What's cooking (2019/08 #06)
Junio C Hamano committed
Aug 23, 2019 at 15:11 UTC
75517afe5f1b20ba0f8d3355954529a783963a86
1 file changed
+329
-319
whats-cooking.txt
+329
-319
@@ -1,10 +1,10 @@
1
To: git@vger.kernel.org
2
Bcc: lwn@lwn.net
3
-Subject: What's cooking in git.git (Aug 2019, #05; Wed, 21)
4
-X-master-at: 5fa0f5238b0cd46cfe7f6fa76c3f526ea98148d9
5
-X-next-at: 17f5b7556cbe9867bed85982f386e41d9db77e19
3
+Subject: What's cooking in git.git (Aug 2019, #06; Fri, 23)
4
+X-master-at: 745f6812895b31c02b29bdfe4ae8e5498f776c26
5
+X-next-at: 0a3290393f27b805eda46be3998bd9e32518815d
6
7
-What's cooking in git.git (Aug 2019, #05; Wed, 21)
7
+What's cooking in git.git (Aug 2019, #06; Fri, 23)
8
--------------------------------------------------
9
10
Here are the topics that have been cooking. Commits prefixed with
@@ -12,20 +12,248 @@ 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
-The first batch of topics post 2.23 will be merged to 'master' and
16
-'next' will be rewound to become empty, soon. After that, 'next'
17
-will grow merging only smaller fixes and cleanup topics first.
18
-Bigger stuff will then follow after the dust settles. All of the
19
-above hopefully will happen before the end of the month.
15
+The tip of 'next' has been rewound. A handful of topics that have
16
+been in the stalled state need to be woken up, now the release is
17
+over (or they can be dropped without prejudice, just to unclutter
18
+the tree). As I said in an earlier message somewhere, I'd prefer to
19
+see the PCRE/JIT topics sorted out soonish. Ævar? Carlo?
20
21
You can find the changes described here in the integration branches
22
of the repositories listed at
23
24
http://git-blame.blogspot.com/p/git-public-repositories.html
25
26
+--------------------------------------------------
27
+[Graduated to "master"]
28
+
29
+* en/fast-import-merge-doc (2019-08-12) 1 commit
30
+ (merged to 'next' on 2019-08-14 at 65efc63345)
31
+ + git-fast-import.txt: clarify that multiple merge commits are allowed
32
+
33
+ Doc update.
34
+
35
+
36
+* jk/perf-no-dups (2019-08-12) 1 commit
37
+ (merged to 'next' on 2019-08-14 at 3010a7d19f)
38
+ + t/perf: rename duplicate-numbered test script
39
+
40
+ Test & perf scripts must use unique numeric prefix, but a pair
41
+ shared the same number, which is fixed here.
42
+
43
+
44
+* jk/tree-walk-overflow (2019-08-01) 6 commits
45
+ (merged to 'next' on 2019-08-02 at 116467c77a)
46
+ + tree-walk: harden make_traverse_path() length computations
47
+ + tree-walk: add a strbuf wrapper for make_traverse_path()
48
+ + tree-walk: accept a raw length for traverse_path_len()
49
+ + tree-walk: use size_t consistently
50
+ + tree-walk: drop oid from traverse_info
51
+ + setup_traverse_info(): stop copying oid
52
+
53
+ Codepaths to walk tree objects have been audited for integer
54
+ overflows and hardened.
55
+
56
+
57
+* mt/grep-submodules-working-tree (2019-07-30) 1 commit
58
+ (merged to 'next' on 2019-08-02 at e1a46a195f)
59
+ + grep: fix worktree case in submodules
60
+
61
+ "git grep --recurse-submodules" that looks at the working tree
62
+ files looked at the contents in the index in submodules, instead of
63
+ files in the working tree.
64
+
65
+
66
+* rs/nedalloc-fixlets (2019-08-07) 2 commits
67
+ (merged to 'next' on 2019-08-09 at 0e9286fb70)
68
+ + nedmalloc: avoid compiler warning about unused value
69
+ + nedmalloc: do assignments only after the declaration section
70
+
71
+ Compilation fix.
72
+
73
+
74
+* sc/pack-refs-deletion-racefix (2019-08-02) 1 commit
75
+ (merged to 'next' on 2019-08-09 at 6bd6c85fbb)
76
+ + pack-refs: always refresh after taking the lock file
77
+
78
+ "git pack-refs" can lose refs that are created while running, which
79
+ is getting corrected.
80
+ Performance impact unclear but let's let correctness take priority
81
+ for now.
82
+ cf. <20190816204906.GA29853@sigill.intra.peff.net>
83
+
84
+
85
+* sg/commit-graph-validate (2019-08-05) 3 commits
86
+ (merged to 'next' on 2019-08-09 at 87f61abb73)
87
+ + commit-graph: error out on invalid commit oids in 'write --stdin-commits'
88
+ + commit-graph: turn a group of write-related macro flags into an enum
89
+ + t5318-commit-graph: use 'test_expect_code'
90
+
91
+ The code to write commit-graph over given commit object names has
92
+ been made a bit more robust.
93
+
94
+
95
+* sg/do-not-skip-non-httpd-tests (2019-08-02) 3 commits
96
+ (merged to 'next' on 2019-08-02 at e700df0693)
97
+ + t: warn against adding non-httpd-specific tests after sourcing 'lib-httpd'
98
+ + t5703: run all non-httpd-specific tests before sourcing 'lib-httpd.sh'
99
+ + t5510-fetch: run non-httpd-specific test before sourcing 'lib-httpd.sh'
100
+ (this branch uses sg/t5510-test-i18ngrep-fix.)
101
+
102
+ Test fix.
103
+
104
+
105
+* sg/show-failed-test-names (2019-08-05) 2 commits
106
+ (merged to 'next' on 2019-08-09 at c49ece8a0d)
107
+ + tests: show the test name and number at the start of verbose output
108
+ + t0000-basic: use realistic test script names in the verbose tests
109
+
110
+ The first line of verbose output from each test piece now carries
111
+ the test name and number to help scanning with eyeballs.
112
+
113
+
114
+* sg/t5510-test-i18ngrep-fix (2019-07-31) 1 commit
115
+ (merged to 'next' on 2019-08-02 at 678564fda2)
116
+ + t5510-fetch: fix negated 'test_i18ngrep' invocation
117
+ (this branch is used by sg/do-not-skip-non-httpd-tests.)
118
+
119
+ Test fix.
120
+
121
+
122
+* sg/worktree-remove-errormsg (2019-08-13) 1 commit
123
+ (merged to 'next' on 2019-08-14 at e5444969c9)
124
+ + worktree remove: clarify error message on dirty worktree
125
+
126
+ Error message update/clarification.
127
+
128
+
129
+* vn/restore-empty-ita-corner-case-fix (2019-08-02) 2 commits
130
+ (merged to 'next' on 2019-08-09 at ad2d7630d1)
131
+ + restore: add test for deleted ita files
132
+ + checkout.c: unstage empty deleted ita files
133
+
134
+ "git checkout" and "git restore" to re-populate the index from a
135
+ tree-ish (typically HEAD) did not work correctly for a path that
136
+ was removed and then added again with the intent-to-add bit, when
137
+ the corresponding working tree file was empty. This has been
138
+ corrected.
139
+
140
--------------------------------------------------
141
[New Topics]
142
143
+* jt/avoid-ls-refs-with-http (2019-08-22) 2 commits
144
+ - transport: teach all vtables to allow fetch first
145
+ - transport-helper: skip ls-refs if unnecessary
146
+
147
+ The http transport lacked some optimization the native transports
148
+ learned to avoid unnecessary ref advertisement, which has been
149
+ corrected.
150
+
151
+ Will merge to 'next'.
152
+
153
+--------------------------------------------------
154
+[Stalled]
155
+
156
+* js/builtin-add-i (2019-07-18) 11 commits
157
+ - built-in add -i: implement the `help` command
158
+ - built-in add -i: use color in the main loop
159
+ - built-in add -i: support `?` (prompt help)
160
+ - built-in add -i: show unique prefixes of the commands
161
+ - Add a function to determine unique prefixes for a list of strings
162
+ - built-in add -i: implement the main loop
163
+ - built-in add -i: color the header in the `status` command
164
+ - built-in add -i: refresh the index before running `status`
165
+ - built-in add -i: implement the `status` command
166
+ - diff: export diffstat interface
167
+ - Start to implement a built-in version of `git add --interactive`
168
+
169
+ The beginning of rewriting "git add -i" in C.
170
+
171
+ Expecting a response and possibly a reroll.
172
+ cf. <xmqqwofyozka.fsf@gitster-ct.c.googlers.com>
173
+ cf. <xmqqsgqmoyz1.fsf@gitster-ct.c.googlers.com>
174
+
175
+
176
+* js/trace2-json-schema (2019-07-25) 3 commits
177
+ . ci: run trace2 schema validation in the CI suite
178
+ . trace2: add a schema validator for trace2 events
179
+ . trace2: add a JSON schema for trace2 events
180
+
181
+ The JSON output produced by "trace2" subsystem now has JSON schema
182
+ defined on it, to allow us validate the output and catch deviation.
183
+
184
+ The CI integration may be a bit too heavy-handed.
185
+
186
+
187
+* jn/unknown-index-extensions (2018-11-21) 2 commits
188
+ - index: offer advice for unknown index extensions
189
+ - index: do not warn about unrecognized extensions
190
+
191
+ A bit too alarming warning given when unknown index extensions
192
+ exist is getting revamped.
193
+
194
+ Expecting a reroll.
195
+
196
+
197
+* jc/format-patch-delay-message-id (2019-04-05) 1 commit
198
+ - format-patch: move message-id and related headers to the end
199
+
200
+ The location "git format-patch --thread" adds the Message-Id:
201
+ header in the series of header fields has been moved down, which
202
+ may help working around a suspected bug in GMail MSA, reported at
203
+ <CAHk-=whP1stFZNAaJiMi5eZ9rj0MRt20Y_yHVczZPH+O01d+sA@mail.gmail.com>
204
+
205
+ Waiting for feedback to see if it truly helps.
206
+ Needs tests.
207
+
208
+
209
+* jt/fetch-cdn-offload (2019-03-12) 9 commits
210
+ - SQUASH???
211
+ - upload-pack: send part of packfile response as uri
212
+ - fetch-pack: support more than one pack lockfile
213
+ - upload-pack: refactor reading of pack-objects out
214
+ - Documentation: add Packfile URIs design doc
215
+ - Documentation: order protocol v2 sections
216
+ - http-fetch: support fetching packfiles by URL
217
+ - http: improve documentation of http_pack_request
218
+ - http: use --stdin when getting dumb HTTP pack
219
+
220
+ WIP for allowing a response to "git fetch" to instruct the bulk of
221
+ the pack contents to be instead taken from elsewhere (aka CDN).
222
+
223
+
224
+* js/protocol-advertise-multi (2018-12-28) 1 commit
225
+ - protocol: advertise multiple supported versions
226
+
227
+ The transport layer has been updated so that the protocol version
228
+ used can be negotiated between the parties, by the initiator
229
+ listing the protocol versions it is willing to talk, and the other
230
+ side choosing from one of them.
231
+
232
+ Expecting a reroll.
233
+ cf. <CANq=j3u-zdb_FvNJGPCmygNMScseav63GhVvBX3NcVS4f7TejA@mail.gmail.com>
234
+
235
+
236
+* mk/use-size-t-in-zlib (2018-10-15) 1 commit
237
+ - zlib.c: use size_t for size
238
+
239
+ The wrapper to call into zlib followed our long tradition to use
240
+ "unsigned long" for sizes of regions in memory, which have been
241
+ updated to use "size_t".
242
+
243
+
244
+* dl/remote-save-to-push (2018-12-11) 1 commit
245
+ - remote: add --save-to-push option to git remote set-url
246
+
247
+ "git remote set-url" learned a new option that moves existing value
248
+ of the URL field to pushURL field of the remote before replacing
249
+ the URL field with a new value.
250
+
251
+ Anybody who wants to champion this topic?
252
+ I am personally not yet quite convinced if this is worth pursuing.
253
+
254
+--------------------------------------------------
255
+[Cooking]
256
+
257
* cb/curl-use-xmalloc (2019-08-15) 1 commit
258
- http: use xmalloc with cURL
259
(this branch uses jk/drop-release-pack-memory.)
@@ -37,22 +265,24 @@ of the repositories listed at
265
266
267
* sg/diff-indent-heuristic-non-experimental (2019-08-15) 1 commit
40
- - diff: 'diff.indentHeuristic' is no longer experimental
268
+ (merged to 'next' on 2019-08-23 at faab5e2a1e)
269
+ + diff: 'diff.indentHeuristic' is no longer experimental
270
271
We promoted the "indent heuristics" that decides where to split
272
diff hunks from experimental to the default a few years ago, but
273
some stale documentation still marked it as experimental, which has
274
been corrected.
275
47
- Will merge to 'next'.
276
+ Will merge to 'master'.
277
278
279
* en/checkout-mismerge-fix (2019-08-16) 1 commit
51
- - checkout: remove duplicate code
280
+ (merged to 'next' on 2019-08-23 at 8ecb69414a)
281
+ + checkout: remove duplicate code
282
283
Fix a mismerge that happened in 2.22 timeframe.
284
55
- Will merge to 'next'.
285
+ Will merge to 'master'.
286
287
288
* en/merge-recursive-cleanup (2019-08-19) 24 commits
@@ -145,11 +375,12 @@ of the repositories listed at
375
376
377
* bm/repository-layout-typofix (2019-08-19) 1 commit
148
- - repository-layout.txt: correct pluralization of 'object'
378
+ (merged to 'next' on 2019-08-23 at 7b1fffbaea)
379
+ + repository-layout.txt: correct pluralization of 'object'
380
381
Typofix.
382
152
- Will merge to 'next'.
383
+ Will merge to 'master'.
384
385
386
* bw/rebase-autostash-keep-current-branch (2019-08-21) 1 commit
@@ -159,6 +390,9 @@ of the repositories listed at
390
different from the current branch, incorrectly moved the tip of the
391
current branch, which has been corrected.
392
393
+ Getting there.
394
+ cf. <nycvar.QRO.7.76.6.1908221418540.46@tvgsbejvaqbjf.bet>
395
+
396
397
* dl/format-patch-cover-letter-subject (2019-08-21) 1 commit
398
- format-patch: learn --infer-cover-subject option
@@ -189,41 +423,47 @@ of the repositories listed at
423
The documentation and tests for "git format-patch" have been
424
cleaned up.
425
426
+ Excpecting a reroll.
427
+
428
429
* ds/midx-expire-repack (2019-08-20) 1 commit
194
- - packfile.h: drop extern from function declaration
430
+ (merged to 'next' on 2019-08-23 at a998356b3b)
431
+ + packfile.h: drop extern from function declaration
432
433
Code cleanup.
434
198
- Will merge to 'next'.
435
+ Will merge to 'master'.
436
437
438
* jt/diff-lazy-fetch-submodule-fix (2019-08-20) 1 commit
202
- - diff: skip GITLINK when lazy fetching missing objs
439
+ (merged to 'next' on 2019-08-23 at 16997cac26)
440
+ + diff: skip GITLINK when lazy fetching missing objs
441
442
On-demand object fetching in lazy clone incorrectly tried to fetch
443
commits from submodule projects, while still working in the
444
superproject, which has been corrected.
445
208
- Will merge to 'next'.
446
+ Will merge to 'master'.
447
448
211
-* mp/for-each-ref-missing-name-or-email (2019-08-19) 1 commit
212
- - ref-filter: initialize empty name or email fields
449
+* mp/for-each-ref-missing-name-or-email (2019-08-22) 1 commit
450
+ (merged to 'next' on 2019-08-23 at d257cb2138)
451
+ + ref-filter: initialize empty name or email fields
452
453
"for-each-ref" and friends that shows refs did not protect themselves
454
against ancient tags that did not record tagger names when asked to
455
show "%(taggername)", which have been corrected.
456
218
- May want a test or two.
457
+ Will merge to 'master'.
458
459
460
* nd/diff-parseopt (2019-08-20) 1 commit
222
- - parseopt: move definition of enum parse_opt_result up
461
+ (merged to 'next' on 2019-08-23 at d80a83b1e4)
462
+ + parseopt: move definition of enum parse_opt_result up
463
464
Compilation fix.
465
226
- Will merge to 'next'.
466
+ Will merge to 'master'.
467
468
469
* pw/rebase-i-show-HEAD-to-reword (2019-08-19) 3 commits
@@ -238,31 +478,34 @@ of the repositories listed at
478
479
480
* rs/pax-extended-header-length-fix (2019-08-19) 4 commits
241
- - archive-tar: turn length miscalculation warning into BUG
242
- - archive-tar: use size_t in strbuf_append_ext_header()
243
- - archive-tar: fix pax extended header length calculation
244
- - archive-tar: report wrong pax extended header length
481
+ (merged to 'next' on 2019-08-23 at de88e8e58a)
482
+ + archive-tar: turn length miscalculation warning into BUG
483
+ + archive-tar: use size_t in strbuf_append_ext_header()
484
+ + archive-tar: fix pax extended header length calculation
485
+ + archive-tar: report wrong pax extended header length
486
487
"git archive" recorded incorrect length in extended pax header in
488
some corner cases, which has been corrected.
489
249
- Will merge to 'next'.
490
+ Will merge to 'master'.
491
492
493
* rs/sort-oid-array-thread-safe (2019-08-20) 1 commit
253
- - sha1-name: make sort_ambiguous_oid_array() thread-safe
494
+ (merged to 'next' on 2019-08-23 at 5ce3126af2)
495
+ + sha1-name: make sort_ambiguous_oid_array() thread-safe
496
497
Prepare get_short_oid() codepath to be thread-safe.
498
257
- Will merge to 'next'.
499
+ Will merge to 'master'.
500
501
260
-* sb/userdiff-dts (2019-08-19) 1 commit
261
- - userdiff: Add a builtin pattern for dts files
502
+* sb/userdiff-dts (2019-08-21) 1 commit
503
+ (merged to 'next' on 2019-08-23 at 2bce0b8f91)
504
+ + userdiff: add a builtin pattern for dts files
505
506
Device-tree files learned their own userdiff patterns.
507
265
- Will merge to 'next'.
508
+ Will merge to 'master'.
509
510
511
* sg/line-log-tree-diff-optim (2019-08-21) 2 commits
@@ -274,145 +517,35 @@ of the repositories listed at
517
Will merge to 'next'.
518
519
277
-* tg/t0021-racefix (2019-08-21) 1 commit
278
- - t0021: make sure clean filter runs
520
+* tg/t0021-racefix (2019-08-22) 1 commit
521
+ (merged to 'next' on 2019-08-23 at 0a3290393f)
522
+ + t0021: make sure clean filter runs
523
524
A test fix.
525
282
- Will merge to 'next'.
283
-
284
-
285
---------------------------------------------------
286
-[Stalled]
287
-
288
-* js/trace2-json-schema (2019-07-25) 3 commits
289
- . ci: run trace2 schema validation in the CI suite
290
- . trace2: add a schema validator for trace2 events
291
- . trace2: add a JSON schema for trace2 events
292
-
293
- The JSON output produced by "trace2" subsystem now has JSON schema
294
- defined on it, to allow us validate the output and catch deviation.
295
-
296
- The CI integration may be a bit too heavy-handed.
297
-
298
-
299
-* jn/unknown-index-extensions (2018-11-21) 2 commits
300
- - index: offer advice for unknown index extensions
301
- - index: do not warn about unrecognized extensions
302
-
303
- A bit too alarming warning given when unknown index extensions
304
- exist is getting revamped.
305
-
306
- Expecting a reroll.
307
-
308
-
309
-* jc/format-patch-delay-message-id (2019-04-05) 1 commit
310
- - format-patch: move message-id and related headers to the end
311
-
312
- The location "git format-patch --thread" adds the Message-Id:
313
- header in the series of header fields has been moved down, which
314
- may help working around a suspected bug in GMail MSA, reported at
315
- <CAHk-=whP1stFZNAaJiMi5eZ9rj0MRt20Y_yHVczZPH+O01d+sA@mail.gmail.com>
316
-
317
- Waiting for feedback to see if it truly helps.
318
- Needs tests.
319
-
320
-
321
-* jt/fetch-cdn-offload (2019-03-12) 9 commits
322
- - SQUASH???
323
- - upload-pack: send part of packfile response as uri
324
- - fetch-pack: support more than one pack lockfile
325
- - upload-pack: refactor reading of pack-objects out
326
- - Documentation: add Packfile URIs design doc
327
- - Documentation: order protocol v2 sections
328
- - http-fetch: support fetching packfiles by URL
329
- - http: improve documentation of http_pack_request
330
- - http: use --stdin when getting dumb HTTP pack
331
-
332
- WIP for allowing a response to "git fetch" to instruct the bulk of
333
- the pack contents to be instead taken from elsewhere (aka CDN).
334
-
335
-
336
-* js/protocol-advertise-multi (2018-12-28) 1 commit
337
- - protocol: advertise multiple supported versions
338
-
339
- The transport layer has been updated so that the protocol version
340
- used can be negotiated between the parties, by the initiator
341
- listing the protocol versions it is willing to talk, and the other
342
- side choosing from one of them.
343
-
344
- Expecting a reroll.
345
- cf. <CANq=j3u-zdb_FvNJGPCmygNMScseav63GhVvBX3NcVS4f7TejA@mail.gmail.com>
346
-
347
-
348
-* mk/use-size-t-in-zlib (2018-10-15) 1 commit
349
- - zlib.c: use size_t for size
350
-
351
- The wrapper to call into zlib followed our long tradition to use
352
- "unsigned long" for sizes of regions in memory, which have been
353
- updated to use "size_t".
354
-
355
-
356
-* dl/remote-save-to-push (2018-12-11) 1 commit
357
- - remote: add --save-to-push option to git remote set-url
358
-
359
- "git remote set-url" learned a new option that moves existing value
360
- of the URL field to pushURL field of the remote before replacing
361
- the URL field with a new value.
362
-
363
- Anybody who wants to champion this topic?
364
- I am personally not yet quite convinced if this is worth pursuing.
526
+ Will merge to 'master'.
527
366
---------------------------------------------------
367
-[Cooking]
528
529
* py/git-gui-do-quit (2019-08-07) 2 commits
370
- - Merge branch 'py/call-do-quit-before-exit' of github.com:gitster/git-gui into py/git-gui-do-quit
371
- - git-gui: call do_quit before destroying the main window
530
+ (merged to 'next' on 2019-08-23 at 6544c9dbb3)
531
+ + Merge branch 'py/call-do-quit-before-exit' of github.com:gitster/git-gui into py/git-gui-do-quit
532
+ + git-gui: call do_quit before destroying the main window
533
534
"git gui" learned to call the clean-up procedure before exiting.
535
375
- Will merge to 'next'.
376
-
377
-
378
-* rs/nedalloc-fixlets (2019-08-07) 2 commits
379
- (merged to 'next' on 2019-08-09 at 0e9286fb70)
380
- + nedmalloc: avoid compiler warning about unused value
381
- + nedmalloc: do assignments only after the declaration section
382
-
383
- Compilation fix.
384
-
385
- Will merge to 'master'.
386
-
387
-
388
-* jk/perf-no-dups (2019-08-12) 1 commit
389
- (merged to 'next' on 2019-08-14 at 3010a7d19f)
390
- + t/perf: rename duplicate-numbered test script
391
-
392
- Test & perf scripts must use unique numeric prefix, but a pair
393
- shared the same number, which is fixed here.
394
-
395
- Will merge to 'master'.
396
-
397
-
398
-* en/fast-import-merge-doc (2019-08-12) 1 commit
399
- (merged to 'next' on 2019-08-14 at 65efc63345)
400
- + git-fast-import.txt: clarify that multiple merge commits are allowed
401
-
402
- Doc update.
403
-
536
Will merge to 'master'.
537
538
539
* bc/reread-attributes-during-rebase (2019-08-19) 2 commits
408
- - apply: reload .gitattributes after patching it
409
- - path: add a function to check for path suffix
540
+ (merged to 'next' on 2019-08-23 at 14aaf533d6)
541
+ + apply: reload .gitattributes after patching it
542
+ + path: add a function to check for path suffix
543
544
The "git am" based backend of "git rebase" ignored the result of
545
updating ".gitattributes" done in one step when replaying
546
subsequent steps.
547
415
- Will merge to 'next'.
548
+ Will merge to 'master'.
549
550
551
* jk/drop-release-pack-memory (2019-08-13) 1 commit
@@ -447,21 +580,18 @@ of the repositories listed at
580
Will merge to 'next'.
581
582
450
-* sg/worktree-remove-errormsg (2019-08-13) 1 commit
451
- (merged to 'next' on 2019-08-14 at e5444969c9)
452
- + worktree remove: clarify error message on dirty worktree
453
-
454
- Error message update/clarification.
455
-
456
- Will merge to 'master'.
457
-
458
-
583
* mt/threaded-grep-in-object-store (2019-08-13) 4 commits
584
- grep: re-enable threads in some non-worktree cases
585
- grep: disable grep_read_mutex when possible
586
- grep: allow locks to be enabled individually
587
- object-store: add lock to read_object_file_extended()
588
589
+ Traditionally, we avoided threaded grep while searching in objects
590
+ (as opposed to files in the working tree) as accesses to the object
591
+ layer is not thread-safe. This limitation is getting lifted.
592
+
593
+ cf. <cover.1565468806.git.matheus.bernardino@usp.br>
594
+
595
596
* pd/fetch-jobs (2019-08-13) 5 commits
597
. fetch: make --jobs control submodules and remotes
@@ -475,33 +605,13 @@ of the repositories listed at
605
606
607
* cb/fetch-set-upstream (2019-08-19) 1 commit
478
- - pull, fetch: add --set-upstream option
608
+ (merged to 'next' on 2019-08-23 at cf8c2ea0a0)
609
+ + pull, fetch: add --set-upstream option
610
611
"git fetch" learned "--set-upstream" option to help those who first
612
clone from their private fork they intend to push to, add the true
613
upstream via "git remote add" and then "git fetch" from it.
614
484
-
485
-* sg/commit-graph-validate (2019-08-05) 3 commits
486
- (merged to 'next' on 2019-08-09 at 87f61abb73)
487
- + commit-graph: error out on invalid commit oids in 'write --stdin-commits'
488
- + commit-graph: turn a group of write-related macro flags into an enum
489
- + t5318-commit-graph: use 'test_expect_code'
490
-
491
- The code to write commit-graph over given commit object names has
492
- been made a bit more robust.
493
-
494
- Will merge to 'master'.
495
-
496
-
497
-* sg/show-failed-test-names (2019-08-05) 2 commits
498
- (merged to 'next' on 2019-08-09 at c49ece8a0d)
499
- + tests: show the test name and number at the start of verbose output
500
- + t0000-basic: use realistic test script names in the verbose tests
501
-
502
- The first line of verbose output from each test piece now carries
503
- the test name and number to help scanning with eyeballs.
504
-
615
Will merge to 'master'.
616
617
@@ -517,9 +627,10 @@ of the repositories listed at
627
628
629
* jk/eoo (2019-08-06) 3 commits
520
- - gitcli: document --end-of-options
521
- - parse-options: allow --end-of-options as a synonym for "--"
522
- - revision: allow --end-of-options to end option parsing
630
+ (merged to 'next' on 2019-08-23 at 217e97127a)
631
+ + gitcli: document --end-of-options
632
+ + parse-options: allow --end-of-options as a synonym for "--"
633
+ + revision: allow --end-of-options to end option parsing
634
635
The command line parser learned "--end-of-options" notation; the
636
standard convention for scripters to have hardcoded set of options
@@ -528,16 +639,19 @@ of the repositories listed at
639
that would not work for commands that use "--" as a delimiter
640
between revs and pathspec.
641
531
- Will merge to 'next'.
642
+ Will merge to 'master'.
643
644
645
* jk/repo-init-cleanup (2019-08-06) 3 commits
535
- - config: stop checking whether the_repository is NULL
536
- - common-main: delay trace2 initialization
537
- - t1309: use short branch name in includeIf.onbranch test
646
+ (merged to 'next' on 2019-08-23 at 1a5e4c4f06)
647
+ + config: stop checking whether the_repository is NULL
648
+ + common-main: delay trace2 initialization
649
+ + t1309: use short branch name in includeIf.onbranch test
650
651
Further clean-up of the initialization code.
652
653
+ Will merge to 'master'.
654
+
655
656
* cb/pcre2-chartables-leakfix (2019-08-06) 3 commits
657
- grep: avoid leak of chartables in PCRE2
@@ -561,92 +675,18 @@ of the repositories listed at
675
Will merge to 'next'.
676
677
564
-* sc/pack-refs-deletion-racefix (2019-08-02) 1 commit
565
- (merged to 'next' on 2019-08-09 at 6bd6c85fbb)
566
- + pack-refs: always refresh after taking the lock file
567
-
568
- "git pack-refs" can lose refs that are created while running, which
569
- is getting corrected.
570
-
571
- Will merge to 'master'.
572
- Performance impact unclear but let's let correctness take priority
573
- for now.
574
- cf. <20190816204906.GA29853@sigill.intra.peff.net>
575
-
576
-
577
-* vn/restore-empty-ita-corner-case-fix (2019-08-02) 2 commits
578
- (merged to 'next' on 2019-08-09 at ad2d7630d1)
579
- + restore: add test for deleted ita files
580
- + checkout.c: unstage empty deleted ita files
581
-
582
- "git checkout" and "git restore" to re-populate the index from a
583
- tree-ish (typically HEAD) did not work correctly for a path that
584
- was removed and then added again with the intent-to-add bit, when
585
- the corresponding working tree file was empty. This has been
586
- corrected.
587
-
588
- Will merge to 'master'.
589
-
590
-
678
* ds/feature-macros (2019-08-13) 6 commits
592
- - repo-settings: create feature.experimental setting
593
- - repo-settings: create feature.manyFiles setting
594
- - repo-settings: parse core.untrackedCache
595
- - commit-graph: turn on commit-graph by default
596
- - t6501: use 'git gc' in quiet mode
597
- - repo-settings: consolidate some config settings
679
+ (merged to 'next' on 2019-08-23 at 115f0c8df9)
680
+ + repo-settings: create feature.experimental setting
681
+ + repo-settings: create feature.manyFiles setting
682
+ + repo-settings: parse core.untrackedCache
683
+ + commit-graph: turn on commit-graph by default
684
+ + t6501: use 'git gc' in quiet mode
685
+ + repo-settings: consolidate some config settings
686
687
A mechanism to affect the default setting for a (related) group of
688
configuration variables is introduced.
689
602
- Will merge to 'next'.
603
-
604
-
605
-* mt/grep-submodules-working-tree (2019-07-30) 1 commit
606
- (merged to 'next' on 2019-08-02 at e1a46a195f)
607
- + grep: fix worktree case in submodules
608
-
609
- "git grep --recurse-submodules" that looks at the working tree
610
- files looked at the contents in the index in submodules, instead of
611
- files in the working tree.
612
-
613
- Will merge to 'master'.
614
-
615
-
616
-* jk/tree-walk-overflow (2019-08-01) 6 commits
617
- (merged to 'next' on 2019-08-02 at 116467c77a)
618
- + tree-walk: harden make_traverse_path() length computations
619
- + tree-walk: add a strbuf wrapper for make_traverse_path()
620
- + tree-walk: accept a raw length for traverse_path_len()
621
- + tree-walk: use size_t consistently
622
- + tree-walk: drop oid from traverse_info
623
- + setup_traverse_info(): stop copying oid
624
-
625
- Codepaths to walk tree objects have been audited for integer
626
- overflows and hardened.
627
-
628
- Will merge to 'master'.
629
-
630
-
631
-* sg/t5510-test-i18ngrep-fix (2019-07-31) 1 commit
632
- (merged to 'next' on 2019-08-02 at 678564fda2)
633
- + t5510-fetch: fix negated 'test_i18ngrep' invocation
634
- (this branch is used by sg/do-not-skip-non-httpd-tests.)
635
-
636
- Test fix.
637
-
638
- Will merge to 'master'.
639
-
640
-
641
-* sg/do-not-skip-non-httpd-tests (2019-08-02) 3 commits
642
- (merged to 'next' on 2019-08-02 at e700df0693)
643
- + t: warn against adding non-httpd-specific tests after sourcing 'lib-httpd'
644
- + t5703: run all non-httpd-specific tests before sourcing 'lib-httpd.sh'
645
- + t5510-fetch: run non-httpd-specific test before sourcing 'lib-httpd.sh'
646
- (this branch uses sg/t5510-test-i18ngrep-fix.)
647
-
648
- Test fix.
649
-
690
Will merge to 'master'.
691
692
@@ -697,33 +737,17 @@ of the repositories listed at
737
Will merge to 'next'.
738
739
700
-* js/builtin-add-i (2019-07-18) 11 commits
701
- - built-in add -i: implement the `help` command
702
- - built-in add -i: use color in the main loop
703
- - built-in add -i: support `?` (prompt help)
704
- - built-in add -i: show unique prefixes of the commands
705
- - Add a function to determine unique prefixes for a list of strings
706
- - built-in add -i: implement the main loop
707
- - built-in add -i: color the header in the `status` command
708
- - built-in add -i: refresh the index before running `status`
709
- - built-in add -i: implement the `status` command
710
- - diff: export diffstat interface
711
- - Start to implement a built-in version of `git add --interactive`
712
-
713
- The beginning of rewriting "git add -i" in C.
714
-
715
- Expecting a response and possibly a reroll.
716
- cf. <xmqqwofyozka.fsf@gitster-ct.c.googlers.com>
717
- cf. <xmqqsgqmoyz1.fsf@gitster-ct.c.googlers.com>
718
-
719
-
740
* es/walken-tutorial (2019-08-07) 1 commit
741
- documentation: add tutorial for revision walking
742
743
A tutorial on object enumeration.
744
745
+ Perhaps take the thing with as little tweaks as possible, retitling
746
+ it to "my first object enumeration" or something?
747
+ cf. <20190814183328.GA40797@google.com>
748
+
749
726
-* ab/pcre-jit-fixes (2019-08-19) 8 commits
750
+* ab/pcre-jit-fixes (2019-08-19) 18 commits
751
- grep: under --debug, show whether PCRE JIT is enabled
752
- grep: do not enter PCRE2_UTF mode on fixed matching
753
- grep: stess test PCRE v2 on invalid UTF-8 data
@@ -732,7 +756,16 @@ of the repositories listed at
756
- grep: stop using a custom JIT stack with PCRE v1
757
- grep: stop "using" a custom JIT stack with PCRE v2
758
- grep: remove overly paranoid BUG(...) code
735
- (this branch uses ab/no-kwset.)
759
+ - grep: use PCRE v2 for optimized fixed-string search
760
+ - grep: remove the kwset optimization
761
+ - grep: drop support for \0 in --fixed-strings <pattern>
762
+ - grep: make the behavior for NUL-byte in patterns sane
763
+ - grep tests: move binary pattern tests into their own file
764
+ - grep tests: move "grep binary" alongside the rest
765
+ - grep: inline the return value of a function call used only once
766
+ - t4210: skip more command-line encoding tests on MinGW
767
+ - grep: don't use PCRE2?_UTF8 with "log --encoding=<non-utf8>"
768
+ - log tests: test regex backends in "--encode=<enc>" tests
769
770
A few simplification and bugfixes to PCRE interface.
771
@@ -807,26 +840,3 @@ of the repositories listed at
840
On hold.
841
cf. <20190508001252.15752-1-avarab@gmail.com>
842
cf. <20190719210156.GA9688@archbookpro.localdomain>
810
-
811
-
812
---------------------------------------------------
813
-[Discarded]
814
-
815
-* ab/no-kwset (2019-07-01) 10 commits
816
- (merged to 'next' on 2019-07-15 at ed0479ce3d)
817
- + grep: use PCRE v2 for optimized fixed-string search
818
- + grep: remove the kwset optimization
819
- + grep: drop support for \0 in --fixed-strings <pattern>
820
- + grep: make the behavior for NUL-byte in patterns sane
821
- + grep tests: move binary pattern tests into their own file
822
- + grep tests: move "grep binary" alongside the rest
823
- + grep: inline the return value of a function call used only once
824
- + t4210: skip more command-line encoding tests on MinGW
825
- + grep: don't use PCRE2?_UTF8 with "log --encoding=<non-utf8>"
826
- + log tests: test regex backends in "--encode=<enc>" tests
827
- (this branch is used by ab/pcre-jit-fixes.)
828
-
829
- Retire use of kwset library, which is an optimization for looking
830
- for fixed strings, with use of pcre2 JIT.
831
-
832
- Kicked out of 'next' to give the topic a chance to get rebooted.