What's cooking (2020/07 #06)
Junio C Hamano committed
Jul 30, 2020 at 16:45 UTC
93d331e2abc3aa7e50a0413fea259939601d1d4e
1 file changed
+300
-274
whats-cooking.txt
+300
-274
@@ -1,10 +1,10 @@
1
To: git@vger.kernel.org
2
Bcc: lwn@lwn.net
3
-Subject: What's cooking in git.git (Jul 2020, #05; Mon, 27)
4
-X-master-at: 47ae905ffb98cc4d4fd90083da6bc8dab55d9ecc
5
-X-next-at: 6104cc2f0b60537336b3f2ee6fc82132e2b84cd3
3
+Subject: What's cooking in git.git (Jul 2020, #06; Thu, 30)
4
+X-master-at: 79bcaf006215b91de981c30c0268467e06faaf39
5
+X-next-at: 392f21bd87383fb9d4b700d1d69cb3af52040784
6
7
-What's cooking in git.git (Jul 2020, #05; Mon, 27)
7
+What's cooking in git.git (Jul 2020, #06; Thu, 30)
8
--------------------------------------------------
9
10
Here are the topics that have been cooking. Commits prefixed with '-' are
@@ -12,13 +12,9 @@ only in 'seen' (formerly 'pu'---proposed updates) while commits prefixed
12
with '+' 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 2.28 has been tagged. We'll go slow for a few days to gauge the
16
-initial reaction to it and see if there is no major brown-paper-bag
17
-issues discovered. Following that, majority of topics that have
18
-been cooking in 'next' will be merged to 'master', the tip of 'next'
19
-will be rewound and rebuilt on top of the updated 'master', near the
20
-end of this week. The 'next' branch then will start taking new
21
-topics after that.
15
+Many topics that have been cooking in 'next' have now been merged to
16
+'master'. The tip of 'next' will be rewound and rebuilt on top of
17
+the updated 'master', perhaps tomorrow.
18
19
During the next cycle for 2.29, we will decide the direction of
20
follow-up work, if any is needed, for the init.defaultBranch topic
@@ -29,43 +25,283 @@ repositories listed at
25
26
http://git-blame.blogspot.com/p/git-public-repositories.html
27
28
+--------------------------------------------------
29
+[Graduated to 'master']
30
+
31
+* bc/push-cas-cquoted-refname (2020-07-20) 1 commit
32
+ (merged to 'next' on 2020-07-21 at fd7bb5c85e)
33
+ + remote-curl: make --force-with-lease work with non-ASCII ref names
34
+
35
+ Pushing a ref whose name contains non-ASCII character with the
36
+ "--force-with-lease" option did not work over smart HTTP protocol,
37
+ which has been corrected.
38
+
39
+
40
+* bw/fail-cloning-into-non-empty (2020-07-10) 1 commit
41
+ (merged to 'next' on 2020-07-14 at a52b5ce7ab)
42
+ + git clone: don't clone into non-empty directory
43
+
44
+ "git clone --separate-git-dir=$elsewhere" used to stomp on the
45
+ contents of the existing directory $elsewhere, which has been
46
+ taught to fail when $elsewhere is not an empty directory.
47
+
48
+
49
+* cc/pretty-contents-size (2020-07-16) 3 commits
50
+ (merged to 'next' on 2020-07-20 at e7c4b8f9ae)
51
+ + ref-filter: add support for %(contents:size)
52
+ + t6300: test refs pointing to tree and blob
53
+ + Documentation: clarify %(contents:XXXX) doc
54
+
55
+ "git for-each-ref --format=<>" learned %(contents:size).
56
+
57
+
58
+* ct/mv-unmerged-path-error (2020-07-20) 1 commit
59
+ (merged to 'next' on 2020-07-21 at c17007237a)
60
+ + git-mv: improve error message for conflicted file
61
+
62
+ "git mv src dst", when src is an unmerged path, errored out
63
+ correctly but with an incorrect error message to claim that src is
64
+ not tracked, which has been clarified.
65
+
66
+
67
+* dl/test-must-fail-fixes-6 (2020-07-07) 6 commits
68
+ (merged to 'next' on 2020-07-20 at bd4a9946a6)
69
+ + test-lib-functions: restrict test_must_fail usage
70
+ + t9400: don't use test_must_fail with cvs
71
+ + t9834: remove use of `test_might_fail p4`
72
+ + t7107: don't use test_must_fail()
73
+ + t5324: reorder `run_with_limited_open_files test_might_fail`
74
+ + t3701: stop using `env` in force_color()
75
+
76
+ Dev support to limit the use of test_must_fail to only git commands.
77
+
78
+
79
+* ds/commit-graph-bloom-updates (2020-07-01) 10 commits
80
+ (merged to 'next' on 2020-07-06 at 177e6b362e)
81
+ + commit-graph: check all leading directories in changed path Bloom filters
82
+ + revision: empty pathspecs should not use Bloom filters
83
+ + revision.c: fix whitespace
84
+ + commit-graph: check chunk sizes after writing
85
+ + commit-graph: simplify chunk writes into loop
86
+ + commit-graph: unify the signatures of all write_graph_chunk_*() functions
87
+ + commit-graph: persist existence of changed-paths
88
+ + bloom: fix logic in get_bloom_filter()
89
+ + commit-graph: change test to die on parse, not load
90
+ + commit-graph: place bloom_settings in context
91
+ (this branch uses sg/commit-graph-cleanups.)
92
+
93
+ Updates to the changed-paths bloom filter.
94
+
95
+
96
+* en/fill-directory-exponential (2020-07-20) 1 commit
97
+ (merged to 'next' on 2020-07-22 at 63662d6646)
98
+ + dir: check pathspecs before returning `path_excluded`
99
+
100
+ Fix to a regression introduced during 2.27 cycle.
101
+
102
+
103
+* hn/reftable (2020-07-10) 4 commits
104
+ (merged to 'next' on 2020-07-14 at e524883fec)
105
+ + reflog: cleanse messages in the refs.c layer
106
+ + bisect: treat BISECT_HEAD as a pseudo ref
107
+ + t3432: use git-reflog to inspect the reflog for HEAD
108
+ + lib-t6000.sh: write tag using git-update-ref
109
+ (this branch is used by hn/reftable-prep-part-2.)
110
+
111
+ Preliminary clean-up of the refs API in preparation for adding a
112
+ new refs backend "reftable".
113
+
114
+
115
+* jk/reject-newer-extensions-in-v0 (2020-07-16) 1 commit
116
+ (merged to 'next' on 2020-07-16 at 85f634d5a6)
117
+ + verify_repository_format(): complain about new extensions in v0 repo
118
+
119
+ With the base fix to 2.27 regresion, any new extensions in a v0
120
+ repository would still be silently honored, which is not quite
121
+ right. Instead, complain and die loudly.
122
+
123
+
124
+* jk/tests-timestamp-fix (2020-07-15) 6 commits
125
+ (merged to 'next' on 2020-07-16 at 2c69e97b92)
126
+ + t9100: stop depending on commit timestamps
127
+ (merged to 'next' on 2020-07-14 at 66ee0e6b45)
128
+ + test-lib: set deterministic default author/committer date
129
+ + t9100: explicitly unset GIT_COMMITTER_DATE
130
+ + t5539: make timestamp requirements more explicit
131
+ + t9700: loosen ident timezone regex
132
+ (merged to 'next' on 2020-07-09 at 633bcd552f)
133
+ + t6000: use test_tick consistently
134
+
135
+ The test framework has been updated so that most tests will run
136
+ with predictable (artificial) timestamps.
137
+
138
+
139
+* jt/avoid-lazy-fetching-upon-have-check (2020-07-16) 1 commit
140
+ (merged to 'next' on 2020-07-20 at 28e85288e5)
141
+ + upload-pack: do not lazy-fetch "have" objects
142
+
143
+ Fetching from a lazily cloned repository resulted at the server
144
+ side in attempts to lazy fetch objects that the client side has,
145
+ many of which will not be available from the third-party anyway.
146
+
147
+
148
+* pb/log-rev-list-doc (2020-07-08) 6 commits
149
+ (merged to 'next' on 2020-07-14 at f1104dfb77)
150
+ + git-log.txt: include rev-list-description.txt
151
+ + git-rev-list.txt: move description to separate file
152
+ + git-rev-list.txt: tweak wording in set operations
153
+ + git-rev-list.txt: fix Asciidoc syntax
154
+ + revisions.txt: describe 'rev1 rev2 ...' meaning for ranges
155
+ + git-log.txt: add links to 'rev-list' and 'diff' docs
156
+
157
+ "git help log" has been enhanced by sharing more material from the
158
+ documentation for the underlying "git rev-list" command.
159
+
160
+
161
+* rs/add-index-entry-optim-fix (2020-07-16) 1 commit
162
+ (merged to 'next' on 2020-07-20 at 617fe6aecc)
163
+ + read-cache: remove bogus shortcut
164
+
165
+ Fix to an ancient bug caused by an over-eager attempt for
166
+ optimization.
167
+
168
+
169
+* sg/commit-graph-cleanups (2020-06-08) 10 commits
170
+ (merged to 'next' on 2020-07-06 at 15c9d77eb9)
171
+ + commit-graph: simplify write_commit_graph_file() #2
172
+ + commit-graph: simplify write_commit_graph_file() #1
173
+ + commit-graph: simplify parse_commit_graph() #2
174
+ + commit-graph: simplify parse_commit_graph() #1
175
+ + commit-graph: clean up #includes
176
+ + diff.h: drop diff_tree_oid() & friends' return value
177
+ + commit-slab: add a function to deep free entries on the slab
178
+ + commit-graph-format.txt: all multi-byte numbers are in network byte order
179
+ + commit-graph: fix parsing the Chunk Lookup table
180
+ + tree-walk.c: don't match submodule entries for 'submod/anything'
181
+ (this branch is used by ds/commit-graph-bloom-updates.)
182
+
183
+ The changed-path Bloom filter is improved using ideas from an
184
+ independent implementation.
185
+
186
--------------------------------------------------
187
[New Topics]
188
35
-* jb/doc-packfile-name (2020-07-22) 1 commit
36
- - pack-write/docs: update regarding pack naming
189
+* en/typofixes (2020-07-28) 2 commits
190
+ (merged to 'next' on 2020-07-30 at 64776daa9a)
191
+ + hashmap: fix typo in usage docs
192
+ + Remove doubled words in various comments
193
38
- Doc update.
194
+ Will merge to 'master'.
195
+
196
+
197
+* jk/log-fp-implies-m (2020-07-29) 7 commits
198
+ - doc/git-log: clarify handling of merge commit diffs
199
+ - doc/git-log: move "-t" into diff-options list
200
+ - doc/git-log: drop "-r" diff option
201
+ - doc/git-log: move "Diff Formatting" from rev-list-options
202
+ - log: enable "-m" automatically with "--first-parent"
203
+ - revision: add "--no-diff-merges" option to counteract "-m"
204
+ - log: drop "--cc implies -m" logic
205
+
206
+ "git log --first-parent -p" showed patches only for single-parent
207
+ commits on the first-parent chain; the "--first-parent" option has
208
+ been made to imply "-m". Use "--no-diff-merges" to restore the
209
+ previous behaviour to omit patches for merge commits.
210
211
Will merge to 'next'.
212
213
43
-* dd/send-email-config (2020-07-23) 1 commit
44
- - git-send-email: die if sendmail.* config is set
214
+* jk/strvec (2020-07-28) 11 commits
215
+ - strvec: rename struct fields
216
+ - strvec: drop argv_array compatibility layer
217
+ - strvec: update documention to avoid argv_array
218
+ - strvec: fix indentation in renamed calls
219
+ - strvec: convert remaining callers away from argv_array name
220
+ - strvec: convert more callers away from argv_array name
221
+ - strvec: convert builtin/ callers away from argv_array name
222
+ - quote: rename sq_dequote_to_argv_array to mention strvec
223
+ - strvec: rename files from argv-array to strvec
224
+ - argv-array: rename to strvec
225
+ - argv-array: use size_t for count and alloc
226
+
227
+ The argv_array API is useful for not just managing argv but any
228
+ "vector" (NULL-terminated array) of strings, and has seen adoption
229
+ to a certain degree. It has been renamed to "strvec" to reduce the
230
+ barrier to adoption.
231
46
- Stop when "sendmail.*" configuration variables are defined, which
47
- could be a mistaken attempt to define "sendemail.*" variables.
232
+ Will merge to 'next'.
233
+
234
+
235
+* rs/grep-simpler-parse-object-or-die-call (2020-07-28) 1 commit
236
+ (merged to 'next' on 2020-07-30 at 6d22dd3058)
237
+ + grep: avoid using oid_to_hex() with parse_object_or_die()
238
+
239
+ Will merge to 'master'.
240
+
241
+
242
+* ar/help-guides-doc (2020-07-29) 1 commit
243
+ (merged to 'next' on 2020-07-30 at e4b0370bfa)
244
+ + git-help.txt: fix mentions of option --guides
245
+
246
+ Will merge to 'master'.
247
+
248
+
249
+* sk/typofixes (2020-07-29) 1 commit
250
+ (merged to 'next' on 2020-07-30 at c56d9e5313)
251
+ + comment: fix spelling mistakes inside comments
252
+
253
+ Will merge to 'master'.
254
+
255
+
256
+* pd/mergetool-nvimdiff (2020-07-29) 2 commits
257
+ - mergetools: add support for nvimdiff (neovim) family
258
+ - mergetool--lib: improve support for vimdiff-style tool variants
259
+
260
+ The existing backends for "git mergetool" based on variants of vim
261
+ have been refactored and then support for "nvim" has been added.
262
+
263
+
264
+* al/bisect-first-parent (2020-07-29) 3 commits
265
+ - bisect: combine args passed to find_bisection()
266
+ - bisect: introduce first-parent flag
267
+ - rev-list: allow bisect and first-parent flags
268
+
269
+ "git bisect" learns the "--first-parent" option to find the first
270
+ breakage along the first-parent chain.
271
+
272
+
273
+* jc/fmt-merge-msg-suppress-destination (2020-07-30) 2 commits
274
+ (merged to 'next' on 2020-07-30 at c44f57f46d)
275
+ + fmt-merge-msg: allow merge destination to be omitted again
276
+ + Revert "fmt-merge-msg: stop treating `master` specially"
277
+
278
+ "git merge" learned to selectively omit " into <branch>" at the end
279
+ of the title of default merge message with merge.suppressDest
280
+ configuration.
281
+
282
+ Will merge to 'master'.
283
284
--------------------------------------------------
285
[Stalled]
286
287
* jx/proc-receive-hook (2020-05-18) 11 commits
53
- - doc: add documentation for the proc-receive hook
54
- - transport: parse report options for tracking refs
55
- - t5411: test updates of remote-tracking branches
56
- - receive-pack: new config receive.procReceiveRefs
57
- - refs.c: refactor to reuse ref_is_hidden()
58
- - receive-pack: feed report options to post-receive
59
- - doc: add document for capability report-status-v2
60
- - New capability "report-status-v2" for git-push
61
- - receive-pack: add new proc-receive hook
62
- - t5411: add basic test cases for proc-receive hook
63
- - transport: not report a non-head push as a branch
288
+ . doc: add documentation for the proc-receive hook
289
+ . transport: parse report options for tracking refs
290
+ . t5411: test updates of remote-tracking branches
291
+ . receive-pack: new config receive.procReceiveRefs
292
+ . refs.c: refactor to reuse ref_is_hidden()
293
+ . receive-pack: feed report options to post-receive
294
+ . doc: add document for capability report-status-v2
295
+ . New capability "report-status-v2" for git-push
296
+ . receive-pack: add new proc-receive hook
297
+ . t5411: add basic test cases for proc-receive hook
298
+ . transport: not report a non-head push as a branch
299
300
"git receive-pack" that accepts requests by "git push" learned to
301
outsource most of the ref updates to the new "proc-receive" hook.
302
68
- Needs review.
303
+ Ejected out of 'seen'; somehow its tests seem to break with clang
304
+ cf. https://travis-ci.org/github/git/git/builds/713443572
305
306
307
* mf/submodule-summary-with-correct-repository (2020-06-24) 2 commits
@@ -93,24 +329,26 @@ repositories listed at
329
330
331
* mr/bisect-in-c-2 (2020-07-17) 14 commits
96
- - SQUASH??? do not add new users of git_path_bisect_head()
97
- - bisect--helper: retire `--bisect-autostart` subcommand
98
- - bisect--helper: retire `--write-terms` subcommand
99
- - bisect--helper: retire `--check-expected-revs` subcommand
100
- - bisect--helper: reimplement `bisect_state` & `bisect_head` shell functions in C
101
- - bisect--helper: retire `--next-all` subcommand
102
- - bisect--helper: retire `--bisect-clean-state` subcommand
103
- - bisect--helper: finish porting `bisect_start()` to C
104
- - bisect--helper: reimplement `bisect_next` and `bisect_auto_next` shell functions in C
105
- - bisect: call 'clear_commit_marks_all()' in 'bisect_next_all()'
106
- - bisect--helper: reimplement `bisect_autostart` shell function in C
107
- - bisect--helper: introduce new `write_in_file()` function
108
- - bisect--helper: use '-res' in 'cmd_bisect__helper' return
109
- - bisect--helper: BUG() in cmd_*() on invalid subcommand
332
+ . SQUASH??? do not add new users of git_path_bisect_head()
333
+ . bisect--helper: retire `--bisect-autostart` subcommand
334
+ . bisect--helper: retire `--write-terms` subcommand
335
+ . bisect--helper: retire `--check-expected-revs` subcommand
336
+ . bisect--helper: reimplement `bisect_state` & `bisect_head` shell functions in C
337
+ . bisect--helper: retire `--next-all` subcommand
338
+ . bisect--helper: retire `--bisect-clean-state` subcommand
339
+ . bisect--helper: finish porting `bisect_start()` to C
340
+ . bisect--helper: reimplement `bisect_next` and `bisect_auto_next` shell functions in C
341
+ . bisect: call 'clear_commit_marks_all()' in 'bisect_next_all()'
342
+ . bisect--helper: reimplement `bisect_autostart` shell function in C
343
+ . bisect--helper: introduce new `write_in_file()` function
344
+ . bisect--helper: use '-res' in 'cmd_bisect__helper' return
345
+ . bisect--helper: BUG() in cmd_*() on invalid subcommand
346
347
Rewrite of the remainder of "git bisect" script in C continues.
348
349
Needs more work.
350
+ Ejected out of 'seen'; al/bisect-first-parent topic has a bit of
351
+ textual conflict with this topic.
352
353
354
* mk/use-size-t-in-zlib (2018-10-15) 1 commit
@@ -123,46 +361,20 @@ repositories listed at
361
--------------------------------------------------
362
[Cooking]
363
126
-* jk/reject-newer-extensions-in-v0 (2020-07-16) 1 commit
127
- (merged to 'next' on 2020-07-16 at 85f634d5a6)
128
- + verify_repository_format(): complain about new extensions in v0 repo
129
-
130
- With the base fix to 2.27 regresion, any new extensions in a v0
131
- repository would still be silently honored, which is not quite
132
- right. Instead, complain and die loudly.
133
-
134
- Will cook in 'next'.
135
-
136
-
137
-* en/fill-directory-exponential (2020-07-20) 1 commit
138
- (merged to 'next' on 2020-07-22 at 63662d6646)
139
- + dir: check pathspecs before returning `path_excluded`
140
-
141
- Fix to a regression introduced during 2.27 cycle.
142
-
143
- Will cook in 'next'.
144
-
145
-
146
-* bc/push-cas-cquoted-refname (2020-07-20) 1 commit
147
- (merged to 'next' on 2020-07-21 at fd7bb5c85e)
148
- + remote-curl: make --force-with-lease work with non-ASCII ref names
149
-
150
- Pushing a ref whose name contains non-ASCII character with the
151
- "--force-with-lease" option did not work over smart HTTP protocol,
152
- which has been corrected.
364
+* jb/doc-packfile-name (2020-07-22) 1 commit
365
+ (merged to 'next' on 2020-07-30 at b46c3f6675)
366
+ + pack-write/docs: update regarding pack naming
367
154
- Will cook in 'next'.
368
+ Doc update.
369
370
+ Will merge to 'master'.
371
157
-* ct/mv-unmerged-path-error (2020-07-20) 1 commit
158
- (merged to 'next' on 2020-07-21 at c17007237a)
159
- + git-mv: improve error message for conflicted file
372
161
- "git mv src dst", when src is an unmerged path, errored out
162
- correctly but with an incorrect error message to claim that src is
163
- not tracked, which has been clarified.
373
+* dd/send-email-config (2020-07-23) 1 commit
374
+ - git-send-email: die if sendmail.* config is set
375
165
- Will cook in 'next'.
376
+ Stop when "sendmail.*" configuration variables are defined, which
377
+ could be a mistaken attempt to define "sendemail.*" variables.
378
379
380
* jt/pack-objects-prefetch-in-batch (2020-07-21) 2 commits
@@ -191,22 +403,23 @@ repositories listed at
403
404
405
* sg/ci-git-path-fix-with-pyenv (2020-07-23) 1 commit
194
- - ci: use absolute PYTHON_PATH in the Linux jobs
406
+ (merged to 'next' on 2020-07-30 at afe304633d)
407
+ + ci: use absolute PYTHON_PATH in the Linux jobs
408
409
CI fixup---tests of Python scripts didn't use the version of Git
410
that is being tested.
411
199
- Will merge to 'next'.
412
+ Will merge to 'master'.
413
414
202
-* bc/sha-256-part-3 (2020-07-22) 39 commits
415
+* bc/sha-256-part-3 (2020-07-30) 39 commits
416
- t: remove test_oid_init in tests
417
- docs: add documentation for extensions.objectFormat
418
- ci: run tests with SHA-256
419
- t: make SHA1 prerequisite depend on default hash
420
- t: allow testing different hash algorithms via environment
421
- t: add test_oid option to select hash algorithm
209
- - Enable SHA-256 support by default
422
+ - repository: enable SHA-256 support by default
423
- setup: add support for reading extensions.objectformat
424
- bundle: add new version for use with SHA-256
425
- builtin/verify-pack: implement an --object-format option
@@ -242,25 +455,10 @@ repositories listed at
455
456
The final leg of SHA-256 transition.
457
245
-
246
-* jk/tests-timestamp-fix (2020-07-15) 6 commits
247
- (merged to 'next' on 2020-07-16 at 2c69e97b92)
248
- + t9100: stop depending on commit timestamps
249
- (merged to 'next' on 2020-07-14 at 66ee0e6b45)
250
- + test-lib: set deterministic default author/committer date
251
- + t9100: explicitly unset GIT_COMMITTER_DATE
252
- + t5539: make timestamp requirements more explicit
253
- + t9700: loosen ident timezone regex
254
- (merged to 'next' on 2020-07-09 at 633bcd552f)
255
- + t6000: use test_tick consistently
256
-
257
- The test framework has been updated so that most tests will run
258
- with predictable (artificial) timestamps.
259
-
260
- Will cook in 'next'.
458
+ Will merge to 'next'.
459
460
263
-* jc/no-update-fetch-head (2020-07-13) 1 commit
461
+* jc/no-update-fetch-head (2020-07-29) 1 commit
462
- fetch: optionally allow disabling FETCH_HEAD update
463
464
"git fetch" learned the "--[no-]write-fetch-head" option to
@@ -278,50 +476,14 @@ repositories listed at
476
A follow-up patch to reduce duplication may be warranted.
477
478
281
-* hn/reftable-prep-part-2 (2020-07-16) 3 commits
479
+* hn/reftable-prep-part-2 (2020-07-27) 3 commits
480
- Make HEAD a PSEUDOREF rather than PER_WORKTREE.
481
- Modify pseudo refs through ref backend storage
482
- t1400: use git rev-parse for testing PSEUDOREF existence
285
- (this branch uses hn/reftable.)
483
484
Further preliminary change to refs API.
485
486
290
-* jt/avoid-lazy-fetching-upon-have-check (2020-07-16) 1 commit
291
- (merged to 'next' on 2020-07-20 at 28e85288e5)
292
- + upload-pack: do not lazy-fetch "have" objects
293
-
294
- Fetching from a lazily cloned repository resulted at the server
295
- side in attempts to lazy fetch objects that the client side has,
296
- many of which will not be available from the third-party anyway.
297
-
298
- Will cook in 'next'.
299
-
300
-
301
-* rs/add-index-entry-optim-fix (2020-07-16) 1 commit
302
- (merged to 'next' on 2020-07-20 at 617fe6aecc)
303
- + read-cache: remove bogus shortcut
304
-
305
- Fix to an ancient bug caused by an over-eager attempt for
306
- optimization.
307
-
308
- Will cook in 'next'.
309
-
310
-
311
-* dl/test-must-fail-fixes-6 (2020-07-07) 6 commits
312
- (merged to 'next' on 2020-07-20 at bd4a9946a6)
313
- + test-lib-functions: restrict test_must_fail usage
314
- + t9400: don't use test_must_fail with cvs
315
- + t9834: remove use of `test_might_fail p4`
316
- + t7107: don't use test_must_fail()
317
- + t5324: reorder `run_with_limited_open_files test_might_fail`
318
- + t3701: stop using `env` in force_color()
319
-
320
- Dev support to limit the use of test_must_fail to only git commands.
321
-
322
- Will cook in 'next'.
323
-
324
-
487
* ds/maintenance (2020-07-23) 18 commits
488
- maintenance: add trace2 regions for task execution
489
- midx: use start_delayed_progress()
@@ -347,28 +509,6 @@ repositories listed at
509
cleaning.
510
511
350
-* bw/fail-cloning-into-non-empty (2020-07-10) 1 commit
351
- (merged to 'next' on 2020-07-14 at a52b5ce7ab)
352
- + git clone: don't clone into non-empty directory
353
-
354
- "git clone --separate-git-dir=$elsewhere" used to stomp on the
355
- contents of the existing directory $elsewhere, which has been
356
- taught to fail when $elsewhere is not an empty directory.
357
-
358
- Will cook in 'next'.
359
-
360
-
361
-* cc/pretty-contents-size (2020-07-16) 3 commits
362
- (merged to 'next' on 2020-07-20 at e7c4b8f9ae)
363
- + ref-filter: add support for %(contents:size)
364
- + t6300: test refs pointing to tree and blob
365
- + Documentation: clarify %(contents:XXXX) doc
366
-
367
- "git for-each-ref --format=<>" learned %(contents:size).
368
-
369
- Will cook in 'next'.
370
-
371
-
512
* ls/mergetool-meld-auto-merge (2020-07-12) 2 commits
513
- SQUASH???
514
- Support auto-merge for meld to follow the vim-diff behavior
@@ -380,21 +520,6 @@ repositories listed at
520
Expecting a reroll.
521
522
383
-* pb/log-rev-list-doc (2020-07-08) 6 commits
384
- (merged to 'next' on 2020-07-14 at f1104dfb77)
385
- + git-log.txt: include rev-list-description.txt
386
- + git-rev-list.txt: move description to separate file
387
- + git-rev-list.txt: tweak wording in set operations
388
- + git-rev-list.txt: fix Asciidoc syntax
389
- + revisions.txt: describe 'rev1 rev2 ...' meaning for ranges
390
- + git-log.txt: add links to 'rev-list' and 'diff' docs
391
-
392
- "git help log" has been enhanced by sharing more material from the
393
- documentation for the underlying "git rev-list" command.
394
-
395
- Will cook in 'next'.
396
-
397
-
523
* tb/upload-pack-filters (2020-07-22) 4 commits
524
- upload-pack.c: introduce 'uploadpackfilter.tree.maxDepth'
525
- upload-pack.c: pass 'struct list_objects_filter_options *'
@@ -418,25 +543,6 @@ repositories listed at
543
cf. <CAHd-oW6A2aBHg80R9kyifvF7thwzam5EYYoN9d2TaBcHJrcKWw@mail.gmail.com>
544
545
421
-* ds/commit-graph-bloom-updates (2020-07-01) 10 commits
422
- (merged to 'next' on 2020-07-06 at 177e6b362e)
423
- + commit-graph: check all leading directories in changed path Bloom filters
424
- + revision: empty pathspecs should not use Bloom filters
425
- + revision.c: fix whitespace
426
- + commit-graph: check chunk sizes after writing
427
- + commit-graph: simplify chunk writes into loop
428
- + commit-graph: unify the signatures of all write_graph_chunk_*() functions
429
- + commit-graph: persist existence of changed-paths
430
- + bloom: fix logic in get_bloom_filter()
431
- + commit-graph: change test to die on parse, not load
432
- + commit-graph: place bloom_settings in context
433
- (this branch uses sg/commit-graph-cleanups.)
434
-
435
- Updates to the changed-paths bloom filter.
436
-
437
- Will cook in 'next'.
438
-
439
-
546
* ss/cmake-build (2020-06-26) 8 commits
547
- ci: modification of main.yml to use cmake for vs-build job
548
- cmake: support for building git on windows with msvc and clang.
@@ -453,27 +559,7 @@ repositories listed at
559
cf. https://github.com/git/git/runs/892824895
560
561
456
-* sg/commit-graph-cleanups (2020-06-08) 10 commits
457
- (merged to 'next' on 2020-07-06 at 15c9d77eb9)
458
- + commit-graph: simplify write_commit_graph_file() #2
459
- + commit-graph: simplify write_commit_graph_file() #1
460
- + commit-graph: simplify parse_commit_graph() #2
461
- + commit-graph: simplify parse_commit_graph() #1
462
- + commit-graph: clean up #includes
463
- + diff.h: drop diff_tree_oid() & friends' return value
464
- + commit-slab: add a function to deep free entries on the slab
465
- + commit-graph-format.txt: all multi-byte numbers are in network byte order
466
- + commit-graph: fix parsing the Chunk Lookup table
467
- + tree-walk.c: don't match submodule entries for 'submod/anything'
468
- (this branch is used by ds/commit-graph-bloom-updates.)
469
-
470
- The changed-path Bloom filter is improved using ideas from an
471
- independent implementation.
472
-
473
- Will cook in 'next'.
474
-
475
-
476
-* es/config-hooks (2020-05-21) 4 commits
562
+* es/config-hooks (2020-07-29) 4 commits
563
- hook: add --porcelain to list command
564
- hook: add list command
565
- hook: scaffolding for git-hook subcommand
@@ -510,50 +596,9 @@ repositories listed at
596
Review needed on 4/6; otherwise looking sane.
597
cf. <CABPp-BGdEyEeajYZj_rdxp=MyEQdszuyjVTax=hhYj3fOtRQUQ@mail.gmail.com>
598
513
-
514
-* hn/reftable (2020-07-10) 4 commits
515
- (merged to 'next' on 2020-07-14 at e524883fec)
516
- + reflog: cleanse messages in the refs.c layer
517
- + bisect: treat BISECT_HEAD as a pseudo ref
518
- + t3432: use git-reflog to inspect the reflog for HEAD
519
- + lib-t6000.sh: write tag using git-update-ref
520
- (this branch is used by hn/reftable-prep-part-2.)
521
-
522
- Preliminary clean-up of the refs API in preparation for adding a
523
- new refs backend "reftable".
524
-
525
- Will cook in 'next'.
526
-
599
--------------------------------------------------
600
[Discarded]
601
530
-* jc/credential-store-file-format-doc (2020-04-27) 1 commit
531
- . credential-store: document the file format a bit more
532
-
533
- Now has become a part of Carlo's credential-store fix patches.
534
-
535
-
536
-* js/ci-skip-on-github-workflow (2020-05-02) 1 commit
537
- . ci: respect the [skip ci] convention in our GitHub workflow "CI/PR"
538
-
539
- Allow contributors to mark a branch/push that it does not have to
540
- be built via GitHub actions, in a way similar to how Travis lets
541
- them mark the commits with an embedded "[skip ci]" string.
542
-
543
- Superseded by dd/ci-only-on-selective-branches topic.
544
-
545
-
546
-* dd/ci-only-on-selective-branches (2020-05-05) 2 commits
547
- - CI: limit GitHub Actions to designated branches
548
- - SubmittingPatches: advertise GitHub Actions CI
549
-
550
- Instead of always building all branches of all forks of our project
551
- at GitHub via GitHub Actions, only build when branches with known
552
- and specific names are updated, and also a pull request.
553
-
554
- Superseded by jk/ci-only-on-selected-branches
555
-
556
-
602
* jk/fast-export-anonym (2020-06-22) 4 commits
603
(merged to 'next' on 2020-06-22 at b517b2f707)
604
+ fast-export: allow dumping the path mapping
@@ -574,22 +619,3 @@ repositories listed at
619
Test update.
620
621
Now it is part of jk/tests-timestamp-fix with a larger scope.
577
-
578
-
579
-* ds/upgrade-with-existing-extension (2020-07-14) 2 commits
580
- . config: provide extra detail about worktree config
581
- . setup: tweak upgrade policy to grandfather worktreeconfig
582
- (this branch is used by jc/upgrade-v0-fixup.)
583
-
584
- In 2.28-rc0, we corrected a bug that some repository extensions are
585
- honored by mistake even in a version 0 repositories (these
586
- configuration variables in extensions.* namespace were supposed to
587
- have special meaning in repositories whose version numbers are 1 or
588
- higher), but this was a bit too big a change.
589
-
590
- Superseded by jn/v0-with-extensions-fix
591
-
592
-
593
-* jc/upgrade-v0-fixup (2020-07-15) 1 commit
594
- . setup: grandfather other extensions that used to be honored by mistake
595
- (this branch uses ds/upgrade-with-existing-extension.)