What's cooking (2023/08 #02)
Junio C Hamano committed
Aug 4, 2023 at 12:07 UTC
0e87def28284219b77cc00671680bedaa26de158
1 file changed
+228
-348
whats-cooking.txt
+228
-348
@@ -1,10 +1,10 @@
1
To: git@vger.kernel.org
2
-Subject: What's cooking in git.git (Aug 2023, #01; Wed, 2)
3
-X-master-at: 1b0a5129563ebe720330fdc8f5c6843d27641137
4
-X-next-at: ed35a5117f890568bcd8290b0000b7550ab351b8
2
+Subject: What's cooking in git.git (Aug 2023, #02; Fri, 4)
3
+X-master-at: ac83bc5054c2ac489166072334b4147ce6d0fccb
4
+X-next-at: 6953b252045fa7a4c836b56ff7a46e3e85c8275b
5
Bcc: lwn@lwn.net, gitster@pobox.com
6
7
-What's cooking in git.git (Aug 2023, #01; Wed, 2)
7
+What's cooking in git.git (Aug 2023, #02; Fri, 4)
8
--------------------------------------------------
9
10
Here are the topics that have been cooking in my tree. Commits
@@ -17,14 +17,8 @@ topic without enough support may be discarded after a long period of
17
no activity (of course they can be resubmit when new interests
18
arise).
19
20
-We are getting closer to the final phase of this cycle, which begins
21
-when -rc0 preview release is tagged this coming Friday, followed by
22
-about 1 1/2 weeks of stabilization period that begins when -rc1 is
23
-tagged (cf. tinyurl.com/gitCal). There are a handful of topics that
24
-still need reviews before getting merged to 'next', but because the
25
-summer in the northern hemisphere is historically a slower season,
26
-too few reviewers seem to be active, relative to the number of these
27
-topics.
20
+Git 2.42-rc0 has been tagged. Hopefully people can help find and
21
+fix plenty of regressions before we tag -rc1 mid next week ;-)
22
23
Copies of the source code to Git live in many repositories, and the
24
following is a list of the ones I push into or their mirrors. Some
@@ -57,111 +51,191 @@ Release tarballs are available at:
51
--------------------------------------------------
52
[Graduated to 'master']
53
60
-* ah/autoconf-fixes (2023-07-19) 3 commits
61
- (merged to 'next' on 2023-07-25 at 35ff66e0cb)
62
- + configure.ac: always save NO_ICONV to config.status
63
- + configure.ac: don't overwrite NO_CURL option
64
- + configure.ac: don't overwrite NO_EXPAT option
54
+* hy/blame-in-bare-with-contents (2023-07-21) 1 commit
55
+ (merged to 'next' on 2023-07-31 at 39ac96d8d8)
56
+ + blame: allow --contents to work with bare repo
57
+
58
+ "git blame --contents=file" has been taught to work in a bare
59
+ repository.
60
+ source: <20230721035758.61956-1-hanyang.tony@bytedance.com>
61
+
62
+
63
+* ja/worktree-orphan-fix (2023-07-26) 3 commits
64
+ (merged to 'next' on 2023-07-27 at e475016065)
65
+ + t2400: rewrite regex to avoid unintentional PCRE
66
+ + builtin/worktree.c: convert tab in advice to space
67
+ + t2400: drop no-op `--sq` from rev-parse call
68
+
69
+ Fix tests with unportable regex patterns.
70
+ source: <20230726214202.15775-1-jacobabel@nullpo.dev>
71
66
- "./configure --with-expat=no" did not work as a way to refuse use
67
- of the expat library on a system with the library installed, which
68
- has been corrected.
69
- source: <20230719145211.17854-2-aherrmann@suse.de>
72
73
+* jc/branch-in-use-error-message (2023-07-21) 1 commit
74
+ (merged to 'next' on 2023-07-31 at 22f17d131b)
75
+ + branch: update the message to refuse touching a branch in-use
76
72
-* ah/sequencer-rewrite-todo-fix (2023-07-24) 1 commit
73
- (merged to 'next' on 2023-07-26 at 24e74d9eda)
74
- + sequencer: finish parsing the todo list despite an invalid first line
77
+ "git branch -f X" to repoint the branch X said that X was "checked
78
+ out" in another worktree, even when branch X was not and instead
79
+ being bisected or rebased. The message was reworded to say the
80
+ branch was "in use".
81
+ source: <xmqqr0p1szhz.fsf_-_@gitster.g>
82
76
- When the user edits "rebase -i" todo file so that it starts with a
77
- "fixup", which would make it invalid, the command truncated the
78
- rest of the file before giving an error and returning the control
79
- back to the user. Stop truncating to make it easier to correct
80
- such a malformed todo file.
81
- cf. <https://lore.kernel.org/git/0d1c5bfd-3ae5-83f0-a333-bbb8510a973a@gmail.com/>
82
- source: <20230722212830.132135-2-alexhenrie24@gmail.com>
83
84
+* jc/doc-sent-patch-now-what (2023-07-27) 1 commit
85
+ (merged to 'next' on 2023-07-31 at 51f5d9d465)
86
+ + MyFirstContribution: refrain from self-iterating too much
87
85
-* bb/use-trace2-counters-for-fsync-stats (2023-07-20) 1 commit
86
- (merged to 'next' on 2023-07-26 at f2c2e3f2b9)
87
- + wrapper: use trace2 counters to collect fsync stats
88
+ Process document update.
89
+ source: <xmqqmszg987u.fsf_-_@gitster.g>
90
89
- Instead of inventing a custom counter variables for debugging,
90
- use existing trace2 facility in the fsync customization codepath.
91
- source: <20230720164823.625815-1-dev+git@drbeat.li>
91
92
+* jc/parse-options-short-help (2023-07-19) 3 commits
93
+ (merged to 'next' on 2023-07-31 at e076d1f497)
94
+ + short help: allow a gap smaller than USAGE_GAP
95
+ + remote: simplify "remote add --tags" help text
96
+ + short help: allow multi-line opthelp
97
94
-* jc/tree-walk-drop-base-offset (2023-07-07) 2 commits
95
- (merged to 'next' on 2023-07-25 at cc050c60a6)
96
- + tree-walk: drop unused base_offset from do_match()
97
- + tree-walk: lose base_offset that is never used in tree_entry_interesting
98
+ Command line parser fix, and a small parse-options API update.
99
+ source: <xmqq5y6gg8fn.fsf@gitster.g>
100
99
- Code simplification.
100
- source: <20230707222116.4129415-1-gitster@pobox.com>
101
102
+* jc/retire-get-sha1-hex (2023-07-24) 1 commit
103
+ (merged to 'next' on 2023-07-27 at eeb9cc37f5)
104
+ + hex: retire get_sha1_hex()
105
103
-* ks/ref-filter-describe (2023-07-24) 2 commits
104
- (merged to 'next' on 2023-07-26 at f4b3b3b7ef)
105
- + ref-filter: add new "describe" atom
106
- + ref-filter: add multiple-option parsing functions
106
+ The implementation of "get_sha1_hex()" that reads a hexadecimal
107
+ string that spells a full object name has been extended to cope
108
+ with any hash function used in the repository, but the "sha1" in
109
+ its name survived. Rename it to get_hash_hex(), a name that is
110
+ more consistent within its friends like get_hash_hex_algop().
111
+ source: <xmqq1qgwoqgo.fsf_-_@gitster.g>
112
108
- "git branch --list --format=<format>" and friends are taught
109
- a new "%(describe)" placeholder.
110
- source: <20230723162717.68123-1-five231003@gmail.com>
113
+
114
+* la/doc-choose-starting-point (2023-07-14) 5 commits
115
+ (merged to 'next' on 2023-07-19 at 5a807cae46)
116
+ + SubmittingPatches: simplify guidance for choosing a starting point
117
+ + SubmittingPatches: emphasize need to communicate non-default starting points
118
+ + SubmittingPatches: de-emphasize branches as starting points
119
+ + SubmittingPatches: discuss subsystems separately from git.git
120
+ + SubmittingPatches: reword awkward phrasing
121
+ (this branch is used by la/doc-choose-starting-point-fixup.)
122
+
123
+ Clarify how to choose the starting point for a new topic in
124
+ developer guidance document.
125
+
126
+ Will merge to 'master' together with the follow-on topic.
127
+ source: <pull.1556.v2.git.1689314493.gitgitgadget@gmail.com>
128
+
129
+
130
+* la/doc-choose-starting-point-fixup (2023-07-27) 3 commits
131
+ (merged to 'next' on 2023-07-28 at 047dcae31c)
132
+ + SubmittingPatches: use of older maintenance tracks is an exception
133
+ + SubmittingPatches: explain why 'next' and above are inappropriate base
134
+ + SubmittingPatches: choice of base for fixing an older maintenance track
135
+ (this branch uses la/doc-choose-starting-point.)
136
+
137
+ Clarify how to pick a starting point for a new topic in the
138
+ SubmittingPatches document.
139
+
140
+ Will merge to 'master', together with the underlying topic.
141
+ source: <pull.1556.v2.git.1689314493.gitgitgadget@gmail.com>
142
+ source: <pull.1556.v3.git.1690340701.gitgitgadget@gmail.com>
143
+
144
+
145
+* pv/doc-submodule-update-settings (2023-07-25) 1 commit
146
+ (merged to 'next' on 2023-07-27 at e27b5b7ba8)
147
+ + doc: highlight that .gitmodules does not support !command
148
+
149
+ Rewrite the description of giving a custom command to the
150
+ submodule.<name>.update configuration variable.
151
+ source: <20230725212218.711116-1-pvutov@imap.cc>
152
153
--------------------------------------------------
154
[New Topics]
155
115
-* bc/ident-dot-is-no-longer-crud-letter (2023-08-02) 1 commit
116
- - ident: don't consider '.' a crud
156
+* jc/rerere-cleanup (2023-08-03) 5 commits
157
+ - rerere: plug small strbuf leak
158
+ - rerere: modernize use of empty strbuf
159
+ - rerere: try_merge() should use LL_MERGE_ERROR when it means an error
160
+ - rerere: fix comment on handle_file() helper
161
+ - rerere: simplify check_one_conflict() helper function
162
+ (this branch uses jc/unresolve-removal.)
163
118
- Exclude "." from the set of characters to be removed from the
119
- beginning and the end of the human-readable name.
164
+ Code clean-up.
165
121
- Will merge to 'next'?
122
- source: <xmqqsf918k4j.fsf@gitster.g>
166
+ Not ready to be reviewed yet.
167
+ source: <20230731224409.4181277-1-gitster@pobox.com>
168
169
125
-* jc/unresolve-removal (2023-07-31) 7 commits
126
- - checkout: allow "checkout -m path" to unmerge removed paths
127
- - checkout/restore: add basic tests for --merge
128
- - checkout/restore: refuse unmerging paths unless checking out of the index
129
- - update-index: remove stale fallback code for "--unresolve"
130
- - update-index: use unmerge_index_entry() to support removal
131
- - resolve-undo: allow resurrecting conflicted state that resolved to deletion
132
- - update-index: do not read HEAD and MERGE_HEAD unconditionally
170
+* ma/locate-in-path-for-windows (2023-08-04) 3 commits
171
+ - docs: update when `git bisect visualize` uses `gitk`
172
+ - compat/mingw: implement a native locate_in_PATH()
173
+ - run-command: conditionally define locate_in_PATH()
174
134
- "checkout --merge -- path" and "update-index --unresolve path" did
135
- not resurrect conflicted state that was resolved to remove path,
136
- but now they do.
175
+ "git bisect visualize" stopped running "gitk" on Git for Windows
176
+ when the command was reimplemented in C around Git 2.34 timeframe.
177
+ This has been corrected.
178
138
- Needs review.
139
- source: <20230731224409.4181277-1-gitster@pobox.com>
179
+ Will merge to 'next'.
180
+ source: <pull.1560.v2.git.1691122124.gitgitgadget@gmail.com>
181
182
142
-* ew/hash-with-openssl-evp (2023-08-01) 2 commits
143
- - avoid SHA-1 functions deprecated in OpenSSL 3+
144
- - sha256: avoid functions deprecated in OpenSSL 3+
183
+* pw/rebase-skip-commit-message-fix (2023-08-03) 1 commit
184
+ - rebase --skip: fix commit message clean up when skipping squash
185
146
- Adjust to OpenSSL 3+, which deprecates its SHA-1 functions based on
147
- its traditional API, by using its EVP API instead.
186
+ "git rebase -i" with a series of squash/fixup, when one of the
187
+ steps stopped in conflicts and ended up getting skipped, did not
188
+ handle the accumulated commit log messages, which has been
189
+ corrected.
190
149
- Will merge to 'next'.
150
- source: <20230801025454.1137802-1-e@80x24.org>
191
+ Will merge to 'next'.
192
+ source: <pull.1558.git.git.1691068176051.gitgitgadget@gmail.com>
193
194
153
-* rj/status-bisect-while-rebase (2023-08-01) 1 commit
154
- - status: fix branch shown when not only bisecting
195
+* bc/ignore-cland-cache (2023-08-04) 1 commit
196
+ (merged to 'next' on 2023-08-04 at 5cf8d41911)
197
+ + gitignore: ignore clangd .cache directory
198
156
- "git status" is taught to show both the branch being bisected and
157
- being rebased when both are in effect at the same time.
199
+ .gitignore update.
200
159
- Needs review.
160
- source: <48745298-f12b-8efb-4e48-90d2c22a8349@gmail.com>
201
+ Will merge to 'master'.
202
+ source: <20230804171328.1737188-1-sandals@crustytoothpaste.net>
203
204
--------------------------------------------------
205
[Stalled]
206
207
+* mh/credential-libsecret-attrs (2023-06-16) 1 commit
208
+ - credential/libsecret: store new attributes
209
+
210
+ The way authentication related data other than passwords (e.g.
211
+ oath token and password expiration data) are stored in libsecret
212
+ keyrings has been rethought.
213
+
214
+ Needs review.
215
+ source: <pull.1469.v5.git.git.1686945306242.gitgitgadget@gmail.com>
216
+
217
+
218
+* cc/git-replay (2023-06-03) 15 commits
219
+ - replay: stop assuming replayed branches do not diverge
220
+ - replay: add --contained to rebase contained branches
221
+ - replay: add --advance or 'cherry-pick' mode
222
+ - replay: disallow revision specific options and pathspecs
223
+ - replay: use standard revision ranges
224
+ - replay: make it a minimal server side command
225
+ - replay: remove HEAD related sanity check
226
+ - replay: remove progress and info output
227
+ - replay: add an important FIXME comment about gpg signing
228
+ - replay: don't simplify history
229
+ - replay: introduce pick_regular_commit()
230
+ - replay: die() instead of failing assert()
231
+ - replay: start using parse_options API
232
+ - replay: introduce new builtin
233
+ - t6429: remove switching aspects of fast-rebase
234
+
235
+ No reviews?
236
+ source: <20230602102533.876905-1-christian.couder@gmail.com>
237
+
238
+
239
* tk/cherry-pick-sequence-requires-clean-worktree (2023-06-01) 1 commit
240
- cherry-pick: refuse cherry-pick sequence if index is dirty
241
@@ -187,19 +261,8 @@ Release tarballs are available at:
261
Stalled for too long.
262
source: <cover-v2-0.3-00000000000-20221230T011725Z-avarab@gmail.com>
263
190
-
191
-* ob/revert-of-revert (2023-05-05) 1 commit
192
- - sequencer: beautify subject of reverts of reverts
193
-
194
- Instead of "Revert "Revert "original"", give "Reapply "original""
195
- as the title for a revert of a revert.
196
-
197
- Will discard.
198
- Have been expecting a hopefully final reroll for too long.
199
- Looking much better, except for minor cosmetic issues.
200
- cf. <xmqqmt21txid.fsf@gitster.g>
201
- source: <20230428083528.1699221-1-oswald.buddenhagen@gmx.de>
202
-
264
+--------------------------------------------------
265
+[Cooking]
266
267
* pw/rebase-i-after-failure (2023-08-01) 7 commits
268
- rebase -i: fix adding failed command to the todo list
@@ -213,13 +276,63 @@ Release tarballs are available at:
276
Various fixes to the behaviour of "rebase -i" when the command got
277
interrupted by conflicting changes.
278
216
- Will merge to 'next'?
217
- cf. <xmqqa5vad6ea.fsf@gitster.g>
218
- cf. <xmqq5y5yd6d7.fsf@gitster.g>
279
+ Expecting a reroll.
280
+ The latter half of the series should be reviewed by those who have
281
+ more stake in the sequencer code than myself.
282
+ cf. <619e458b-218b-a790-dfb4-9200e201b513@gmail.com>
283
source: <pull.1492.v3.git.1690903412.gitgitgadget@gmail.com>
284
221
---------------------------------------------------
222
-[Cooking]
285
+
286
+* bc/ident-dot-is-no-longer-crud-letter (2023-08-02) 1 commit
287
+ (merged to 'next' on 2023-08-03 at e786442a9b)
288
+ + ident: don't consider '.' a crud
289
+
290
+ Exclude "." from the set of characters to be removed from the
291
+ beginning and the end of the human-readable name.
292
+
293
+ Will merge to 'master'.
294
+ source: <xmqqsf918k4j.fsf@gitster.g>
295
+
296
+
297
+* jc/unresolve-removal (2023-07-31) 7 commits
298
+ - checkout: allow "checkout -m path" to unmerge removed paths
299
+ - checkout/restore: add basic tests for --merge
300
+ - checkout/restore: refuse unmerging paths unless checking out of the index
301
+ - update-index: remove stale fallback code for "--unresolve"
302
+ - update-index: use unmerge_index_entry() to support removal
303
+ - resolve-undo: allow resurrecting conflicted state that resolved to deletion
304
+ - update-index: do not read HEAD and MERGE_HEAD unconditionally
305
+ (this branch is used by jc/rerere-cleanup.)
306
+
307
+ "checkout --merge -- path" and "update-index --unresolve path" did
308
+ not resurrect conflicted state that was resolved to remove path,
309
+ but now they do.
310
+
311
+ Needs review.
312
+ source: <20230731224409.4181277-1-gitster@pobox.com>
313
+
314
+
315
+* ew/hash-with-openssl-evp (2023-08-01) 2 commits
316
+ (merged to 'next' on 2023-08-02 at 996db74865)
317
+ + avoid SHA-1 functions deprecated in OpenSSL 3+
318
+ + sha256: avoid functions deprecated in OpenSSL 3+
319
+
320
+ Adjust to OpenSSL 3+, which deprecates its SHA-1 functions based on
321
+ its traditional API, by using its EVP API instead.
322
+
323
+ Will merge to 'master'.
324
+ source: <20230801025454.1137802-1-e@80x24.org>
325
+
326
+
327
+* rj/status-bisect-while-rebase (2023-08-01) 1 commit
328
+ - status: fix branch shown when not only bisecting
329
+
330
+ "git status" is taught to show both the branch being bisected and
331
+ being rebased when both are in effect at the same time.
332
+
333
+ Needs review.
334
+ source: <48745298-f12b-8efb-4e48-90d2c22a8349@gmail.com>
335
+
336
337
* ew/sha256-gcrypt-leak-fixes (2023-07-31) 3 commits
338
(merged to 'next' on 2023-08-01 at eed83801c3)
@@ -243,32 +356,6 @@ Release tarballs are available at:
356
source: <2dcb915f-b926-e024-6394-23aff200955c@web.de>
357
358
246
-* pv/doc-submodule-update-settings (2023-07-25) 1 commit
247
- (merged to 'next' on 2023-07-27 at e27b5b7ba8)
248
- + doc: highlight that .gitmodules does not support !command
249
-
250
- Rewrite the description of giving a custom command to the
251
- submodule.<name>.update configuraiton variable.
252
-
253
- Will merge to 'master'.
254
- source: <20230725212218.711116-1-pvutov@imap.cc>
255
-
256
-
257
-* la/doc-choose-starting-point-fixup (2023-07-27) 3 commits
258
- (merged to 'next' on 2023-07-28 at 047dcae31c)
259
- + SubmittingPatches: use of older maintenance tracks is an exception
260
- + SubmittingPatches: explain why 'next' and above are inappropriate base
261
- + SubmittingPatches: choice of base for fixing an older maintenance track
262
- (this branch uses la/doc-choose-starting-point.)
263
-
264
- Clarify how to pick a starting point for a new topic in the
265
- SubmittingPatches document.
266
-
267
- Will merge to 'master', together with the underlying topic.
268
- source: <pull.1556.v2.git.1689314493.gitgitgadget@gmail.com>
269
- source: <pull.1556.v3.git.1690340701.gitgitgadget@gmail.com>
270
-
271
-
359
* am/doc-sha256 (2023-07-31) 1 commit
360
(merged to 'next' on 2023-08-01 at d7419bf527)
361
+ doc: sha256 is no longer experimental
@@ -283,43 +370,6 @@ Release tarballs are available at:
370
source: <ZMe6KmzZGVubYpvO@adams>
371
372
286
-* hy/blame-in-bare-with-contents (2023-07-21) 1 commit
287
- (merged to 'next' on 2023-07-31 at 39ac96d8d8)
288
- + blame: allow --contents to work with bare repo
289
-
290
- "git blame --contents=file" has been taught to work in a bare
291
- repository.
292
-
293
- Will merge to 'master'.
294
- source: <20230721035758.61956-1-hanyang.tony@bytedance.com>
295
-
296
-
297
-* ja/worktree-orphan-fix (2023-07-26) 3 commits
298
- (merged to 'next' on 2023-07-27 at e475016065)
299
- + t2400: rewrite regex to avoid unintentional PCRE
300
- + builtin/worktree.c: convert tab in advice to space
301
- + t2400: drop no-op `--sq` from rev-parse call
302
-
303
- Fix tests with unportable regex patterns.
304
-
305
- Will merge to 'master'.
306
- source: <20230726214202.15775-1-jacobabel@nullpo.dev>
307
-
308
-
309
-* jc/retire-get-sha1-hex (2023-07-24) 1 commit
310
- (merged to 'next' on 2023-07-27 at eeb9cc37f5)
311
- + hex: retire get_sha1_hex()
312
-
313
- The implementation of "get_sha1_hex()" that reads a hexadecimal
314
- string that spells a full object name has been extended to cope
315
- with any hash function used in the repository, but the "sha1" in
316
- its name survived. Rename it to get_hash_hex(), a name that is
317
- more consistent within its friends like get_hash_hex_algop().
318
-
319
- Will merge to 'master'.
320
- source: <xmqq1qgwoqgo.fsf_-_@gitster.g>
321
-
322
-
373
* rs/parse-options-negation-help (2023-07-24) 5 commits
374
- parse-options: show negatability of options in short help
375
- t1502: test option negation
@@ -330,9 +380,11 @@ Release tarballs are available at:
380
"git cmd -h" learned to signal which options can be negated by
381
listing such options like "--[no-]opt".
382
333
- Comments?
334
- Would showing "--[[no-]no-]opt" for "no-opt" be worth it?
335
- cf. <9e8225dd-1e8b-8af2-c3e1-0c5834694244@web.de>
383
+ Will merge to 'next'.
384
+ I personally liked a later suggestion to deal with "--[no-]no-foo"
385
+ but that can be done as a separate enhancement.
386
+ cf. <31b71333-de8b-d9a8-3ec4-1bad9cae2bf3@web.de>
387
+ cf. <xmqqjzug14jo.fsf@gitster.g>
388
source: <4d01e971-07cb-4f11-3cc6-9d9f21e590c1@web.de>
389
390
@@ -350,44 +402,6 @@ Release tarballs are available at:
402
source: <cover.1690216758.git.me@ttaylorr.com>
403
404
353
-* la/doc-choose-starting-point (2023-07-14) 5 commits
354
- (merged to 'next' on 2023-07-19 at 5a807cae46)
355
- + SubmittingPatches: simplify guidance for choosing a starting point
356
- + SubmittingPatches: emphasize need to communicate non-default starting points
357
- + SubmittingPatches: de-emphasize branches as starting points
358
- + SubmittingPatches: discuss subsystems separately from git.git
359
- + SubmittingPatches: reword awkward phrasing
360
- (this branch is used by la/doc-choose-starting-point-fixup.)
361
-
362
- Clarify how to choose the starting point for a new topic in
363
- developer guidance document.
364
-
365
- Will merge to 'master' together with the follow-on topic.
366
- source: <pull.1556.v2.git.1689314493.gitgitgadget@gmail.com>
367
-
368
-
369
-* jc/doc-sent-patch-now-what (2023-07-27) 1 commit
370
- (merged to 'next' on 2023-07-31 at 51f5d9d465)
371
- + MyFirstContribution: refrain from self-iterating too much
372
-
373
- Process document update.
374
-
375
- Will merge to 'master'.
376
- source: <xmqqmszg987u.fsf_-_@gitster.g>
377
-
378
-
379
-* jc/parse-options-short-help (2023-07-19) 3 commits
380
- (merged to 'next' on 2023-07-31 at e076d1f497)
381
- + short help: allow a gap smaller than USAGE_GAP
382
- + remote: simplify "remote add --tags" help text
383
- + short help: allow multi-line opthelp
384
-
385
- Command line parser fix, and a small parse-options API update.
386
-
387
- Will merge to 'master'.
388
- source: <xmqq5y6gg8fn.fsf@gitster.g>
389
-
390
-
405
* sl/sparse-check-attr (2023-07-18) 3 commits
406
- check-attr: integrate with sparse-index
407
- attr.c: read attributes in a sparse directory
@@ -401,19 +415,6 @@ Release tarballs are available at:
415
source: <20230718232916.31660-1-cheskaqiqi@gmail.com>
416
417
404
-* jc/branch-in-use-error-message (2023-07-21) 1 commit
405
- (merged to 'next' on 2023-07-31 at 22f17d131b)
406
- + branch: update the message to refuse touching a branch in-use
407
-
408
- "git branch -f X" to repoint the branch X seid that X was "checked
409
- out" in another worktree, even when branch X was not and instead
410
- being bisected or rebased. The message was reworded to say the
411
- branch was "in use".
412
-
413
- Will merge to 'master'.
414
- source: <xmqqr0p1szhz.fsf_-_@gitster.g>
415
-
416
-
418
* mh/credential-erase-improvements-more (2023-07-26) 2 commits
419
- credential/wincred: erase matching creds only
420
- credential/libsecret: erase matching creds only
@@ -454,140 +455,19 @@ Release tarballs are available at:
455
456
457
* jt/path-filter-fix (2023-08-01) 7 commits
457
- - commit-graph: new filter ver. that fixes murmur3
458
- - repo-settings: introduce commitgraph.changedPathsVersion
459
- - t4216: test changed path filters with high bit paths
460
- - t/helper/test-read-graph: implement `bloom-filters` mode
461
- - bloom.h: make `load_bloom_filter_from_graph()` public
462
- - t/helper/test-read-graph.c: extract `dump_graph_info()`
463
- - gitformat-commit-graph: describe version 2 of BDAT
458
+ (merged to 'next' on 2023-08-03 at d99958c287)
459
+ + commit-graph: new filter ver. that fixes murmur3
460
+ + repo-settings: introduce commitgraph.changedPathsVersion
461
+ + t4216: test changed path filters with high bit paths
462
+ + t/helper/test-read-graph: implement `bloom-filters` mode
463
+ + bloom.h: make `load_bloom_filter_from_graph()` public
464
+ + t/helper/test-read-graph.c: extract `dump_graph_info()`
465
+ + gitformat-commit-graph: describe version 2 of BDAT
466
467
The Bloom filter used for path limited history traversal was broken
468
on systems whose "char" is unsigned; update the implementation and
469
bump the format version to 2.
470
469
- Still under discussion.
470
- cf. <20230801185232.1457172-1-jonathantanmy@google.com>
471
+ Will merge to 'master'.
472
+ cf. <ZMqp6K2iXixWH/zT@nand.local>
473
source: <cover.1690912539.git.jonathantanmy@google.com>
472
-
473
-
474
-* mh/credential-libsecret-attrs (2023-06-16) 1 commit
475
- - credential/libsecret: store new attributes
476
-
477
- The way authentication related data other than passwords (e.g.
478
- oath token and password expiration data) are stored in libsecret
479
- keyrings has been rethought.
480
-
481
- Needs review.
482
- source: <pull.1469.v5.git.git.1686945306242.gitgitgadget@gmail.com>
483
-
484
-
485
-* cc/git-replay (2023-06-03) 15 commits
486
- - replay: stop assuming replayed branches do not diverge
487
- - replay: add --contained to rebase contained branches
488
- - replay: add --advance or 'cherry-pick' mode
489
- - replay: disallow revision specific options and pathspecs
490
- - replay: use standard revision ranges
491
- - replay: make it a minimal server side command
492
- - replay: remove HEAD related sanity check
493
- - replay: remove progress and info output
494
- - replay: add an important FIXME comment about gpg signing
495
- - replay: don't simplify history
496
- - replay: introduce pick_regular_commit()
497
- - replay: die() instead of failing assert()
498
- - replay: start using parse_options API
499
- - replay: introduce new builtin
500
- - t6429: remove switching aspects of fast-rebase
501
-
502
- What's the status of this thing?
503
- source: <20230602102533.876905-1-christian.couder@gmail.com>
504
-
505
---------------------------------------------------
506
-[Discarded]
507
-
508
-* jc/doc-submodule-update-settings (2023-07-13) 1 commit
509
- . submodule: clarify that "!custom command" is the only oddball
510
-
511
- Rewrite the description of giving a custom command to the
512
- submodule.<name>.update configuraiton variable.
513
-
514
- Superseded by pv/doc-submodule-update-settings topic.
515
- source: <xmqqwmz3oacg.fsf@gitster.g>
516
-
517
-
518
-* jc/rerere-read-rr-fix (2023-07-21) 1 commit
519
- . rerere: match the hash algorithm with its length
520
-
521
- SHA-256 fix.
522
-
523
- Superseded by jc/retire-get-sha1-hex
524
- source: <xmqqa5vou9ar.fsf@gitster.g>
525
-
526
-
527
-* cb/checkout-same-branch-twice (2023-03-22) 2 commits
528
- . SQUASH??? the test marked to expect failure passes from day one
529
- . checkout/switch: disallow checking out same branch in multiple worktrees
530
-
531
- "git checkout -B $branch" failed to protect against checking out
532
- a branch that is checked out elsewhere, unlike "git branch -f" did.
533
-
534
- Have been expecting a hopefully minor and final reroll for too long.
535
- cf. <CAPUEspj_Bh+LgYLnWfeBdcq_uV5Cbou-7H51GLFjzSa5Qzby9w@mail.gmail.com>
536
- source: <20230120113553.24655-1-carenas@gmail.com>
537
-
538
-
539
-* ed/fsmonitor-windows-named-pipe (2023-03-24) 1 commit
540
- . fsmonitor: handle differences between Windows named pipe functions
541
-
542
- Fix fsmonitor on Windows when the filesystem path contains certain
543
- characters.
544
-
545
- Have been expecting a reroll for too long.
546
- cf. <b9cf67e4-22a7-2ff0-8310-9223bea10d6d@jeffhostetler.com>
547
- source: <pull.1503.git.1679678090412.gitgitgadget@gmail.com>
548
-
549
-
550
-* rn/sparse-diff-index (2023-04-10) 1 commit
551
- . diff-index: enable sparse index
552
-
553
- "git diff-index" command has been taught to work better with the
554
- sparse index.
555
-
556
- Have been expecting a reroll for too long.
557
- cf. <62821012-4fc3-5ad8-695c-70f7ab14a8c9@github.com>
558
- source: <20230408112342.404318-1-nanth.raghul@gmail.com>
559
-
560
-
561
-* es/recurse-submodules-option-is-a-bool (2023-04-10) 1 commit
562
- . usage: clarify --recurse-submodules as a boolean
563
-
564
- The "--[no-]recurse-submodules" option of "git checkout" and others
565
- supported an undocumented syntax --recurse-submodules=<value> where
566
- the value can spell a Boolean in various ways. The support for the
567
- syntax is being dropped.
568
-
569
- Have been expecting a reroll for too long.
570
- cf. <ZDSTFwMFO7vbj/du@google.com>
571
- source: <ZDSTFwMFO7vbj/du@google.com>
572
-
573
-
574
-* jc/checkout-merge-fix (2023-07-28) 2 commits
575
- . checkout/restore: add basic tests for --merge
576
- . checkout/restore: refuse unmerging paths unless checking out of the index
577
-
578
- "git checkout/restore --merge -- $path" improvements.
579
-
580
- Superseded by jc/unresolve-removal
581
- source: <xmqq7cqj4rme.fsf@gitster.g>
582
-
583
-
584
-* jc/resolve-undo-fixes (2023-07-28) 4 commits
585
- . update-index: remove stale fallback code for "--unresolve"
586
- . update-index: use unmerge_index_entry() to support removal
587
- . resolve-undo: allow resurrecting conflicted state that resolved to deletion
588
- . update-index: do not read HEAD and MERGE_HEAD unconditionally
589
-
590
- Assorted fixes and clean-up around resolve-undo data.
591
-
592
- Superseded by jc/unresolve-removal
593
- source: <xmqqo7jv4y0t.fsf_-_@gitster.g>