What's cooking (2021/04 #03)
Junio C Hamano committed
Apr 12, 2021 at 16:58 UTC
432c4f6cfdd725dce2e17ef228cbedebcc61c636
1 file changed
+279
-201
whats-cooking.txt
+279
-201
@@ -1,10 +1,10 @@
1
To: git@vger.kernel.org
2
Bcc: lwn@lwn.net
3
-Subject: What's cooking in git.git (Apr 2021, #02; Thu, 8)
3
+Subject: What's cooking in git.git (Apr 2021, #03; Mon, 12)
4
X-master-at: 89b43f80a514aee58b662ad606e6352e03eaeee4
5
-X-next-at: be11c130afeab11780014fb555625b10f13c8463
5
+X-next-at: 7d1e84936f59976b1fce260a447d8781a07cd620
6
7
-What's cooking in git.git (Apr 2021, #02; Thu, 8)
7
+What's cooking in git.git (Apr 2021, #03; Mon, 12)
8
--------------------------------------------------
9
10
Here are the topics that have been cooking. Commits prefixed with '-' are
@@ -41,174 +41,249 @@ Release tarballs are available at:
41
https://www.kernel.org/pub/software/scm/git/
42
43
--------------------------------------------------
44
-[Graduated to 'master']
44
+[New Topics]
45
46
-* ab/fsck-api-cleanup (2021-03-28) 19 commits
47
- (merged to 'next' on 2021-03-31 at 30c3303dba)
48
- + fetch-pack: use new fsck API to printing dangling submodules
49
- + fetch-pack: use file-scope static struct for fsck_options
50
- + fetch-pack: don't needlessly copy fsck_options
51
- + fsck.c: move gitmodules_{found,done} into fsck_options
52
- + fsck.c: add an fsck_set_msg_type() API that takes enums
53
- + fsck.c: pass along the fsck_msg_id in the fsck_error callback
54
- + fsck.[ch]: move FOREACH_FSCK_MSG_ID & fsck_msg_id from *.c to *.h
55
- + fsck.c: give "FOREACH_MSG_ID" a more specific name
56
- + fsck.c: undefine temporary STR macro after use
57
- + fsck.c: call parse_msg_type() early in fsck_set_msg_type()
58
- + fsck.h: re-order and re-assign "enum fsck_msg_type"
59
- + fsck.h: move FSCK_{FATAL,INFO,ERROR,WARN,IGNORE} into an enum
60
- + fsck.c: refactor fsck_msg_type() to limit scope of "int msg_type"
61
- + fsck.c: rename remaining fsck_msg_id "id" to "msg_id"
62
- + fsck.c: remove (mostly) redundant append_msg_id() function
63
- + fsck.c: rename variables in fsck_set_msg_type() for less confusion
64
- + fsck.h: use "enum object_type" instead of "int"
65
- + fsck.h: use designed initializers for FSCK_OPTIONS_{DEFAULT,STRICT}
66
- + fsck.c: refactor and rename common config callback
46
+* jt/fetch-pack-request-fix (2021-04-08) 1 commit
47
+ - fetch-pack: buffer object-format with other args
48
+ (this branch is used by jt/push-negotiation.)
49
68
- Fsck API clean-up.
50
+ Will merge to 'next'.
51
52
71
-* ab/make-tags-quiet (2021-04-01) 1 commit
72
- (merged to 'next' on 2021-04-01 at 9ea45b61b8)
73
- + Makefile: add QUIET_GEN to "tags" and "TAGS" targets
53
+* jt/push-negotiation (2021-04-08) 6 commits
54
+ - send-pack: support push negotiation
55
+ - fetch: teach independent negotiation (no packfile)
56
+ - fetch-pack: refactor command and capability write
57
+ - fetch-pack: refactor add_haves()
58
+ - fetch-pack: refactor process_acks()
59
+ - Merge branch 'jt/fetch-pack-request-fix' into jt/push-negotiation
60
+ (this branch uses jt/fetch-pack-request-fix.)
61
75
- Generate [ec]tags under $(QUIET_GEN).
62
+ "git push" learns to discover common ancestor with the receiving
63
+ end over protocol v2.
64
65
+ Waiting for reviews.
66
78
-* ah/plugleaks (2021-03-21) 9 commits
79
- (merged to 'next' on 2021-03-30 at 78636ede89)
80
- + transport: also free remote_refs in transport_disconnect()
81
- + parse-options: don't leak alias help messages
82
- + parse-options: convert bitfield values to use binary shift
83
- + init-db: silence template_dir leak when converting to absolute path
84
- + init: remove git_init_db_config() while fixing leaks
85
- + worktree: fix leak in dwim_branch()
86
- + clone: free or UNLEAK further pointers when finished
87
- + reset: free instead of leaking unneeded ref
88
- + symbolic-ref: don't leak shortened refname in check_symref()
67
+* ab/doc-lint (2021-04-10) 7 commits
68
+ - docs: fix linting issues due to incorrect relative section order
69
+ - doc lint: lint relative section order
70
+ - doc lint: lint and fix missing "GIT" end sections
71
+ - doc lint: fix bugs in, simplify and improve lint script
72
+ - doc lint: Perl "strict" and "warnings" in lint-gitlink.perl
73
+ - Documentation/Makefile: make doc.dep dependencies a variable again
74
+ - Documentation/Makefile: make $(wildcard howto/*.txt) a var
75
+
76
+ Dev support.
77
+
78
+ Waiting for reviews to conclude.
79
+
80
+
81
+* ab/rebase-no-reschedule-failed-exec (2021-04-10) 2 commits
82
+ - rebase: don't override --no-reschedule-failed-exec with config
83
+ - rebase tests: camel-case rebase.rescheduleFailedExec consistently
84
+
85
+ "git rebase --[no-]reschedule-failed-exec" did not work well with
86
+ its configuration variable, which has been corrected.
87
+
88
+ Waiting for reviews to conclude.
89
+
90
+
91
+* ah/plugleaks (2021-04-10) 9 commits
92
+ - builtin/rm: avoid leaking pathspec and seen
93
+ - builtin/rebase: release git_format_patch_opt too
94
+ - builtin/for-each-ref: free filter and UNLEAK sorting.
95
+ - builtin/checkout: clear pending objects after diffing
96
+ - builtin/check-ignore: clear_pathspec before returning
97
+ - builtin/bugreport: don't leak prefixed filename
98
+ - branch: FREE_AND_NULL instead of NULL'ing real_ref
99
+ - bloom: clear each bloom_key after use
100
+ - wt-status: fix multiple small leaks
101
+
102
+ Plug various leans reported by LSAN.
103
+
104
+ Waiting for reroll.
105
+ What is queued here is a subset of the original series that
106
+ looked (at least to me) to be obviously correct.
107
+
108
+
109
+* bc/hash-transition-interop-part-1 (2021-04-10) 16 commits
110
+ - fixup! hash: provide per-algorithm null OIDs
111
+ - hex: print objects using the hash algorithm member
112
+ - hex: default to the_hash_algo on zero algorithm value
113
+ - builtin/pack-objects: avoid using struct object_id for pack hash
114
+ - commit-graph: don't store file hashes as struct object_id
115
+ - builtin/show-index: set the algorithm for object IDs
116
+ - hash: provide per-algorithm null OIDs
117
+ - hash: set and copy algo field in struct object_id
118
+ - cache: compare the entire buffer for struct object_id
119
+ - builtin/pack-redundant: avoid casting buffers to struct object_id
120
+ - Use the final_oid_fn to finalize hashing of object IDs
121
+ - hash: add a function to finalize object IDs
122
+ - Always use oidread to read into struct object_id
123
+ - cache: add an algo member to struct object_id
124
+ - builtin/hash-object: allow literally hashing with a given algorithm
125
+ - sha1-file: allow hashing objects literally with any algorithm
126
+
127
+ SHA-256 transition.
128
+
129
+ Seems to break the tests when merged to 'seen'.
130
+
131
+
132
+* ps/rev-list-object-type-filter (2021-04-12) 8 commits
133
+ - rev-list: allow filtering of provided items
134
+ - pack-bitmap: implement combined filter
135
+ - pack-bitmap: implement object type filter
136
+ - list-objects: implement object type filter
137
+ - list-objects: support filtering by tag and commit
138
+ - list-objects: move tag processing into its own function
139
+ - revision: mark commit parents as NOT_USER_GIVEN
140
+ - uploadpack.txt: document implication of `uploadpackfilter.allow`
141
+
142
+ "git rev-list" learns the "--filter=object:type=<type>" option,
143
+ which can be used to exclude objects of the given kind from the
144
+ packfile generated by pack-objects.
145
+
146
+ Waiting for reviews to conclude.
147
+ cf. <cover.1618234575.git.ps@pks.im>
148
+
149
+
150
+* tb/multi-pack-bitmaps (2021-04-10) 23 commits
151
+ - p5326: perf tests for MIDX bitmaps
152
+ - p5310: extract full and partial bitmap tests
153
+ - midx: respect 'GIT_TEST_MULTI_PACK_INDEX_WRITE_BITMAP'
154
+ - t7700: update to work with MIDX bitmap test knob
155
+ - t5319: don't write MIDX bitmaps in t5319
156
+ - t5310: disable GIT_TEST_MULTI_PACK_INDEX_WRITE_BITMAP
157
+ - t5326: test multi-pack bitmap behavior
158
+ - t/helper/test-read-midx.c: add --checksum mode
159
+ - t5310: move some tests to lib-bitmap.sh
160
+ - pack-bitmap: write multi-pack bitmaps
161
+ - pack-bitmap: read multi-pack bitmaps
162
+ - pack-bitmap.c: introduce 'bitmap_is_preferred_refname()'
163
+ - pack-bitmap.c: introduce 'nth_bitmap_object_oid()'
164
+ - pack-bitmap.c: introduce 'bitmap_num_objects()'
165
+ - midx: respect 'core.multiPackIndex' when writing
166
+ - midx: clear auxiliary .rev after replacing the MIDX
167
+ - midx: make a number of functions non-static
168
+ - Documentation: describe MIDX-based bitmaps
169
+ - Documentation: build 'technical/bitmap-format' by default
170
+ - pack-bitmap-write.c: free existing bitmaps
171
+ - pack-bitmap-write.c: gracefully fail to write non-closed bitmaps
172
+ - pack-bitmap.c: harden 'test_bitmap_walk()' to check type bitmaps
173
+ - Merge branch 'tb/pack-preferred-tips-to-give-bitmap' into tb/multi-pack-bitmaps
174
+ (this branch uses tb/pack-preferred-tips-to-give-bitmap.)
175
+
176
+ The reachability bitmap file used to be generated only for a single
177
+ pack, but now we've learned to generate bitmaps for history that
178
+ span across multiple packfiles.
179
90
- Plug or annotate remaining leaks that trigger while running the
91
- very basic set of tests.
180
+ Waiting for reviews.
181
+ cf. <cover.1617991824.git.me@ttaylorr.com>
182
183
94
-* cc/downcase-opt-help (2021-03-29) 1 commit
95
- (merged to 'next' on 2021-03-31 at 0df5c647b9)
96
- + column, range-diff: downcase option description
97
-
98
- A few option description strings started with capital letters,
99
- which were corrected.
100
-
101
-
102
-* ds/clarify-hashwrite (2021-03-26) 1 commit
103
- (merged to 'next' on 2021-03-30 at 701f5c0696)
104
- + csum-file: make hashwrite() more readable
184
+* ab/svn-tests-set-e-fix (2021-04-12) 2 commits
185
+ - svn tests: refactor away a "set -e" in test body
186
+ - svn tests: remove legacy re-setup from init-clone test
187
106
- The hashwrite() API uses a buffering mechanism to avoid calling
107
- write(2) too frequently. This logic has been refactored to be
108
- easier to understand.
188
+ Test clean-up.
189
190
+ Waiting for reviews.
191
111
-* en/ort-perf-batch-9 (2021-03-10) 8 commits
112
- (merged to 'next' on 2021-04-01 at cc5254aadc)
113
- + diffcore-rename: avoid doing basename comparisons for irrelevant sources
114
- + merge-ort: skip rename detection entirely if possible
115
- + merge-ort: use relevant_sources to filter possible rename sources
116
- + merge-ort: precompute whether directory rename detection is needed
117
- + merge-ort: introduce wrappers for alternate tree traversal
118
- + merge-ort: add data structures for an alternate tree traversal
119
- + merge-ort: precompute subset of sources for which we need rename detection
120
- + diffcore-rename: enable filtering possible rename sources
121
- (this branch is used by en/ort-perf-batch-10, en/ort-perf-batch-11 and en/ort-readiness.)
192
123
- The ort merge backend has been optimized by skipping irrelevant
124
- renames.
193
+* ab/test-lib-updates (2021-04-12) 16 commits
194
+ - test-lib-functions: remove last two parameter count assertions
195
+ - Revert and amend "test-lib-functions: assert correct parameter count"
196
+ - test-lib-functions: use "return 1" instead of "false"
197
+ - test-lib-functions: normalize test_path_is_missing() debugging
198
+ - test-lib: modernize test_create_repo() function
199
+ - test-lib: do not show advice about init.defaultBranch under --verbose
200
+ - test-lib: reformat argument list in test_create_repo()
201
+ - submodule tests: use symbolic-ref --short to discover branch name
202
+ - test-lib functions: add --printf option to test_commit
203
+ - describe tests: convert setup to use test_commit
204
+ - test-lib functions: add an --annotated-tag option to "test_commit"
205
+ - test-lib-functions: document test_commit --no-tag
206
+ - test-lib-functions: reword "test_commit --append" docs
207
+ - test-lib tests: remove dead GIT_TEST_FRAMEWORK_SELFTEST variable
208
+ - test-lib: bring $remove_trash out of retirement
209
+ - check-non-portable-shell: complain about "test" a/-o instead of &&/||
210
+ (this branch is used by ab/describe-tests-fix and ab/pickaxe-pcre2.)
211
212
+ Test clean-up.
213
127
-* en/sequencer-edit-upon-conflict-fix (2021-03-31) 1 commit
128
- (merged to 'next' on 2021-04-01 at 2e3c705a73)
129
- + sequencer: fix edit handling for cherry-pick and revert messages
214
+ Waiting for reviews.
215
131
- "git cherry-pick/revert" with or without "--[no-]edit" did not spawn
132
- the editor as expected (e.g. "revert --no-edit" after a conflict
133
- still asked to edit the message), which has been corrected.
216
217
+* ah/merge-ort-ubsan-fix (2021-04-12) 1 commit
218
+ - merge-ort: only do pointer arithmetic for non-empty lists
219
136
-* js/cmake-vsbuild (2021-03-29) 4 commits
137
- (merged to 'next' on 2021-03-31 at e0c4369f65)
138
- + cmake(install): include vcpkg dlls
139
- + cmake: add a preparatory work-around to accommodate `vcpkg`
140
- + cmake(install): fix double .exe suffixes
141
- + cmake: support SKIP_DASHED_BUILT_INS
220
+ Code clean-up for merge-ort backend.
221
143
- CMake update for vsbuild.
222
+ Will merge to 'next'?
223
224
146
-* js/security-md (2021-03-27) 2 commits
147
- (merged to 'next' on 2021-03-31 at 28f64fee38)
148
- + Document how we do embargoed releases
149
- + SECURITY: describe how to report vulnerabilities
225
+* ao/p4-avoid-decoding (2021-04-12) 2 commits
226
+ - git-p4: do not decode data from perforce by default
227
+ - git-p4: avoid decoding more data from perforce
228
151
- SECURITY.md that is facing individual contributors and end users
152
- has been introduced. Also a procedure to follow when preparing
153
- embargoed releases has been spelled out.
229
+ "git p4" in Python-2 days used to accept a lot more kinds of data
230
+ from Perforce server as uninterrupted byte sequence, but after
231
+ switching to Python-3, too many things are expected to be in UTF-8,
232
+ which broke traditional use cases.
233
234
+ Waiting for reviews.
235
156
-* ll/clone-reject-shallow (2021-04-01) 1 commit
157
- (merged to 'next' on 2021-04-01 at 9b13372e1a)
158
- + builtin/clone.c: add --reject-shallow option
236
160
- "git clone --reject-shallow" option fails the clone as soon as we
161
- notice that we are cloning from a shallow repository.
237
+* hn/refs-trace-errno (2021-04-12) 1 commit
238
+ - refs: print errno for read_raw_ref if GIT_TRACE_REFS is set
239
240
+ Show errno in the trace output in the error codepath that calls
241
+ read_raw_ref method.
242
164
-* ps/pack-bitmap-optim (2021-03-22) 1 commit
165
- (merged to 'next' on 2021-03-31 at 12bb775d9a)
166
- + pack-bitmap: avoid traversal of objects referenced by uninteresting tag
243
+ Waiting for reviews to conclude.
244
+ cf. <xmqq4kgbb2ic.fsf@gitster.g>
245
168
- Optimize "rev-list --use-bitmap-index --objects" corner case that
169
- uses negative tags as the stopping points.
246
247
+* hn/reftable-tables-doc-update (2021-04-12) 1 commit
248
+ - reftable: document an alternate cleanup method on Windows
249
172
-* rs/daemon-sanitize-dir-sep (2021-03-26) 1 commit
173
- (merged to 'next' on 2021-04-01 at c8f43b1c8d)
174
- + daemon: sanitize all directory separators
250
+ Doc updte.
251
+
252
+ Will merge to 'next'.
253
+
254
+
255
+* jk/pack-objects-bitmap-progress-fix (2021-04-12) 1 commit
256
+ - pack-objects: update "nr_seen" progress based on pack-reused count
257
+
258
+ When "git pack-objects" makes a literal copy of a part of existing
259
+ packfile using the reachability bitmaps, its update to the progress
260
+ meter was broken.
261
+
262
+ Will merge to 'next'.
263
176
- "git daemon" has been tightened against systems that take backslash
177
- as directory separator.
264
265
+* ma/t0091-bugreport-fix (2021-04-12) 1 commit
266
+ - t0091-bugreport.sh: actually verify some content of report
267
180
-* tb/reverse-midx (2021-04-01) 17 commits
181
- (merged to 'next' on 2021-04-01 at 17c9542606)
182
- + midx.c: improve cache locality in midx_pack_order_cmp()
183
- + pack-revindex: write multi-pack reverse indexes
184
- + pack-write.c: extract 'write_rev_file_order'
185
- + pack-revindex: read multi-pack reverse indexes
186
- + Documentation/technical: describe multi-pack reverse indexes
187
- + midx: make some functions non-static
188
- + midx: keep track of the checksum
189
- + midx: don't free midx_name early
190
- + midx: allow marking a pack as preferred
191
- + t/helper/test-read-midx.c: add '--show-objects'
192
- + builtin/multi-pack-index.c: display usage on unrecognized command
193
- + builtin/multi-pack-index.c: don't enter bogus cmd_mode
194
- + builtin/multi-pack-index.c: split sub-commands
195
- + builtin/multi-pack-index.c: define common usage with a macro
196
- + builtin/multi-pack-index.c: don't handle 'progress' separately
197
- + builtin/multi-pack-index.c: inline 'flags' with options
198
- + Merge branch 'ds/chunked-file-api' into tb/reverse-midx
268
+ Test fix.
269
200
- An on-disk reverse-index to map the in-pack location of an object
201
- back to its object name across multiple packfiles is introduced.
270
+ Waiting for an Ack.
271
272
204
-* zh/commit-trailer (2021-03-23) 1 commit
205
- (merged to 'next' on 2021-03-31 at f17d196ee1)
206
- + commit: add --trailer option
207
- (this branch is used by zh/trailer-cmd.)
273
+* ps/config-global-override (2021-04-12) 3 commits
274
+ - config: allow overriding of global and system configuration
275
+ - config: unify code paths to get global config paths
276
+ - config: rename `git_etc_config()`
277
209
- "git commit" learned "--trailer <key>[=<value>]" option; together
210
- with the interpret-trailers command, this will make it easier to
211
- support custom trailers.
278
+ Replace GIT_CONFIG_NOSYSTEM mechanism to decline from reading the
279
+ system-wide configuration file with GIT_CONFIG_SYSTEM that lets
280
+ users specify from which file to read the system-wide configuration
281
+ (setting it to an empty file would essentially be the same as
282
+ setting NOSYSTEM), and introduce GIT_CONFIG_GLOBAL to override the
283
+ per-user configuration in $HOME/.gitconfig.
284
+
285
+ Expecting a (hopefully final) reroll.
286
+ cf. <xmqq4kgbfn8o.fsf@gitster.g>
287
288
--------------------------------------------------
289
[Stalled]
@@ -270,29 +345,24 @@ Release tarballs are available at:
345
blob", which is all that we can reliably say.
346
347
273
-* ab/describe-tests-fix (2021-03-01) 10 commits
274
- - test-lib: return 1 from test_expect_{success,failure}
275
- - svn tests: refactor away a "set -e" in test body
276
- - svn tests: remove legacy re-setup from init-clone test
348
+* ab/describe-tests-fix (2021-04-12) 5 commits
349
- describe tests: support -C in "check_describe"
350
- describe tests: fix nested "test_expect_success" call
279
- - describe tests: convert setup to use test_commit
280
- - test-lib functions: add an --annotated-tag option to "test_commit"
281
- - describe tests: always assert empty stderr from "describe"
351
+ - describe tests: don't rely on err.actual from "check_describe"
352
- describe tests: refactor away from glob matching
353
- describe tests: improve test for --work-tree & --dirty
354
+ (this branch uses ab/test-lib-updates.)
355
356
Various updates to tests around "git describe"
357
287
- Expecting a reroll.
288
- cf. <xmqq1rcj6hzr.fsf@gitster.g>
289
-
358
291
-* ab/pickaxe-pcre2 (2021-02-18) 24 commits
359
+* ab/pickaxe-pcre2 (2021-04-12) 22 commits
360
+ - xdiff-interface: replace discard_hunk_line() with a flag
361
+ - xdiff users: use designated initializers for out_line
362
- pickaxe -G: don't special-case create/delete
363
- pickaxe -G: terminate early on matching lines
294
- - xdiff-interface: support early exit in xdiff_outf()
295
- - xdiff-interface: allow early return from xdiff_emit_{line,hunk}_fn
364
+ - xdiff-interface: allow early return from xdiff_emit_line_fn
365
+ - xdiff-interface: prepare for allowing early return
366
- pickaxe -S: slightly optimize contains()
367
- pickaxe: rename variables in has_changes() for brevity
368
- pickaxe -S: support content with NULs under --pickaxe-regex
@@ -303,23 +373,17 @@ Release tarballs are available at:
373
- diff.h: move pickaxe fields together again
374
- pickaxe: die when --find-object and --pickaxe-all are combined
375
- pickaxe: die when -G and --pickaxe-regex are combined
376
+ - pickaxe tests: add missing test for --no-pickaxe-regex being an error
377
- pickaxe tests: test for -G, -S and --find-object incompatibility
378
- pickaxe tests: add test for "log -S" not being a regex
379
- pickaxe tests: add test for diffgrep_consume() internals
380
- pickaxe tests: refactor to use test_commit --append --printf
310
- - test-lib functions: add --printf option to test_commit
311
- - test-lib-functions: reword "test_commit --append" docs
312
- - test-lib-functions: document and test test_commit --no-tag
381
- grep/pcre2 tests: reword comments referring to kwset
314
- - Merge branch 'jk/rev-list-disk-usage' into ab/pickaxe-pcre2
315
- - Merge branch 'ab/test-lib' into ab/pickaxe-pcre2
382
+ (this branch uses ab/test-lib-updates.)
383
384
Rewrite the backend for "diff -G/-S" to use pcre2 engine when
385
available.
386
320
- Needs (hopefully the final round of) reroll and its review.
321
- cf. <xmqqh7ksb5j7.fsf@gitster.g>, <xmqqa6qkb5fi.fsf@gitster.g>, etc.
322
-
387
388
* es/config-hooks (2021-03-10) 36 commits
389
. run-command: stop thinking about hooks
@@ -362,7 +426,7 @@ Release tarballs are available at:
426
The "hooks defined in config" topic.
427
428
--------------------------------------------------
365
-[New Topics]
429
+[Cooking]
430
431
* cc/test-helper-bloom-usage-fix (2021-04-05) 1 commit
432
(merged to 'next' on 2021-04-06 at 4895f33fee)
@@ -374,7 +438,8 @@ Release tarballs are available at:
438
439
440
* jz/apply-run-3way-first (2021-04-06) 1 commit
377
- - git-apply: try threeway first when "--3way" is used
441
+ (merged to 'next' on 2021-04-08 at f826bcc6b8)
442
+ + git-apply: try threeway first when "--3way" is used
443
(this branch is used by jz/apply-3way-cached.)
444
445
"git apply --3way" has always been "to fall back to 3-way merge
@@ -383,14 +448,12 @@ Release tarballs are available at:
448
is given, of course) and then straight patch application is used as
449
a fallback when it fails.
450
386
- Will merge to 'next'.
451
+ Will merge to 'master'.
452
453
389
-* ds/maintenance-prefetch-fix (2021-04-06) 5 commits
390
- - maintenance: allow custom refspecs during prefetch
391
- - test-tool: test refspec input/output
392
- - refspec: output a refspec item
393
- - test-lib: use exact match for test_subcommand
454
+* ds/maintenance-prefetch-fix (2021-04-10) 3 commits
455
+ - maintenance: use 'git fetch --prefetch'
456
+ - fetch: add --prefetch option
457
- maintenance: simplify prefetch logic
458
459
The prefetch task in "git maintenance" assumed that "git fetch"
@@ -399,19 +462,20 @@ Release tarballs are available at:
462
branches there.
463
464
Expecting a reroll.
402
- cf. <35fa99cf-ba24-f13e-0607-f9acce2905cf@gmail.com>
465
+ cf. <e7f09182-3b75-f2a3-c9c7-0055cc377a32@gmail.com>
466
467
468
* ab/complete-cherry-pick-head (2021-04-07) 1 commit
406
- - bash completion: complete CHERRY_PICK_HEAD
469
+ (merged to 'next' on 2021-04-09 at 5810d0e2c2)
470
+ + bash completion: complete CHERRY_PICK_HEAD
471
472
The command line completion (in contrib/) has learned that
473
CHERRY_PICK_HEAD is a possible pseudo-ref.
474
411
- Will merge to 'next'?
475
+ Will merge to 'master'.
476
477
414
-* mr/bisect-in-c-4 (2021-04-07) 4 commits
478
+* mr/bisect-in-c-4 (2021-04-11) 4 commits
479
- bisect--helper: retire `--bisect-next-check` subcommand
480
- bisect--helper: reimplement `bisect_run` shell function in C
481
- bisect--helper: reimplement `bisect_visualize()`shell function in C
@@ -420,36 +484,39 @@ Release tarballs are available at:
484
The codepaths involved in running "git bisect visualize" and "git
485
bisect run" has been rewritten in C.
486
423
- Expecting a reroll.
424
- cf. <xmqqo8epydwx.fsf@gitster.g>, <xmqqeeflycf9.fsf@gitster.g>
487
+
488
+ cf. <xmqq35vwh8qk.fsf@gitster.g>, <xmqqy2doftrj.fsf@gitster.g>
489
490
491
* ab/detox-config-gettext (2021-04-08) 1 commit
428
- - config.c: remove last remnant of GIT_TEST_GETTEXT_POISON
492
+ (merged to 'next' on 2021-04-08 at 01d6ff5069)
493
+ + config.c: remove last remnant of GIT_TEST_GETTEXT_POISON
494
495
The last remnant of gettext-poison has been removed.
496
432
- Will merge to 'next'.
497
+ Will merge to 'master'.
498
499
500
* jz/apply-3way-cached (2021-04-07) 1 commit
436
- - git-apply: allow simultaneous --cached and --3way options
501
+ (merged to 'next' on 2021-04-09 at 7d1e84936f)
502
+ + git-apply: allow simultaneous --cached and --3way options
503
(this branch uses jz/apply-run-3way-first.)
504
505
"git apply" now takes "--3way" and "--cached" at the same time, and
506
work and record results only in the index.
507
442
- Will merge to 'next'.
508
+ Will merge to 'master'.
509
510
511
* vs/completion-with-set-u (2021-04-08) 1 commit
446
- - completion: audit and guard $GIT_* against unset use
512
+ (merged to 'next' on 2021-04-08 at b9b8bde89d)
513
+ + completion: audit and guard $GIT_* against unset use
514
515
The command-line completion script (in contrib/) had a couple of
516
references that would have given a warning under the "-u" (nounset)
517
option.
518
452
- Will merge to 'next'.
519
+ Will merge to 'master'.
520
521
522
* sg/bugreport-fixes (2021-04-08) 1 commit
@@ -459,10 +526,9 @@ Release tarballs are available at:
526
when the former was split out of the latter, which has been
527
corrected.
528
462
- Will merge to 'next'.
529
+ Expecting a reroll.
530
+ cf. <20210409212044.GD2947267@szeder.dev>
531
464
---------------------------------------------------
465
-[Cooking]
532
533
* mt/add-rm-in-sparse-checkout (2021-04-08) 7 commits
534
- rm: honor sparse checkout patterns
@@ -518,6 +584,7 @@ Release tarballs are available at:
584
+ builtin/pack-objects.c: respect 'pack.preferBitmapTips'
585
+ t/helper/test-bitmap.c: initial commit
586
+ pack-bitmap: add 'test_bitmap_commits()' helper
587
+ (this branch is used by tb/multi-pack-bitmaps.)
588
589
A configuration variable has been added to force tips of certain
590
refs to be given a reachability bitmap.
@@ -535,8 +602,9 @@ Release tarballs are available at:
602
Will merge to 'master'.
603
604
538
-* zh/trailer-cmd (2021-04-02) 1 commit
539
- - trailer: add new trailer.<token>.cmd config option
605
+* zh/trailer-cmd (2021-04-12) 2 commits
606
+ - trailer: add new .cmd config option
607
+ - docs: correct descript of trailer.<token>.command
608
609
The way the command line specified by the trailer.<token>.command
610
configuration variable receives the end-user supplied value was
@@ -544,8 +612,10 @@ Release tarballs are available at:
612
same goal in a safer and more intuitive way has been added, as
613
the trailer.<token>.cmd configuration variable, to replace it.
614
547
- Expecting a reroll.
548
- cf. <CAP8UFD08Lagij6oqpOJkpbzMLRgQ08=mDfiyf1qXyD5VW6Ae7A@mail.gmail.com>
615
+ Waiting for reviews.
616
+ I doubt it is a good idea to copy the "run once even without being
617
+ asked" from .command interface.
618
+ cf. <xmqqwnt7b5fg.fsf@gitster.g>
619
620
621
* fm/user-manual-use-preface (2021-04-03) 1 commit
@@ -606,8 +676,9 @@ Release tarballs are available at:
676
Will merge to 'next'?
677
678
609
-* gk/gitweb-redacted-email (2021-03-31) 1 commit
610
- - gitweb: add "e-mail privacy" feature to redact e-mail addresses
679
+* gk/gitweb-redacted-email (2021-04-08) 1 commit
680
+ (merged to 'next' on 2021-04-08 at 8a19c3cd6e)
681
+ + gitweb: add "e-mail privacy" feature to redact e-mail addresses
682
683
"gitweb" learned "e-mail privacy" feature to redact strings that
684
look like e-mail addresses on various pages.
@@ -675,9 +746,10 @@ Release tarballs are available at:
746
- unpack-trees: add basic support for parallel checkout
747
748
678
-* ds/sparse-index-protections (2021-03-31) 25 commits
749
+* ds/sparse-index-protections (2021-04-12) 26 commits
750
- name-hash: use expand_to_path()
751
- sparse-index: expand_to_path()
752
+ - name-hash: don't add directories to name_hash
753
- revision: ensure full index
754
- resolve-undo: ensure full index
755
- read-cache: ensure full index
@@ -709,34 +781,40 @@ Release tarballs are available at:
781
782
783
* en/ort-readiness (2021-03-20) 13 commits
712
- - Add testing with merge-ort merge strategy
713
- - t6423: mark remaining expected failure under merge-ort as such
714
- - Revert "merge-ort: ignore the directory rename split conflict for now"
715
- - merge-recursive: add a bunch of FIXME comments documenting known bugs
716
- - merge-ort: write $GIT_DIR/AUTO_MERGE whenever we hit a conflict
717
- - t: mark several submodule merging tests as fixed under merge-ort
718
- - merge-ort: implement CE_SKIP_WORKTREE handling with conflicted entries
719
- - t6428: new test for SKIP_WORKTREE handling and conflicts
720
- - merge-ort: support subtree shifting
721
- - merge-ort: let renormalization change modify/delete into clean delete
722
- - merge-ort: have ll_merge() use a special attr_index for renormalization
723
- - merge-ort: add a special minimal index just for renormalization
724
- - merge-ort: use STABLE_QSORT instead of QSORT where required
784
+ (merged to 'next' on 2021-04-08 at 20283a3f10)
785
+ + Add testing with merge-ort merge strategy
786
+ + t6423: mark remaining expected failure under merge-ort as such
787
+ + Revert "merge-ort: ignore the directory rename split conflict for now"
788
+ + merge-recursive: add a bunch of FIXME comments documenting known bugs
789
+ + merge-ort: write $GIT_DIR/AUTO_MERGE whenever we hit a conflict
790
+ + t: mark several submodule merging tests as fixed under merge-ort
791
+ + merge-ort: implement CE_SKIP_WORKTREE handling with conflicted entries
792
+ + t6428: new test for SKIP_WORKTREE handling and conflicts
793
+ + merge-ort: support subtree shifting
794
+ + merge-ort: let renormalization change modify/delete into clean delete
795
+ + merge-ort: have ll_merge() use a special attr_index for renormalization
796
+ + merge-ort: add a special minimal index just for renormalization
797
+ + merge-ort: use STABLE_QSORT instead of QSORT where required
798
(this branch is used by en/ort-perf-batch-11; uses en/ort-perf-batch-10.)
799
800
Plug the ort merge backend throughout the rest of the system, and
801
start testing it as a replacement for the recursive backend.
802
730
- Will merge to 'next'?
803
+ Will merge to 'next'.
804
805
733
-* hn/reftable (2021-03-12) 15 commits
806
+* hn/reftable (2021-04-12) 20 commits
807
- Add "test-tool dump-reftable" command.
808
- git-prompt: prepare for reftable refs backend
809
- Reftable support for git-core
737
- - reftable: rest of library
810
+ - reftable: add dump utility
811
+ - reftable: implement stack, a mutable database of reftable files.
812
+ - reftable: implement refname validation
813
+ - reftable: add merged table view
814
+ - reftable: add a heap-based priority queue for reftable records
815
- reftable: reftable file level tests
816
- reftable: read reftable files
817
+ - reftable: generic interface to tables
818
- reftable: write reftable files
819
- reftable: a generic binary tree implementation
820
- reftable: reading/writing blocks
@@ -749,7 +827,7 @@ Release tarballs are available at:
827
828
The "reftable" backend for the refs API.
829
752
- What's the status of this topic?
830
+ Waiting for reviews.
831
832
--------------------------------------------------
833
[Discarded]