What's cooking (2017/05 #03)
Junio C Hamano committed
May 10, 2017 at 14:17 UTC
813e7e553bda0475e51ec1b8334b9c0d387ac7a5
1 file changed
+326
-281
whats-cooking.txt
+326
-281
@@ -1,10 +1,10 @@
1
To: git@vger.kernel.org
2
Bcc: lwn@lwn.net
3
-Subject: What's cooking in git.git (May 2017, #02; Mon, 8)
4
-X-master-at: 4fa66c85f11bc5a541462ca5ae3246aa0ce02e74
5
-X-next-at: 57267f22770ebcdb20d119053d99e73c84a1c8da
3
+Subject: What's cooking in git.git (May 2017, #03; Wed, 10)
4
+X-master-at: b06d3643105c8758ed019125a4399cb7efdcce2c
5
+X-next-at: 4ebf3021692df4cb51da8d806fbb8b909ee7e111
6
7
-What's cooking in git.git (May 2017, #02; Mon, 8)
7
+What's cooking in git.git (May 2017, #03; Wed, 10)
8
--------------------------------------------------
9
10
Here are the topics that have been cooking. Commits prefixed with
@@ -12,6 +12,14 @@ 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
+Git 2.13, together with maintenance releases for 9 maintenance
16
+tracks, has been tagged. Let's wait for a few days to see if users
17
+find embarrassing regressions and then start the next cycle. In the
18
+meantime, please pay closer attention to the topics that have been
19
+stalled or waiting to be reviewed for too long and think of ways to
20
+help moving them forward, before you start sending your new shiny
21
+toys to the list.
22
+
23
You can find the changes described here in the integration branches
24
of the repositories listed at
25
@@ -20,52 +28,283 @@ of the repositories listed at
28
--------------------------------------------------
29
[Graduated to "master"]
30
23
-* ja/i18n-cleanup (2017-05-01) 2 commits
24
- (merged to 'next' on 2017-04-30 at 8002e53820)
25
- + i18n: read-cache: typofix
26
- + i18n: remove i18n from tag reflog message
31
+* jh/verify-index-checksum-only-in-fsck (2017-04-27) 1 commit
32
+ (merged to 'next' on 2017-04-28 at afb4c70061)
33
+ + t1450: avoid use of "sed" on the index, which is a binary file
34
+
35
+ Update an unportable constructin a new test.
36
+
37
+--------------------------------------------------
38
+[New Topics]
39
40
+* ab/compat-regex-update (2017-05-09) 2 commits
41
+ - compat/regex: update the gawk regex engine from upstream
42
+ - compat/regex: add a README with a maintenance guide
43
29
-* rg/a-the-typo (2017-05-01) 1 commit
30
- (merged to 'next' on 2017-04-30 at a648cf2961)
31
- + fix minor typos
44
+ Will merge to 'next'.
45
33
- Typofix.
46
47
+* jc/apply-fix-mismerge (2017-05-08) 1 commit
48
+ (merged to 'next' on 2017-05-09 at e0b89532d0)
49
+ + apply.c: fix whitespace-only mismerge
50
36
-* rg/doc-pull-typofix (2017-05-01) 1 commit
37
- (merged to 'next' on 2017-04-30 at c2edb4d813)
38
- + doc: git-pull.txt use US spelling, fix minor typo
51
+ Will merge to 'master'.
52
53
41
-* rg/doc-submittingpatches-wordfix (2017-05-01) 1 commit
42
- (merged to 'next' on 2017-04-30 at 4065036588)
43
- + doc: update SubmittingPatches
54
+* jt/push-options-doc (2017-05-10) 2 commits
55
+ - receive-pack: verify push options in cert
56
+ - docs: correct receive.advertisePushOptions default
57
58
+ The receive-pack program now makes sure that the push certificate
59
+ records the same set of push options used for pushing.
60
46
-* sr/hooks-cwd-doc (2017-05-01) 1 commit
47
- (merged to 'next' on 2017-04-30 at 063dd5cb02)
48
- + githooks.txt: clarify push hooks are always executed in $GIT_DIR
61
+ Will merge to 'next'.
62
+
63
+
64
+* dt/unpack-save-untracked-cache-extension (2017-05-09) 1 commit
65
+ - DONTMERGE: unpack-trees: preserve index extensions
66
67
--------------------------------------------------
51
-[New Topics]
68
+[Stalled]
69
+
70
+* mg/status-in-progress-info (2017-04-14) 1 commit
71
+ - status: show in-progress info for short status
72
+
73
+ "git status" learns an option to report various operations
74
+ (e.g. "merging") that the user is in the middle of.
75
+
76
+ It is still unclear if the participants decided that it is OK to
77
+ spell "--inprogress" as a single word.
78
+
79
+
80
+* mg/name-rev-debug (2017-03-31) 4 commits
81
+ - describe: pass --debug down to name-rev
82
+ - name-rev: provide debug output
83
+ - name-rev: favor describing with tags and use committer date to tiebreak
84
+ - name-rev: refactor logic to see if a new candidate is a better name
85
+
86
+ "git describe --debug --contains" did not add any meaningful
87
+ information, even though without "--contains" it did.
88
+
89
+ Have been expecting a reroll of the tip two, but it has not seen
90
+ any activity for too long.
91
+ cf. <xmqqshltxnwt.fsf@gitster.mtv.corp.google.com>
92
+
93
+
94
+* nd/worktree-move (2017-04-20) 6 commits
95
+ - worktree remove: new command
96
+ - worktree move: refuse to move worktrees with submodules
97
+ - worktree move: accept destination as directory
98
+ - worktree move: new command
99
+ - worktree.c: add update_worktree_location()
100
+ - worktree.c: add validate_worktree()
101
+
102
+ "git worktree" learned move and remove subcommands.
103
+
104
+ Expecting a reroll.
105
+ cf. <20170420101024.7593-1-pclouds@gmail.com>
106
+ cf. <20170421145916.mknekgqzhxffu7di@sigill.intra.peff.net>
107
+ cf. <d0e81b1e-5869-299e-f462-4d43dc997bd1@ramsayjones.plus.com>
108
+
109
+
110
+* df/dir-iter-remove-subtree (2017-04-19) 5 commits
111
+ - remove_subtree(): reimplement using iterators
112
+ - dir_iterator: rewrite state machine model
113
+ - dir_iterator: refactor dir_iterator_advance
114
+ - remove_subtree(): test removing nested directories
115
+ - dir_iterator: add tests for dir_iterator API
116
+
117
+ Update the dir-iterator API and use it to reimplement
118
+ remove_subtree().
119
+
120
+ A reroll exists that is based on the updated 'master', but I ran
121
+ out of time trying to get it to work with other topics in flight
122
+ in 'pu'.
123
+ GSoC microproject.
124
+
125
+
126
+* jc/bundle (2016-03-03) 6 commits
127
+ - index-pack: --clone-bundle option
128
+ - Merge branch 'jc/index-pack' into jc/bundle
129
+ - bundle v3: the beginning
130
+ - bundle: keep a copy of bundle file name in the in-core bundle header
131
+ - bundle: plug resource leak
132
+ - bundle doc: 'verify' is not about verifying the bundle
133
+
134
+ The beginning of "split bundle", which could be one of the
135
+ ingredients to allow "git clone" traffic off of the core server
136
+ network to CDN.
137
+
138
+ This was surrected from a "to be discarded" pile, as from time to
139
+ time people wonder about resumable clone that can be primed without
140
+ bothering Git servers with dynamic packfile creation, and some
141
+ people seem to think that the topic could serve as a useful
142
+ building block for that goal. But nothing seem to have happend.
143
+ Unless people really want it, I am inclined to discard this topic.
144
+ Opinions?
145
+
146
+
147
+* ja/doc-l10n (2017-03-20) 3 commits
148
+ . SQUASH???
149
+ . l10n: add git-add.txt to localized man pages
150
+ . l10n: introduce framework for localizing man pages
151
+
152
+ A proposal to use po4a to localize our manual pages.
153
+
154
+ Will discard.
155
+ Has been stalled for too long.
156
+
157
+
158
+* ls/filter-process-delayed (2017-03-06) 1 commit
159
+ - convert: add "status=delayed" to filter process protocol
160
+
161
+ What's the status of this one???
162
+ cf. <xmqq60jmnmef.fsf@junio-linux.mtv.corp.google.com>
163
+
164
+
165
+* jk/parse-options-no-no-no (2017-04-25) 2 commits
166
+ . SQUASH???
167
+ . parse-options: disallow double-negations of options starting with no-
168
+
169
+ Command line options that begin with "--no-" (e.g. "--no-checkout"
170
+ option of "git clone") can be negated by removing "--no-"; we
171
+ historically also allowed prefixing an extra "no" to the option
172
+ (e.g. "--no-no-checkout"), which made the command line look ugly
173
+ and unusual. This proposes to forbid it.
174
+
175
+ While I agree there is no need to support "--no-no-checkout", this
176
+ looks more like "if it looks ugly and unusual, you do not have to
177
+ use it". Perhaps we can drop it?
178
+
179
+ Will discard.
180
+
181
+
182
+* sk/dash-is-previous (2017-03-01) 5 commits
183
+ - revert.c: delegate handling of "-" shorthand to setup_revisions
184
+ - sha1_name.c: teach get_sha1_1 "-" shorthand for "@{-1}"
185
+ - revision.c: args starting with "-" might be a revision
186
+ - revision.c: swap if/else blocks
187
+ - revision.c: do not update argv with unknown option
188
+
189
+ A dash "-" can be written to mean "the branch that was previously
190
+ checked out" in more places.
191
+
192
+ Needs review.
193
+ cf. <1488007487-12965-1-git-send-email-kannan.siddharth12@gmail.com>
194
+
195
+
196
+* pb/bisect (2017-02-18) 28 commits
197
+ . fixup! bisect--helper: `bisect_next_check` & bisect_voc shell function in C
198
+ . bisect--helper: remove the dequote in bisect_start()
199
+ . bisect--helper: retire `--bisect-auto-next` subcommand
200
+ . bisect--helper: retire `--bisect-autostart` subcommand
201
+ . bisect--helper: retire `--bisect-write` subcommand
202
+ . bisect--helper: `bisect_replay` shell function in C
203
+ . bisect--helper: `bisect_log` shell function in C
204
+ . bisect--helper: retire `--write-terms` subcommand
205
+ . bisect--helper: retire `--check-expected-revs` subcommand
206
+ . bisect--helper: `bisect_state` & `bisect_head` shell function in C
207
+ . bisect--helper: `bisect_autostart` shell function in C
208
+ . bisect--helper: retire `--next-all` subcommand
209
+ . bisect--helper: retire `--bisect-clean-state` subcommand
210
+ . bisect--helper: `bisect_next` and `bisect_auto_next` shell function in C
211
+ . t6030: no cleanup with bad merge base
212
+ . bisect--helper: `bisect_start` shell function partially in C
213
+ . bisect--helper: `get_terms` & `bisect_terms` shell function in C
214
+ . bisect--helper: `bisect_next_check` & bisect_voc shell function in C
215
+ . bisect--helper: `check_and_set_terms` shell function in C
216
+ . bisect--helper: `bisect_write` shell function in C
217
+ . bisect--helper: `is_expected_rev` & `check_expected_revs` shell function in C
218
+ . bisect--helper: `bisect_reset` shell function in C
219
+ . wrapper: move is_empty_file() and rename it as is_empty_or_missing_file()
220
+ . t6030: explicitly test for bisection cleanup
221
+ . bisect--helper: `bisect_clean_state` shell function in C
222
+ . bisect--helper: `write_terms` shell function in C
223
+ . bisect: rewrite `check_term_format` shell function in C
224
+ . bisect--helper: use OPT_CMDMODE instead of OPT_BOOL
225
+
226
+ Move more parts of "git bisect" to C.
227
+
228
+ Will discard.
229
+ It has been in the stalled state for too long.
230
+ Expecting a reroll.
231
+ cf. <CAFZEwPPXPPHi8KiEGS9ggzNHDCGhuqMgH9Z8-Pf9GLshg8+LPA@mail.gmail.com>
232
+ cf. <CAFZEwPM9RSTGN54dzaw9gO9iZmsYjJ_d1SjUD4EzSDDbmh-XuA@mail.gmail.com>
233
+ cf. <CAFZEwPNUXcNY9Qdz=_B7q2kQuaecPzJtTMGdv8YMUPEz2vnp8A@mail.gmail.com>
234
+
235
+
236
+* sh/grep-tree-obj-tweak-output (2017-01-20) 2 commits
237
+ - grep: use '/' delimiter for paths
238
+ - grep: only add delimiter if there isn't one already
239
+
240
+ "git grep", when fed a tree-ish as an input, shows each hit
241
+ prefixed with "<tree-ish>:<path>:<lineno>:". As <tree-ish> is
242
+ almost always either a commit or a tag that points at a commit, the
243
+ early part of the output "<tree-ish>:<path>" can be used as the
244
+ name of the blob and given to "git show". When <tree-ish> is a
245
+ tree given in the extended SHA-1 syntax (e.g. "<commit>:", or
246
+ "<commit>:<dir>"), however, this results in a string that does not
247
+ name a blob (e.g. "<commit>::<path>" or "<commit>:<dir>:<path>").
248
+ "git grep" has been taught to be a bit more intelligent about these
249
+ cases and omit a colon (in the former case) or use slash (in the
250
+ latter case) to produce "<commit>:<path>" and
251
+ "<commit>:<dir>/<path>" that can be used as the name of a blob.
252
+
253
+ Will discard.
254
+ Nobody seems to be championing this change, after asked in all
255
+ recent "What's cooking" report:
256
+ "Is this good enough with known limitations?"
257
+
258
+
259
+* jc/diff-b-m (2015-02-23) 5 commits
260
+ . WIPWIP
261
+ . WIP: diff-b-m
262
+ . diffcore-rename: allow easier debugging
263
+ . diffcore-rename.c: add locate_rename_src()
264
+ . diffcore-break: allow debugging
265
+
266
+ "git diff -B -M" produced incorrect patch when the postimage of a
267
+ completely rewritten file is similar to the preimage of a removed
268
+ file; such a resulting file must not be expressed as a rename from
269
+ other place.
270
+
271
+ The fix in this patch is broken, unfortunately.
272
+
273
+ Will discard.
274
+
275
+--------------------------------------------------
276
+[Cooking]
277
+
278
+* sg/clone-refspec-from-command-line-config (2017-05-09) 2 commits
279
+ - git-clone --config order & fetching extra refs during initial clone
280
+ - clone: respect configured fetch respecs during initial fetch
281
+
282
+ "git clone -c var=val" is a way to set configuration variables in
283
+ the resulting repository, but it is more useful to also make these
284
+ variables take effect while the initial clone is happening,
285
+ e.g. these configuration variables could be fetch refspecs.
286
+
287
+ Will merge to 'next'.
288
+
289
290
* jn/clone-add-empty-config-from-command-line (2017-05-02) 1 commit
54
- - clone: handle empty config values in -c
291
+ (merged to 'next' on 2017-05-09 at c56ac3f782)
292
+ + clone: handle empty config values in -c
293
294
"git clone --config var=val" is a way to populate the
295
per-repository configuration file of the new repository, but it did
296
not work well when val is an empty string. This has been fixed.
297
60
- Will merge to 'next'.
298
+ Will merge to 'master'.
299
300
301
* jn/credential-doc-on-clear (2017-05-02) 1 commit
64
- - credential doc: make multiple-helper behavior more prominent
302
+ (merged to 'next' on 2017-05-09 at 96fa65d122)
303
+ + credential doc: make multiple-helper behavior more prominent
304
305
Doc update.
306
68
- Will merge to 'next'.
307
+ Will merge to 'master'.
308
309
310
* sb/checkout-recurse-submodules (2017-05-04) 3 commits
@@ -74,17 +313,18 @@ of the repositories listed at
313
- submodule_move_head: reuse child_process structure for futher commands
314
315
"git checkout --recurse-submodules" did not quite work with a
77
- submodule that itself has submodules.
316
+ submodule that itself has submodules.
317
318
Will merge to 'next'.
319
320
321
* ab/aix-needs-compat-regex (2017-05-04) 1 commit
83
- - config.mak.uname: set NO_REGEX=NeedsStartEnd on AIX
322
+ (merged to 'next' on 2017-05-09 at 881202b6bb)
323
+ + config.mak.uname: set NO_REGEX=NeedsStartEnd on AIX
324
325
Build fix.
326
87
- Will merge to 'next'.
327
+ Will merge to 'master'.
328
329
330
* ja/do-not-ask-needless-questions (2017-05-04) 3 commits
@@ -100,8 +340,8 @@ of the repositories listed at
340
this is not a good idea?
341
342
103
-* ab/doc-replace-gmane-links (2017-05-08) 2 commits
104
- - SQAUSH???
343
+* ab/doc-replace-gmane-links (2017-05-09) 2 commits
344
+ - doc: replace more gmane links
345
- doc: replace a couple of broken gmane links
346
347
The Web interface to gmane news archive is long gone, even though
@@ -110,6 +350,8 @@ of the repositories listed at
350
based on the actual message-id, it is likely that it will be easier
351
to migrate away from it if/when necessary.
352
353
+ Will merge to 'next'.
354
+
355
356
* ab/fix-poison-tests (2017-05-08) 3 commits
357
- travis-ci: add job to run tests with GETTEXT_POISON
@@ -147,10 +389,13 @@ of the repositories listed at
389
* jk/diff-submodule-diff-inline (2017-05-08) 1 commit
390
- diff: recurse into nested submodules for inline diff
391
150
- "git diff --submodule=diff" didn't recurse into nested submodules.
392
+ "git diff --submodule=diff" now recurses into nested submodules.
393
394
+ Will merge to 'next'.
395
153
-* jk/disable-pack-reuse-when-broken (2017-05-08) 1 commit
396
+
397
+* jk/disable-pack-reuse-when-broken (2017-05-09) 2 commits
398
+ - t5310: fix "; do" style
399
- pack-objects: disable pack reuse for object-selection options
400
401
"pack-objects" can stream a slice of an existing packfile out when
@@ -159,11 +404,12 @@ of the repositories listed at
404
strategy however would not work well when "--local" and other
405
options are in use, and need to be disabled.
406
407
+ Will merge to 'next'.
408
+
409
163
-* js/eol-on-ourselves (2017-05-08) 7 commits
410
+* js/eol-on-ourselves (2017-05-10) 6 commits
411
- t4051: mark supporting files as requiring LF-only line endings
412
- Fix the remaining tests that failed with core.autocrlf=true
166
- - t4003, t4005, t4007 & t4008: handle CR/LF in t/README & t/diff-lib/README
413
- t3901: move supporting files into t/t3901/
414
- completion: mark bash script as LF-only
415
- git-new-workdir: mark script as LF-only
@@ -175,6 +421,8 @@ of the repositories listed at
421
themselves that are meant to be run with /bin/sh, ought to be
422
checked out with eol=LF even on Windows.
423
424
+ Will merge to 'next'.
425
+
426
427
* nd/split-index-unshare (2017-05-08) 2 commits
428
- p3400: add perf tests for rebasing many changes
@@ -194,131 +442,15 @@ of the repositories listed at
442
A few codepaths in "checkout" and "am" working on an unborn branch
443
tried to access an uninitialized piece of memory.
444
197
---------------------------------------------------
198
-[Stalled]
199
-
200
-* ja/doc-l10n (2017-03-20) 3 commits
201
- - SQUASH???
202
- - l10n: add git-add.txt to localized man pages
203
- - l10n: introduce framework for localizing man pages
204
-
205
- A proposal to use po4a to localize our manual pages.
206
-
207
-
208
-* sg/clone-refspec-from-command-line-config (2017-02-27) 1 commit
209
- - clone: respect configured fetch respecs during initial fetch
210
-
211
- Expecting a reroll.
212
- cf. <20170227211217.73gydlxb2qu2sp3m@sigill.intra.peff.net>
213
- cf. <CAM0VKj=rsAfKvVccOMOoo5==Q1yW1U0zJBbUV=faKppWFm-u+g@mail.gmail.com>
214
- Some nits but looks ok.
215
-
216
-
217
-* sk/dash-is-previous (2017-03-01) 5 commits
218
- - revert.c: delegate handling of "-" shorthand to setup_revisions
219
- - sha1_name.c: teach get_sha1_1 "-" shorthand for "@{-1}"
220
- - revision.c: args starting with "-" might be a revision
221
- - revision.c: swap if/else blocks
222
- - revision.c: do not update argv with unknown option
223
-
224
- A dash "-" can be written to mean "the branch that was previously
225
- checked out" in more places.
226
-
227
- Needs review.
228
- cf. <1488007487-12965-1-git-send-email-kannan.siddharth12@gmail.com>
229
-
230
-
231
-* ls/filter-process-delayed (2017-03-06) 1 commit
232
- - convert: add "status=delayed" to filter process protocol
233
-
234
- What's the status of this one???
235
- cf. <xmqq60jmnmef.fsf@junio-linux.mtv.corp.google.com>
236
-
237
-
238
-* pb/bisect (2017-02-18) 28 commits
239
- . fixup! bisect--helper: `bisect_next_check` & bisect_voc shell function in C
240
- . bisect--helper: remove the dequote in bisect_start()
241
- . bisect--helper: retire `--bisect-auto-next` subcommand
242
- . bisect--helper: retire `--bisect-autostart` subcommand
243
- . bisect--helper: retire `--bisect-write` subcommand
244
- . bisect--helper: `bisect_replay` shell function in C
245
- . bisect--helper: `bisect_log` shell function in C
246
- . bisect--helper: retire `--write-terms` subcommand
247
- . bisect--helper: retire `--check-expected-revs` subcommand
248
- . bisect--helper: `bisect_state` & `bisect_head` shell function in C
249
- . bisect--helper: `bisect_autostart` shell function in C
250
- . bisect--helper: retire `--next-all` subcommand
251
- . bisect--helper: retire `--bisect-clean-state` subcommand
252
- . bisect--helper: `bisect_next` and `bisect_auto_next` shell function in C
253
- . t6030: no cleanup with bad merge base
254
- . bisect--helper: `bisect_start` shell function partially in C
255
- . bisect--helper: `get_terms` & `bisect_terms` shell function in C
256
- . bisect--helper: `bisect_next_check` & bisect_voc shell function in C
257
- . bisect--helper: `check_and_set_terms` shell function in C
258
- . bisect--helper: `bisect_write` shell function in C
259
- . bisect--helper: `is_expected_rev` & `check_expected_revs` shell function in C
260
- . bisect--helper: `bisect_reset` shell function in C
261
- . wrapper: move is_empty_file() and rename it as is_empty_or_missing_file()
262
- . t6030: explicitly test for bisection cleanup
263
- . bisect--helper: `bisect_clean_state` shell function in C
264
- . bisect--helper: `write_terms` shell function in C
265
- . bisect: rewrite `check_term_format` shell function in C
266
- . bisect--helper: use OPT_CMDMODE instead of OPT_BOOL
267
-
268
- Move more parts of "git bisect" to C.
269
-
270
- Ejected from 'pu', as it overlaps and gets in the way of bc/object-id topic.
271
- Expecting a reroll.
272
- cf. <CAFZEwPPXPPHi8KiEGS9ggzNHDCGhuqMgH9Z8-Pf9GLshg8+LPA@mail.gmail.com>
273
- cf. <CAFZEwPM9RSTGN54dzaw9gO9iZmsYjJ_d1SjUD4EzSDDbmh-XuA@mail.gmail.com>
274
- cf. <CAFZEwPNUXcNY9Qdz=_B7q2kQuaecPzJtTMGdv8YMUPEz2vnp8A@mail.gmail.com>
275
-
276
-
277
-* sh/grep-tree-obj-tweak-output (2017-01-20) 2 commits
278
- - grep: use '/' delimiter for paths
279
- - grep: only add delimiter if there isn't one already
280
-
281
- "git grep", when fed a tree-ish as an input, shows each hit
282
- prefixed with "<tree-ish>:<path>:<lineno>:". As <tree-ish> is
283
- almost always either a commit or a tag that points at a commit, the
284
- early part of the output "<tree-ish>:<path>" can be used as the
285
- name of the blob and given to "git show". When <tree-ish> is a
286
- tree given in the extended SHA-1 syntax (e.g. "<commit>:", or
287
- "<commit>:<dir>"), however, this results in a string that does not
288
- name a blob (e.g. "<commit>::<path>" or "<commit>:<dir>:<path>").
289
- "git grep" has been taught to be a bit more intelligent about these
290
- cases and omit a colon (in the former case) or use slash (in the
291
- latter case) to produce "<commit>:<path>" and
292
- "<commit>:<dir>/<path>" that can be used as the name of a blob.
293
-
294
- Expecting a reroll? Is this good enough with known limitations?
295
-
296
-
297
-* jc/diff-b-m (2015-02-23) 5 commits
298
- . WIPWIP
299
- . WIP: diff-b-m
300
- - diffcore-rename: allow easier debugging
301
- - diffcore-rename.c: add locate_rename_src()
302
- - diffcore-break: allow debugging
303
-
304
- "git diff -B -M" produced incorrect patch when the postimage of a
305
- completely rewritten file is similar to the preimage of a removed
306
- file; such a resulting file must not be expressed as a rename from
307
- other place.
445
+ Will merge to 'next'.
446
309
- The fix in this patch is broken, unfortunately.
310
-
311
- Will discard.
312
-
313
---------------------------------------------------
314
-[Cooking]
447
448
* bw/submodule-with-bs-path (2017-05-01) 1 commit
449
- t7400: add !CYGWIN prerequisite to 'add with \\ in path'
450
319
- A hotfix to a topic that is already in 'master'.
451
+ A hotfix to a topic that is already in v2.13.
452
321
- Wait for a few days for Acks and then merge to 'master'.
453
+ Will merge to 'next'.
454
455
456
* jk/am-leakfix (2017-04-27) 3 commits
@@ -330,11 +462,11 @@ of the repositories listed at
462
The codepath in "git am" that is used when running "git rebase"
463
leaked memory held for the log message of the commits being rebased.
464
333
- Will cook in 'next'.
465
+ Will merge to 'master'.
466
467
336
-* js/plug-leaks (2017-05-08) 26 commits
337
- - fixup! checkout: fix memory leak
468
+* js/plug-leaks (2017-05-09) 26 commits
469
+ - checkout: fix memory leak
470
- submodule_uses_worktrees(): plug memory leak
471
- show_worktree(): plug memory leak
472
- name-rev: avoid leaking memory in the `deref` case
@@ -361,9 +493,9 @@ of the repositories listed at
493
- winansi: avoid use of uninitialized value
494
- mingw: avoid memory leak when splitting PATH
495
364
- Fix memory leaks pointed out by Coverity.
496
+ Fix memory leaks pointed out by Coverity (and people).
497
366
- All looked sensible.
498
+ Will merge to 'next'.
499
500
501
* sk/status-short-branch-color-config (2017-04-28) 2 commits
@@ -375,7 +507,7 @@ of the repositories listed at
507
of the current branch and its remote-tracking branch are now
508
configurable.
509
378
- Will cook in 'next'.
510
+ Will merge to 'master'.
511
512
513
* jh/close-index-before-stat (2017-04-28) 1 commit
@@ -401,27 +533,26 @@ of the repositories listed at
533
534
Code clean-up and duplicate removal.
535
404
- Will cook in 'next'.
536
+ Will merge to 'master'.
537
538
539
* ls/travis-relays-for-windows-ci (2017-05-04) 2 commits
540
- travis-ci: retry if Git for Windows CI returns HTTP error 502 or 503
541
- travis-ci: handle Git for Windows CI status "failed" explicitly
542
411
- Will merge to 'next' after a few CI run in 'pu'.
543
+ Will merge to 'next'.
544
545
414
-* mb/diff-default-to-indent-heuristics (2017-05-02) 4 commits
415
- (merged to 'next' on 2017-05-08 at 158f401a92)
416
- + add--interactive: drop diff.indentHeuristic handling
417
- + diff: enable indent heuristic by default
418
- + diff: have the diff-* builtins configure diff before initializing revisions
419
- + diff: make the indent heuristic part of diff's basic configuration
546
+* mb/diff-default-to-indent-heuristics (2017-05-09) 4 commits
547
+ - add--interactive: drop diff.indentHeuristic handling
548
+ - diff: enable indent heuristic by default
549
+ - diff: have the diff-* builtins configure diff before initializing revisions
550
+ - diff: make the indent heuristic part of diff's basic configuration
551
552
Make the "indent" heuristics the default in "diff" and diff.indentHeuristics
553
configuration variable an escape hatch for those who do no want it.
554
424
- Will cook in 'next'.
555
+ Kicked out of next; it seems it is still getting review suggestions?
556
557
558
* tb/dedup-crlf-tests (2017-05-01) 1 commit
@@ -436,14 +567,7 @@ of the repositories listed at
567
"git repack" learned to accept the --threads=<n> option and pass it
568
to pack-objects.
569
439
-
440
-* jh/verify-index-checksum-only-in-fsck (2017-04-27) 1 commit
441
- (merged to 'next' on 2017-04-28 at afb4c70061)
442
- + t1450: avoid use of "sed" on the index, which is a binary file
443
-
444
- Update an unportable constructin a new test.
445
-
446
- Will cook in 'next'.
570
+ Will merge to 'next'.
571
572
573
* dt/gc-ignore-old-gc-logs (2017-04-24) 3 commits
@@ -455,10 +579,10 @@ of the repositories listed at
579
An attempt to allow us notice "fishy" situation where we fail to
580
remove the temporary directory used during the test.
581
458
- Will cook in 'next'.
582
+ Will merge to 'master'.
583
584
461
-* nd/fopen-errors (2017-05-04) 21 commits
585
+* nd/fopen-errors (2017-05-09) 23 commits
586
- t1308: add a test case on open a config directory
587
- config.c: handle error on failing to fopen()
588
- xdiff-interface.c: report errno on failure to stat() or fopen()
@@ -477,18 +601,16 @@ of the repositories listed at
601
- attr.c: use fopen_or_warn()
602
- wrapper.c: add fopen_or_warn()
603
- wrapper.c: add warn_on_fopen_errors()
604
+ - config.mak.uname: set FREAD_READS_DIRECTORIES for Darwin, too
605
- config.mak.uname: set FREAD_READS_DIRECTORIES for Linux and FreeBSD
606
- clone: use xfopen() instead of fopen()
607
- Use xfopen() in more places
608
+ - git_fopen: fix a sparse 'not declared' warning
609
610
We often try to open a file for reading whose existence is
611
optional, and silently ignore errors from open/fopen; report such
612
errors if they are not due to missing files.
613
488
- Need to resurrect "git_fopen: fix a sparse 'not declared' warning".
489
- It also seems to cause a test failure on MacOS X.
490
- e.g. https://travis-ci.org/git/git/jobs/229585099#L3091
491
-
614
615
* dt/raise-core-packed-git-limit (2017-04-20) 1 commit
616
(merged to 'next' on 2017-04-26 at c72dd8c62f)
@@ -498,11 +620,12 @@ of the repositories listed at
620
avoid "git fetch" from a (recoverable) failure while "gc" is
621
running in parallel.
622
501
- Will cook in 'next'.
623
+ Will merge to 'master'.
624
625
504
-* js/larger-timestamps (2017-05-04) 9 commits
505
- - archive-tar: fix a sparse 'constant too large' warning
626
+* js/larger-timestamps (2017-05-09) 9 commits
627
+ (merged to 'next' on 2017-05-09 at ae0603fd3e)
628
+ + archive-tar: fix a sparse 'constant too large' warning
629
(merged to 'next' on 2017-04-28 at b56a0d38cd)
630
+ use uintmax_t for timestamps
631
+ date.c: abort if the system time cannot handle one of our timestamps
@@ -521,9 +644,7 @@ of the repositories listed at
644
move), and then declare uintmax_t is the type to be used as the
645
timestamp_t
646
524
- The tip one from Ramsay needs to be replaced with a newer one.
525
- cf. <70b431dc-c5cc-e376-b63e-6a58ee1385f1@ramsayjones.plus.com>
526
- Will cook in 'next'.
647
+ Will merge to 'master'.
648
649
650
* ab/grep-pcre-v2 (2017-05-02) 19 commits
@@ -550,6 +671,9 @@ of the repositories listed at
671
PCRE2, which has an API different from and incompatible with PCRE,
672
can now be chosen to support "grep -P -e '<pattern>'" and friends.
673
674
+ Need to make sure that fallback for those without pcre2 works
675
+ without causing build failure.
676
+
677
678
* bc/object-id (2017-05-08) 53 commits
679
- object: convert parse_object* to take struct object_id
@@ -618,7 +742,7 @@ of the repositories listed at
742
Tighten code to update cache-tree so that we won't accidentally
743
write out any 0{40} entry in the tree object.
744
621
- Will cook in 'next'.
745
+ Will merge to 'master'.
746
747
748
* rs/large-zip (2017-05-01) 7 commits
@@ -635,7 +759,7 @@ of the repositories listed at
759
"git archive --format=zip" learned to use zip64 extension when
760
necessary to go beyond the 4GB limit.
761
638
- Will cook in 'next'.
762
+ Will merge to 'master'.
763
764
765
* jc/checkout-working-tree-only (2017-04-27) 1 commit
@@ -644,6 +768,10 @@ of the repositories listed at
768
"git checkout <tree-ish> <pathspec>" learned a variant that does
769
not update the index when doing its thing.
770
771
+ This was more of "the world could have been like this" illustration
772
+ rather than a "let's make this change" proposal. Unless people
773
+ really want it, I am inclined to discard this topic. Opinions?
774
+
775
776
* js/rebase-i-final (2017-05-02) 10 commits
777
- rebase -i: rearrange fixup/squash lines using the rebase--helper
@@ -672,56 +800,7 @@ of the repositories listed at
800
This has been fixed so that an additional sign-off etc. are added
801
after completing the existing incomplete line.
802
675
- Will cook in 'next'.
676
-
677
-
678
-* nd/worktree-move (2017-04-20) 6 commits
679
- - worktree remove: new command
680
- - worktree move: refuse to move worktrees with submodules
681
- - worktree move: accept destination as directory
682
- - worktree move: new command
683
- - worktree.c: add update_worktree_location()
684
- - worktree.c: add validate_worktree()
685
-
686
- "git worktree" learned move and remove subcommands.
687
-
688
- Expecting a reroll.
689
- cf. <20170420101024.7593-1-pclouds@gmail.com>
690
- cf. <20170421145916.mknekgqzhxffu7di@sigill.intra.peff.net>
691
- cf. <d0e81b1e-5869-299e-f462-4d43dc997bd1@ramsayjones.plus.com>
692
-
693
-
694
-* jk/parse-options-no-no-no (2017-04-25) 2 commits
695
- - SQUASH???
696
- - parse-options: disallow double-negations of options starting with no-
697
-
698
- Command line options that begin with "--no-" (e.g. "--no-checkout"
699
- option of "git clone") can be negated by removing "--no-"; we
700
- historically also allowed prefixing an extra "no" to the option
701
- (e.g. "--no-no-checkout"), which made the command line look ugly
702
- and unusual. This proposes to forbid it.
703
-
704
- While I agree there is no need to support "--no-no-checkout", this
705
- looks more like "if it looks ugly and unusual, you do not have to
706
- use it". Perhaps we can drop it?
707
-
708
- Will discard.
709
-
710
-
711
-* df/dir-iter-remove-subtree (2017-04-19) 5 commits
712
- - remove_subtree(): reimplement using iterators
713
- - dir_iterator: rewrite state machine model
714
- - dir_iterator: refactor dir_iterator_advance
715
- - remove_subtree(): test removing nested directories
716
- - dir_iterator: add tests for dir_iterator API
717
-
718
- Update the dir-iterator API and use it to reimplement
719
- remove_subtree().
720
-
721
- A reroll exists that is based on the updated 'master', but I ran
722
- out of time trying to get it to work with other topics in flight
723
- in 'pu'.
724
- GSoC microproject.
803
+ Will merge to 'master'.
804
805
806
* ab/clone-no-tags (2017-05-01) 3 commits
@@ -734,7 +813,7 @@ of the repositories listed at
813
initially, and also set up the tagopt not to follow any tags in
814
subsequent fetches.
815
737
- Will cook in 'next'.
816
+ Will merge to 'master'.
817
818
819
* bw/forking-and-threading (2017-04-25) 13 commits
@@ -757,7 +836,7 @@ of the repositories listed at
836
The "run-command" API implementation has been made more robust
837
against dead-locking in a threaded environment.
838
760
- Will cook in 'next'.
839
+ Will merge to 'master'.
840
841
842
* sb/reset-recurse-submodules (2017-04-23) 4 commits
@@ -768,6 +847,8 @@ of the repositories listed at
847
848
"git reset" learned "--recurse-submodules" option.
849
850
+ Will merge to 'next'.
851
+
852
853
* ls/travis-doc-asciidoctor (2017-04-26) 4 commits
854
(merged to 'next' on 2017-05-04 at 813ba54fc3)
@@ -780,17 +861,7 @@ of the repositories listed at
861
Have Travis CI format the documentation with both AsciiDoc and
862
AsciiDoctor.
863
783
- Will cook in 'next'.
784
-
785
-
786
-* mg/status-in-progress-info (2017-04-14) 1 commit
787
- - status: show in-progress info for short status
788
-
789
- "git status" learns an option to report various operations
790
- (e.g. "merging") that the user is in the middle of.
791
-
792
- It is still unclear if the participants decided that it is OK to
793
- spell "--inprogress" as a single word.
864
+ Will merge to 'master'.
865
866
867
* ab/grep-threading-cleanup (2017-04-16) 8 commits
@@ -826,19 +897,6 @@ of the repositories listed at
897
Will merge to 'next'.
898
899
829
-* mg/name-rev-debug (2017-03-31) 4 commits
830
- - describe: pass --debug down to name-rev
831
- - name-rev: provide debug output
832
- - name-rev: favor describing with tags and use committer date to tiebreak
833
- - name-rev: refactor logic to see if a new candidate is a better name
834
-
835
- "git describe --debug --contains" did not add any meaningful
836
- information, even though without "--contains" it did.
837
-
838
- Expecting a reroll of the tip two.
839
- cf. <xmqqshltxnwt.fsf@gitster.mtv.corp.google.com>
840
-
841
-
900
* nd/prune-in-worktree (2017-04-24) 12 commits
901
- rev-list: expose and document --single-worktree
902
- revision.c: --reflog add HEAD reflog from all worktrees
@@ -878,17 +936,4 @@ of the repositories listed at
936
"git gc" did not interact well with "git worktree"-managed
937
per-worktree refs.
938
881
- Will cook in 'next'.
882
-
883
-
884
-* jc/bundle (2016-03-03) 6 commits
885
- - index-pack: --clone-bundle option
886
- - Merge branch 'jc/index-pack' into jc/bundle
887
- - bundle v3: the beginning
888
- - bundle: keep a copy of bundle file name in the in-core bundle header
889
- - bundle: plug resource leak
890
- - bundle doc: 'verify' is not about verifying the bundle
891
-
892
- The beginning of "split bundle", which could be one of the
893
- ingredients to allow "git clone" traffic off of the core server
894
- network to CDN.
939
+ Will merge to 'master'.