817
-* lf/receive-pack-auto-gc-to-client (2016-06-06) 1 commit
818
- (merged to 'next' on 2016-06-22 at 92162f5)
819
- + receive-pack: send auto-gc output over sideband 2
820
-
821
- Allow messages that are generated by auto gc during "git push" on
822
- the receiving end to be explicitly passed back to the sending end
823
- over sideband, so that they are shown with "remote: " prefix to
824
- avoid confusing the users.
825
-
826
- Will merge to 'master'.
827
-
828
-
829
-* mg/cherry-pick-multi-on-unborn (2016-06-06) 1 commit
830
- (merged to 'next' on 2016-06-22 at 183295b)
831
- + cherry-pick: allow to pick to unborn branches
832
-
833
- "git cherry-pick A" worked on an unborn branch, but "git
834
- cherry-pick A..B" didn't.
835
-
836
- Will merge to 'master'.
837
-
838
-
839
-* sg/reflog-past-root (2016-06-06) 1 commit
840
- (merged to 'next' on 2016-06-22 at c5d4e29)
841
- + reflog: continue walking the reflog past root commits
842
-
843
- "git reflog" stopped upon seeing an entry that denotes a branch
844
- creation event (aka "unborn"), which made it appear as if the
845
- reflog was truncated.
846
-
847
- Will merge to 'master'.
848
-
849
-
850
-* tb/complete-status (2016-06-10) 3 commits
851
- (merged to 'next' on 2016-06-22 at 44ae68f)
852
- + completion: add git status
853
- + completion: add __git_get_option_value helper
854
- + completion: factor out untracked file modes into a variable
855
-
856
- The completion script (in contrib/) learned to complete "git
857
- status" options.
858
-
859
- Will merge to 'master'.
860
-
861
-
862
-* tr/doc-tt (2016-06-08) 4 commits
863
- (merged to 'next' on 2016-06-22 at dc6df3b)
864
- + doc: change configuration variables format
865
- + doc: more consistency in environment variables format
866
- + doc: change environment variables format
867
- + doc: clearer rule about formatting literals
868
-
869
- The documentation set has been updated so that literal commands,
870
- configuration variables and environment variables are consistently
871
- typeset in fixed-width font and bold in manpages.
872
-
873
- Will merge to 'master'.
874
-
875
-
876
-* vs/prompt-avoid-unset-variable (2016-06-06) 1 commit
877
- (merged to 'next' on 2016-06-22 at 8bf21d3)
878
- + git-prompt.sh: Don't error on null ${ZSH,BASH}_VERSION, $short_sha
879
-
880
- The git-prompt scriptlet (in contrib/) was not friendly with those
881
- who uses "set -u", which has been fixed.
882
-
883
- Will merge to 'master'.
884
-
885
-
886
-* rj/compat-regex-size-max-fix (2016-06-06) 1 commit
887
- (merged to 'next' on 2016-06-22 at 376c5b1)
888
- + regex: fix a SIZE_MAX macro redefinition warning
889
-
890
- A compilation fix.
891
-
892
- Will merge to 'master'.
893
-
894
-
895
-* et/add-chmod-x (2016-06-07) 1 commit
896
- (merged to 'next' on 2016-06-22 at 71d65a0)
897
- + add: add --chmod=+x / --chmod=-x options
898
-
899
- "git update-index --add --chmod=+x file" may be usable as an escape
900
- hatch, but not a friendly thing to force for people who do need to
901
- use it regularly. "git add --chmod=+x file" can be used instead.
902
-
903
- Will merge to 'master'.
904
-
905
-
1048
* mh/connect (2016-06-06) 10 commits
1049
- connect: [host:port] is legacy for ssh
1050
- connect: move ssh command line preparation to a separate function
@@ -918,8 +1060,6 @@ of the repositories listed at
1060
Rewrite Git-URL parsing routine (hopefully) without changing any
1061
behaviour.
1062
921
- Will merge to 'next'???
922
-
1063
1064
* va/i18n-even-more (2016-06-17) 38 commits
1065
- i18n: branch: mark comment when editing branch description for translation
@@ -1024,13 +1164,14 @@ of the repositories listed at
1164
1165
1166
* ep/http-curl-trace (2016-05-24) 2 commits
1027
- - imap-send.c: introduce the GIT_TRACE_CURL enviroment variable
1028
- - http.c: implement the GIT_TRACE_CURL environment variable
1167
+ (merged to 'next' on 2016-06-27 at c290515)
1168
+ + imap-send.c: introduce the GIT_TRACE_CURL enviroment variable
1169
+ + http.c: implement the GIT_TRACE_CURL environment variable
1170
1171
HTTP transport gained an option to produce more detailed debugging
1172
trace.
1173
1033
- Will merge to 'next'.
1174
+ Will merge to 'master'.
1175
1176
1177
* jc/attr (2016-05-25) 18 commits
@@ -1064,85 +1205,18 @@ of the repositories listed at
1205
back-burner the threading enhancement to another day.
1206
1207
1067
-* cc/apply-introduce-state (2016-06-06) 50 commits
1068
- (merged to 'next' on 2016-06-20 at 4f205b8)
1069
- + builtin/apply: remove misleading comment on lock_file field
1070
- + builtin/apply: move 'newfd' global into 'struct apply_state'
1071
- + builtin/apply: add 'lock_file' pointer into 'struct apply_state'
1072
- + builtin/apply: move applying patches into apply_all_patches()
1073
- + builtin/apply: move 'state' check into check_apply_state()
1074
- + builtin/apply: move 'symlink_changes' global into 'struct apply_state'
1075
- + builtin/apply: move 'fn_table' global into 'struct apply_state'
1076
- + builtin/apply: move 'state_linenr' global into 'struct apply_state'
1077
- + builtin/apply: move 'max_change' and 'max_len' into 'struct apply_state'
1078
- + builtin/apply: move 'ws_ignore_action' into 'struct apply_state'
1079
- + builtin/apply: move 'ws_error_action' into 'struct apply_state'
1080
- + builtin/apply: move 'applied_after_fixing_ws' into 'struct apply_state'
1081
- + builtin/apply: move 'squelch_whitespace_errors' into 'struct apply_state'
1082
- + builtin/apply: remove whitespace_option arg from set_default_whitespace_mode()
1083
- + builtin/apply: move 'whitespace_option' into 'struct apply_state'
1084
- + builtin/apply: move 'whitespace_error' global into 'struct apply_state'
1085
- + builtin/apply: move 'root' global into 'struct apply_state'
1086
- + builtin/apply: move 'p_value_known' global into 'struct apply_state'
1087
- + builtin/apply: move 'p_value' global into 'struct apply_state'
1088
- + builtin/apply: move 'has_include' global into 'struct apply_state'
1089
- + builtin/apply: move 'limit_by_name' global into 'struct apply_state'
1090
- + builtin/apply: move 'patch_input_file' global into 'struct apply_state'
1091
- + builtin/apply: move 'apply' global into 'struct apply_state'
1092
- + builtin/apply: move 'p_context' global into 'struct apply_state'
1093
- + builtin/apply: move 'fake_ancestor' global into 'struct apply_state'
1094
- + builtin/apply: move 'line_termination' global into 'struct apply_state'
1095
- + builtin/apply: move 'unsafe_paths' global into 'struct apply_state'
1096
- + builtin/apply: move 'no_add' global into 'struct apply_state'
1097
- + builtin/apply: move 'threeway' global into 'struct apply_state'
1098
- + builtin/apply: move 'summary' global into 'struct apply_state'
1099
- + builtin/apply: move 'numstat' global into 'struct apply_state'
1100
- + builtin/apply: move 'diffstat' global into 'struct apply_state'
1101
- + builtin/apply: move 'cached' global into 'struct apply_state'
1102
- + builtin/apply: move 'allow_overlap' global into 'struct apply_state'
1103
- + builtin/apply: move 'update_index' global into 'struct apply_state'
1104
- + builtin/apply: move 'apply_verbosely' global into 'struct apply_state'
1105
- + builtin/apply: move 'apply_with_reject' global into 'struct apply_state'
1106
- + builtin/apply: move 'apply_in_reverse' global into 'struct apply_state'
1107
- + builtin/apply: move 'check_index' global into 'struct apply_state'
1108
- + builtin/apply: move 'check' global into 'struct apply_state'
1109
- + builtin/apply: move 'unidiff_zero' global into 'struct apply_state'
1110
- + builtin/apply: move 'state' init into init_apply_state()
1111
- + builtin/apply: introduce 'struct apply_state' to start libifying
1112
- + builtin/apply: move 'read_stdin' global into cmd_apply()
1113
- + builtin/apply: move 'options' variable into cmd_apply()
1114
- + builtin/apply: extract line_by_line_fuzzy_match() from match_fragment()
1115
- + builtin/apply: avoid local variable shadowing 'len' parameter
1116
- + builtin/apply: avoid parameter shadowing 'linenr' global
1117
- + builtin/apply: avoid parameter shadowing 'p_value' global
1118
- + builtin/apply: make gitdiff_verify_name() return void
1119
-
1120
- Originally merged to 'next' on 2016-06-06
1121
-
1122
- The "git apply" standalone program is being libified; this is the
1123
- first step to move many state variables into a structure that can
1124
- be explicitly (re)initialized to make the machinery callable more
1125
- than once.
1126
-
1127
- The next step that moves some remaining state variables into the
1128
- structure and turns die()s into an error return that propagates up
1129
- to the caller is not queued yet but in flight. It would be good to
1130
- review the above first and give the remainder of the series a solid
1131
- base to build on.
1132
-
1133
- Will merge to 'master'.
1134
-
1135
-
1136
-* pb/bisect (2016-05-24) 3 commits
1208
+* pb/bisect (2016-06-27) 9 commits
1209
+ - bisect--helper: `bisect_write` shell function in C
1210
+ - bisect--helper: `is_expected_rev` & `check_expected_revs` shell function in C
1211
+ - bisect--helper: `bisect_reset` shell function in C
1212
+ - wrapper: move is_empty_file() and rename it as is_empty_or_missing_file()
1213
+ - t6030: explicitly test for bisection cleanup
1214
+ - bisect--helper: `bisect_clean_state` shell function in C
1215
- bisect--helper: `write_terms` shell function in C
1216
- bisect: rewrite `check_term_format` shell function in C
1217
- bisect--helper: use OPT_CMDMODE instead of OPT_BOOL
1218
1141
- Beginning of GSoC "git bisect" project.
1142
-
1143
- I know another topic is getting rerolled many times on top of this;
1144
- are people happy with these three patches? If so, will merge to
1145
- 'next'.
1219
+ GSoC "bisect" topic.
1220
1221
1222
* sb/pathspec-label (2016-06-03) 6 commits
@@ -1256,7 +1330,8 @@ of the repositories listed at
1330
This also really needs review.
1331
1332
1259
-* dt/index-helper (2016-05-20) 20 commits
1333
+* dt/index-helper (2016-06-27) 21 commits
1334
+ - unix-socket.c: add stub implementation when unix sockets are not supported
1335
- index-helper: indexhelper.exitafter config
1336
- trace: measure where the time is spent in the index-heavy operations
1337
- index-helper: optionally automatically run
@@ -1283,9 +1358,8 @@ of the repositories listed at
1358
optionally interface with the watchman daemon to further reduce the
1359
refresh cost.
1360
1286
- Belated review comments still trickling in.
1287
- Will hold.
1288
- ($gmane/295106).
1361
+ Expecting a reroll.
1362
+ ($gmane/298331, $gmane/298319).
1363
1364
1365
* jc/bundle (2016-03-03) 6 commits