What's cooking (2020/03 #09)
Junio C Hamano committed
Mar 29, 2020 at 12:24 UTC
2edc701201c4bdba20351d52bf1a368e21c77c3b
1 file changed
+390
-293
whats-cooking.txt
+390
-293
@@ -1,10 +1,10 @@
1
To: git@vger.kernel.org
2
Bcc: lwn@lwn.net
3
-Subject: What's cooking in git.git (Mar 2020, #08; Wed, 25)
4
-X-master-at: a7d14a44285d3ec4b25bf9e3b7df701221350661
5
-X-next-at: 33ef6b2f381b0c2d6c1eba5609779d5de2162ad9
3
+Subject: What's cooking in git.git (Mar 2020, #09; Sun, 29)
4
+X-master-at: 9fadedd637b312089337d73c3ed8447e9f0aa775
5
+X-next-at: 6d04fd2bf2c2653e09ae1d75aa42c176da447be8
6
7
-What's cooking in git.git (Mar 2020, #08; Wed, 25)
7
+What's cooking in git.git (Mar 2020, #09; Sun, 29)
8
--------------------------------------------------
9
10
Here are the topics that have been cooking. Commits prefixed with
@@ -23,128 +23,394 @@ of the repositories listed at
23
--------------------------------------------------
24
[Graduated to "master"]
25
26
-* am/real-path-fix (2020-03-10) 4 commits
27
- (merged to 'next' on 2020-03-11 at 1f843820a7)
28
- + get_superproject_working_tree(): return strbuf
29
- + real_path_if_valid(): remove unsafe API
30
- + real_path: remove unsafe API
31
- + set_git_dir: fix crash when used with real_path()
26
+* ah/force-pull-rebase-configuration (2020-03-10) 1 commit
27
+ (merged to 'next' on 2020-03-11 at c79dbbaf9c)
28
+ + pull: warn if the user didn't say whether to rebase or to merge
29
33
- The real_path() convenience function can easily be misused; with a
34
- bit of code refactoring in the callers' side, its use has been
35
- eliminated.
30
+ "git pull" learned to warn when no pull.rebase configuration
31
+ exists, and neither --[no-]rebase nor --ff-only is given (which
32
+ would result a merge).
33
34
38
-* ds/check-connected-reprepare-packed-git (2020-03-15) 1 commit
39
- (merged to 'next' on 2020-03-16 at 609a322030)
40
- + connected.c: reprepare packs for corner cases
35
+* at/rebase-fork-point-regression-fix (2020-02-11) 1 commit
36
+ (merged to 'next' on 2020-03-02 at a1a84d37a7)
37
+ + rebase: --fork-point regression fix
38
42
- Corner case "git fetch" fix.
39
+ The "--fork-point" mode of "git rebase" regressed when the command
40
+ was rewritten in C back in 2.20 era, which has been corrected.
41
42
45
-* en/oidset-uninclude-hashmap (2020-03-15) 1 commit
46
- (merged to 'next' on 2020-03-16 at a79879866d)
47
- + oidset: remove unnecessary include
43
+* bc/filter-process (2020-03-16) 8 commits
44
+ (merged to 'next' on 2020-03-17 at 2cd9dbf794)
45
+ + t0021: test filter metadata for additional cases
46
+ + builtin/reset: compute checkout metadata for reset
47
+ + builtin/rebase: compute checkout metadata for rebases
48
+ + builtin/clone: compute checkout metadata for clones
49
+ + builtin/checkout: compute checkout metadata for checkouts
50
+ + convert: provide additional metadata to filters
51
+ + convert: permit passing additional metadata to filter processes
52
+ + builtin/checkout: pass branch info down to checkout_worktree
53
49
- Code clean-up.
54
+ Provide more information (e.g. the object of the tree-ish in which
55
+ the blob being converted appears, in addition to its path, which
56
+ has already been given) to smudge/clean conversion filters.
57
58
52
-* hw/advise-ng (2020-03-05) 4 commits
53
- (merged to 'next' on 2020-03-09 at ea9e5a1fa3)
54
- + tag: use new advice API to check visibility
55
- + advice: revamp advise API
56
- + advice: change "setupStreamFailure" to "setUpstreamFailure"
57
- + advice: extract vadvise() from advise()
59
+* bc/sha-256-part-1-of-4 (2020-02-28) 22 commits
60
+ (merged to 'next' on 2020-03-17 at 436c4e64a7)
61
+ + fast-import: add options for rewriting submodules
62
+ + fast-import: add a generic function to iterate over marks
63
+ + fast-import: make find_marks work on any mark set
64
+ + fast-import: add helper function for inserting mark object entries
65
+ + fast-import: permit reading multiple marks files
66
+ + commit: use expected signature header for SHA-256
67
+ + worktree: allow repository version 1
68
+ + init-db: move writing repo version into a function
69
+ + builtin/init-db: add environment variable for new repo hash
70
+ + builtin/init-db: allow specifying hash algorithm on command line
71
+ + setup: allow check_repository_format to read repository format
72
+ + t/helper: make repository tests hash independent
73
+ + t/helper: initialize repository if necessary
74
+ + t/helper/test-dump-split-index: initialize git repository
75
+ + t6300: make hash algorithm independent
76
+ + t6300: abstract away SHA-1-specific constants
77
+ + t: use hash-specific lookup tables to define test constants
78
+ + repository: require a build flag to use SHA-256
79
+ + hex: add functions to parse hex object IDs in any algorithm
80
+ + hex: introduce parsing variants taking hash algorithms
81
+ + hash: implement and use a context cloning function
82
+ + builtin/pack-objects: make hash agnostic
83
59
- Revamping of the advise API to allow more systematic enumeration of
60
- advice knobs in the future.
84
+ SHA-256 transition continues.
85
86
63
-* jc/config-tar (2020-03-18) 1 commit
64
- (merged to 'next' on 2020-03-19 at aa6216fd1a)
65
- + separate tar.* config to its own source file
87
+* ds/default-pack-use-sparse-to-true (2020-03-20) 2 commits
88
+ (merged to 'next' on 2020-03-21 at 25cc87784d)
89
+ + pack-objects: flip the use of GIT_TEST_PACK_SPARSE
90
+ + config: set pack.useSparse=true by default
91
67
- Improve the structure of the documentation source a bit.
92
+ The 'pack.useSparse' configuration variable now defaults to 'true',
93
+ enabling an optimization that has been experimental since Git 2.21.
94
95
70
-* jc/maintain-doc (2020-03-09) 1 commit
71
- (merged to 'next' on 2020-03-09 at 7f1a754f60)
72
- + update how-to-maintain-git
96
+* hi/gpg-prefer-check-signature (2020-03-15) 2 commits
97
+ (merged to 'next' on 2020-03-17 at 2def2d9a7e)
98
+ + gpg-interface: prefer check_signature() for GPG verification
99
+ + t: increase test coverage of signature verification output
100
74
- Doc update.
101
+ The code to interface with GnuPG has been refactored.
102
+
103
+
104
+* jc/describe-misnamed-annotated-tag (2020-02-20) 1 commit
105
+ (merged to 'next' on 2020-03-02 at b4e2ca6a46)
106
+ + describe: force long format for a name based on a mislocated tag
107
+
108
+ When "git describe C" finds an annotated tag with tagname A to be
109
+ the best name to explain commit C, and the tag is stored in a
110
+ "wrong" place in the refs/tags hierarchy, e.g. refs/tags/B, the
111
+ command gave a warning message but used A (not B) to describe C.
112
+ If C is exactly at the tag, the describe output would be "A", but
113
+ "git rev-parse A^0" would not be equal as "git rev-parse C^0". The
114
+ behavior of the command has been changed to use the "long" form
115
+ i.e. A-0-gOBJECTNAME, which is correctly interpreted by rev-parse.
116
+
117
+
118
+* pb/recurse-submodules-fix (2020-02-19) 6 commits
119
+ (merged to 'next' on 2020-03-17 at b46922ddd1)
120
+ + t/lib-submodule-update: add test removing nested submodules
121
+ + unpack-trees: check for missing submodule directory in merged_entry
122
+ + unpack-trees: remove outdated description for verify_clean_submodule
123
+ + t/lib-submodule-update: move a test to the right section
124
+ + t/lib-submodule-update: remove outdated test description
125
+ + t7112: remove mention of KNOWN_FAILURE_SUBMODULE_RECURSIVE_NESTED
126
+
127
+ Fix "git checkout --recurse-submodules" of a nested submodule
128
+ hierarchy.
129
+
130
+
131
+* tg/retire-scripted-stash (2020-03-05) 2 commits
132
+ (merged to 'next' on 2020-03-05 at 8e82eb9dec)
133
+ + stash: remove the stash.useBuiltin setting
134
+ + stash: get git_stash_config at the top level
135
136
+ "git stash" has kept an escape hatch to use the scripted version
137
+ for a few releases, which got stale. It has been removed.
138
77
-* js/https-proxy-config (2020-03-05) 2 commits
78
- (merged to 'next' on 2020-03-09 at 8a06f8501d)
79
- + http: add environment variable support for HTTPS proxies
80
- + http: add client cert support for HTTPS proxies
139
+--------------------------------------------------
140
+[New Topics]
141
82
- A handful of options to configure SSL when talking to proxies have
83
- been added.
142
+* dr/midx-avoid-int-underflow (2020-03-28) 1 commit
143
+ - midx.c: fix an integer underflow
144
145
+ When fed a midx that records no objects, some codepaths tried to
146
+ loop from 0 through (num_objects-1), which, due to integer
147
+ arithmetic wrapping around, made it nonsense operation with out of
148
+ bounds array accesses. The code has been corrected to reject such
149
+ an midx file.
150
86
-* pw/advise-rebase-skip (2019-12-06) 7 commits
87
- (merged to 'next' on 2020-03-11 at 80386de756)
88
- + commit: give correct advice for empty commit during a rebase
89
- + commit: encapsulate determine_whence() for sequencer
90
- + commit: use enum value for multiple cherry-picks
91
- + sequencer: write CHERRY_PICK_HEAD for reword and edit
92
- + cherry-pick: check commit error messages
93
- + cherry-pick: add test for `--skip` advice in `git commit`
94
- + t3404: use test_cmp_rev
151
+ Will merge to 'next'.
152
96
- The mechanism to prevent "git commit" from making an empty commit
97
- or amending during an interrupted cherry-pick was broken during the
98
- rewrite of "git rebase" in C, which has been corrected.
99
- cf. <xmqq7e0e7d9z.fsf@gitster-ct.c.googlers.com>
153
154
+* ak/run-command-on-cygwin-fix (2020-03-27) 1 commit
155
+ - run-command: trigger PATH lookup properly on Cygwin
156
102
-* rs/doc-passthru-fetch-options (2020-03-11) 1 commit
103
- (merged to 'next' on 2020-03-12 at f08dab7228)
104
- + pull: document more passthru options
157
+ Utitiles run via the run_command() API were not spawned correctly
158
+ on Cygwin, when the paths to them are given as a full path with
159
+ backslashes.
160
+
161
+ Will merge to 'next'.
162
106
- Doc update.
163
164
+* dd/ci-musl-libc (2020-03-29) 3 commits
165
+ - travis: build and test on Linux with musl libc and busybox
166
+ - ci: refactor docker runner script
167
+ - ci: libify logic for usage and checking CI_USER
168
109
-* sg/commit-slab-clarify-peek (2020-03-10) 1 commit
110
- (merged to 'next' on 2020-03-11 at 0496b26f23)
111
- + commit-slab: clarify slabname##_peek()'s return value
169
+ A new CI job to build and run test suite on linux with musl libc
170
+ has been added.
171
113
- In-code comment update.
172
173
+* dr/doc-recurse-submodules (2020-03-27) 5 commits
174
+ - doc: --recurse-submodules mostly applies to active submodules
175
+ - doc: be more precise on (fetch|push).recurseSubmodules
176
+ - doc: explain how to deactivate submodule.recurse completely
177
+ - doc: document --recurse-submodules for reset and restore
178
+ - doc: list all commands affected by submodule.recurse
179
+
180
+ Documentation updates around the "--recurse-submodules" option.
181
+
182
+ Will merge to 'next'.
183
116
-* ss/submodule-foreach-cb (2020-03-18) 1 commit
117
- (merged to 'next' on 2020-03-21 at 6651eafe02)
118
- + submodule--helper.c: Rename 'cb_foreach' to 'foreach_cb'
184
120
- Code clean-up.
185
+* dr/push-remoteref-fix (2020-03-27) 1 commit
186
+ - remote.c: fix handling of %(push:remoteref)
187
188
+ The "%(push:remoteref)" placeholder in the "--format=" argument of
189
+ "git format-patch" (and friends) only showed what got explicitly
190
+ configured, not what ref at the receiving end would be updated when
191
+ "git push" was used, as it ignored the default behaviour (e.g. update
192
+ the same ref as the source).
193
123
-* yz/p4-py3 (2020-03-10) 14 commits
124
- (merged to 'next' on 2020-03-11 at 01ca57c2b2)
125
- + ci: use python3 in linux-gcc and osx-gcc and python2 elsewhere
126
- + git-p4: use python3's input() everywhere
127
- + git-p4: simplify regex pattern generation for parsing diff-tree
128
- + git-p4: use dict.items() iteration for python3 compatibility
129
- + git-p4: use functools.reduce instead of reduce
130
- + git-p4: fix freezing while waiting for fast-import progress
131
- + git-p4: use marshal format version 2 when sending to p4
132
- + git-p4: open .gitp4-usercache.txt in text mode
133
- + git-p4: convert path to unicode before processing them
134
- + git-p4: encode/decode communication with git for python3
135
- + git-p4: encode/decode communication with p4 for python3
136
- + git-p4: remove string type aliasing
137
- + git-p4: change the expansion test from basestring to list
138
- + git-p4: make python2.7 the oldest supported version
194
+ Expecting a reroll.
195
+ cf. <20200328222546.gvrtzkcazf3bhjno@doriath>
196
140
- Update "git p4" to work with Python 3.
197
+
198
+* jc/doc-test-leaving-early (2020-03-29) 1 commit
199
+ - t/README: suggest how to leave test early with failure
200
+
201
+ Document the recommended way to abort a failing test early (e.g. by
202
+ exiting a loop), which is to say "return 1".
203
+
204
+ Will merge to 'next'.
205
+
206
+
207
+* jk/build-with-right-curl (2020-03-27) 2 commits
208
+ - Makefile: use curl-config --cflags
209
+ - Makefile: avoid running curl-config multiple times
210
+
211
+ The build procedure did not use the libcurl library and its include
212
+ files correctly for a custom-built installation.
213
+
214
+ Will merge to 'next'.
215
+
216
+
217
+* jk/harden-protocol-v2-delim-handling (2020-03-29) 3 commits
218
+ - test-lib-functions: simplify packetize() stdin code
219
+ - upload-pack: handle unexpected delim packets
220
+ - test-lib-functions: make packetize() more efficient
221
+
222
+ The server-end of the v2 protocol to serve "git clone" and "git
223
+ fetch" was not prepared to see a delim packets at unexpected
224
+ places, which led to a crash.
225
+
226
+ Will merge to 'next'.
227
+
228
+
229
+* jk/p5310-drop-non-bitmap-timing (2020-03-27) 1 commit
230
+ - p5310: stop timing non-bitmap pack-to-disk
231
+
232
+ Perf-test update.
233
+
234
+ Will merge to 'next'.
235
+
236
+
237
+* jk/test-cleanup (2020-03-27) 2 commits
238
+ - t/lib-*.sh: drop executable bit
239
+ - t/lib-credential.sh: drop shebang line
240
+
241
+ Test cleanup.
242
+
243
+ Will merge to 'next'.
244
+
245
+
246
+* ps/transactional-update-ref-stdin (2020-03-27) 9 commits
247
+ - update-ref: implement interactive transaction handling
248
+ - update-ref: read commands in a line-wise fashion
249
+ - update-ref: move transaction handling into `update_refs_stdin()`
250
+ - update-ref: pass end pointer instead of strbuf
251
+ - update-ref: drop unused argument for `parse_refname`
252
+ - update-ref: organize commands in an array
253
+ - strbuf: provide function to append whole lines
254
+ - git-update-ref.txt: add missing word
255
+ - refs: fix segfault when aborting empty transaction
256
+
257
+ "git update-ref --stdin" learned a handful of new verbs to let the
258
+ user control ref update transactions more explicitly, which helps
259
+ as an ingredient to implement two-phase commit-style atomic
260
+ ref-updates across multiple repositories.
261
+
262
+ Good start, but suspect not quite there yet.
263
+ cf. <xmqqo8she9yp.fsf@gitster.c.googlers.com>
264
+
265
+
266
+* ag/sequencer-i18n-messages (2020-03-28) 1 commit
267
+ - sequencer: mark messages for translation
268
+
269
+ Message fix.
270
+
271
+ Will merge to 'next'.
272
+
273
+
274
+* dl/wrapper-fix-indentation (2020-03-28) 1 commit
275
+ - wrapper: indent with tabs
276
+
277
+ Coding style fix.
278
+
279
+ Will merge to 'next'.
280
+
281
+
282
+* en/pull-do-not-rebase-after-fast-forwarding (2020-03-27) 1 commit
283
+ - pull: avoid running both merge and rebase
284
+
285
+ "git pull --rebase" tried to run a rebase even after noticing that
286
+ the pull results in a fast-forward and no rebase is needed nor
287
+ sensible, for the past few years due to a mistake nobody noticed.
288
+
289
+ Will merge to 'next'.
290
+
291
+
292
+* jc/allow-strlen-substitution-in-shell-scripts (2020-03-29) 1 commit
293
+ - CodingGuidelines: allow ${#posix} == strlen($posix)
294
+
295
+ Coding guideline update.
296
+
297
+
298
+* jm/gitweb-fastcgi-utf8 (2020-03-29) 1 commit
299
+ - gitweb: fix UTF-8 encoding when using CGI::Fast
300
+
301
+ Gitweb update.
302
+
303
+
304
+* js/walk-doc-optim (2020-03-28) 1 commit
305
+ - MyFirstObjectWalk: remove unnecessary conditional statement
306
+
307
+ Code cleanup.
308
+
309
+ Will merge to 'next'.
310
+
311
+
312
+* jx/atomic-push (2020-03-29) 4 commits
313
+ - transport-helper: new method reject_atomic_push()
314
+ - transport-helper: mark failure for atomic push
315
+ - send-pack: mark failure of atomic push properly
316
+ - t5543: never report what we do not push
317
+
318
+ "git push --atomic" used to show failures for refs that weren't
319
+ even pushed, which has been corrected.
320
+
321
+ Will merge to 'next'.
322
+
323
+
324
+* ma/doc-discard-docbook-xsl-1.73 (2020-03-29) 6 commits
325
+ - INSTALL: drop support for docbook-xsl before 1.74
326
+ - manpage-normal.xsl: fold in manpage-base.xsl
327
+ - manpage-bold-literal.xsl: stop using git.docbook.backslash
328
+ - Doc: drop support for docbook-xsl before 1.73.0
329
+ - Doc: drop support for docbook-xsl before 1.72.0
330
+ - Doc: drop support for docbook-xsl before 1.71.1
331
+
332
+ Raise the minimum required version of docbook-xsl package to 1.74,
333
+ as 1.74.0 was from late 2008, which is more than 10 years old, and
334
+ drop compatibility cruft from our documentation suite.
335
+
336
+ Will merge to 'next'.
337
+
338
+
339
+* pb/rebase-doc-typofix (2020-03-28) 1 commit
340
+ - git-rebase.txt: fix typo
341
+
342
+ Typofix.
343
+
344
+ Will merge to 'next'.
345
+
346
+
347
+* rs/pull-options-sync-code-and-doc (2020-03-28) 2 commits
348
+ - pull: pass documented fetch options on
349
+ - pull: remove --update-head-ok from documentation
350
+
351
+ "git pull" shares many options with underlying "git fetch", but
352
+ some of them were not documented and some of those that would make
353
+ sense to pass down were not passed down.
354
+
355
+ Will merge to 'next'.
356
357
--------------------------------------------------
143
-[New Topics]
358
+[Stalled]
359
+
360
+* gs/commit-graph-path-filter (2020-02-12) 12 commits
361
+ - (bytesex breakage band-aid)
362
+ - commit-graph: add GIT_TEST_COMMIT_GRAPH_CHANGED_PATHS test flag
363
+ - revision.c: use Bloom filters to speed up path based revision walks
364
+ - commit-graph: add --changed-paths option to write subcommand
365
+ - commit-graph: reuse existing Bloom filters during write.
366
+ - commit-graph: write Bloom filters to commit graph file
367
+ - commit-graph: examine commits by generation number
368
+ - commit-graph: examine changed-path objects in pack order
369
+ - commit-graph: compute Bloom filters for changed paths
370
+ - diff: halt tree-diff early after max_changes
371
+ - bloom: core Bloom filter implementation for changed paths
372
+ - commit-graph: use MAX_NUM_CHUNKS
373
+
374
+ Introduce an extension to the commit-graph to make it efficient to
375
+ check for the paths that were modified at each commit using Bloom
376
+ filters.
377
+
378
+ Expecting a reroll.
379
+ cf. <fdcbd793-57c2-f5ea-ccb9-cf34e911b669@gmail.com>
380
+ Breakage due to byte-order dependency reported.
381
+
382
+
383
+* mk/use-size-t-in-zlib (2018-10-15) 1 commit
384
+ - zlib.c: use size_t for size
385
+
386
+ The wrapper to call into zlib followed our long tradition to use
387
+ "unsigned long" for sizes of regions in memory, which have been
388
+ updated to use "size_t".
389
145
-* dd/test-with-busybox (2020-03-25) 8 commits
390
+--------------------------------------------------
391
+[Cooking]
392
+
393
+* en/fill-directory-exponential (2020-03-26) 7 commits
394
+ - dir: replace exponential algorithm with a linear one
395
+ - dir: refactor treat_directory to clarify control flow
396
+ - dir: fix confusion based on variable tense
397
+ - dir: fix broken comment
398
+ - dir: consolidate treat_path() and treat_one_path()
399
+ - dir: fix simple typo in comment
400
+ - t7063: more thorough status checking
401
+
402
+ The directory traversal code had redundant recursive calls which
403
+ made its performance characteristics exponential with respect to
404
+ the depth of the tree, which was corrected.
405
+
406
+ Expecting a reroll.
407
+ cf. <pull.700.v4.git.git.1585258061.gitgitgadget@gmail.com> (v4)
408
+ cf. <CABPp-BFXWx_MPL1Z0-X5hHf0N1T_13TEbAhVTKh6-atO1+g90Q@mail.gmail.com>
409
+
410
+
411
+* dd/test-with-busybox (2020-03-26) 8 commits
412
- t5703: feed raw data into test-tool unpack-sideband
147
- - t4124: fix test for non-compliant diff(1)
413
+ - t4124: tweak test so that non-compliant diff(1) can also be used
414
- t7063: drop non-POSIX argument "-ls" from find(1)
415
- t5616: use rev-parse instead to get HEAD's object_id
416
- t5003: skip conversion test if unzip -a is unavailable
@@ -167,21 +433,25 @@ of the repositories listed at
433
subcommands have been moved to libgit.a so that they could be used
434
by others.
435
436
+ Will merge to 'next'.
437
+
438
171
-* dl/test-must-fail-fixes-3 (2020-03-24) 8 commits
439
+* dl/test-must-fail-fixes-3 (2020-03-27) 8 commits
440
- t5801: teach compare_refs() to accept !
441
- t5612: stop losing return codes of git commands
442
- t5612: don't use `test_must_fail test_cmp`
443
- t5607: reorder `nongit test_must_fail`
176
- - t5550: remove use of `test_might_fail grep`
444
+ - t5550: simplify no matching line check
445
- t5512: stop losing return codes of git commands
178
- - t5512: generate references with generate_references()
446
+ - t5512: stop losing git exit code in here-docs
447
- t5512: don't use `test_must_fail test_cmp`
448
449
Test clean-up continues.
450
451
+ Will merge to 'next'.
452
+
453
184
-* en/sparse-checkout (2020-03-23) 18 commits
454
+* en/sparse-checkout (2020-03-27) 18 commits
455
- sparse-checkout: provide a new reapply subcommand
456
- unpack-trees: failure to set SKIP_WORKTREE bits always just a warning
457
- unpack-trees: provide warnings on sparse updates for unmerged paths too
@@ -203,6 +473,8 @@ of the repositories listed at
473
474
"sparse-checkout" UI improvements.
475
476
+ Will merge to 'next'.
477
+
478
479
* js/import-tars-do-not-make-phony-files-from-pax-headers (2020-03-24) 1 commit
480
- import-tars: ignore the global PAX header
@@ -224,10 +496,10 @@ of the repositories listed at
496
Will merge to 'next'.
497
498
227
-* js/tests-gpg-integration-on-windows (2020-03-25) 5 commits
499
+* js/tests-gpg-integration-on-windows (2020-03-26) 5 commits
500
- tests: increase the verbosity of the GPG-related prereqs
229
- - tests: do not let lazy prereqs inside `test_expect_*` turn off tracing
501
- tests: turn GPG, GPGSM and RFC1991 into lazy prereqs
502
+ - tests: do not let lazy prereqs inside `test_expect_*` turn off tracing
503
- t/lib-gpg.sh: stop pretending to be a stand-alone script
504
- tests(gpg): allow the gpg-agent to start on Windows
505
@@ -235,68 +507,6 @@ of the repositories listed at
507
508
Will merge to 'next'.
509
238
---------------------------------------------------
239
-[Stalled]
240
-
241
-* gs/commit-graph-path-filter (2020-02-12) 12 commits
242
- - (bytesex breakage band-aid)
243
- - commit-graph: add GIT_TEST_COMMIT_GRAPH_CHANGED_PATHS test flag
244
- - revision.c: use Bloom filters to speed up path based revision walks
245
- - commit-graph: add --changed-paths option to write subcommand
246
- - commit-graph: reuse existing Bloom filters during write.
247
- - commit-graph: write Bloom filters to commit graph file
248
- - commit-graph: examine commits by generation number
249
- - commit-graph: examine changed-path objects in pack order
250
- - commit-graph: compute Bloom filters for changed paths
251
- - diff: halt tree-diff early after max_changes
252
- - bloom: core Bloom filter implementation for changed paths
253
- - commit-graph: use MAX_NUM_CHUNKS
254
-
255
- Introduce an extension to the commit-graph to make it efficient to
256
- check for the paths that were modified at each commit using Bloom
257
- filters.
258
-
259
- Will be rerolled with bytesex fixes squashed in.
260
- Breakage due to byte-order dependency reported.
261
-
262
-
263
-* en/fill-directory-exponential (2020-01-31) 6 commits
264
- - t7063: blindly accept diffs
265
- - dir: replace exponential algorithm with a linear one
266
- - dir: refactor treat_directory to clarify control flow
267
- - dir: fix confusion based on variable tense
268
- - dir: fix broken comment
269
- - dir: consolidate treat_path() and treat_one_path()
270
-
271
- The directory traversal code had redundant recursive calls which
272
- made its performance characteristics exponential with respect to
273
- the depth of the tree, which was corrected.
274
-
275
- Expecting an update.
276
- cf. <CABPp-BEnt4C_7XyxQKxk4aga=JjM9fXCE-7SFp7azO_v5-pQYw@mail.gmail.com>
277
- cf. <pull.700.v2.git.git.1580495486.gitgitgadget@gmail.com>
278
- cf. <CABPp-BF0SO92wxY-dM14Vxv6BGA5=T=CicdwbiaZ7c9w6-j2jQ@mail.gmail.com>
279
-
280
-
281
-* vn/reset-deleted-ita (2019-07-26) 1 commit
282
- - reset: unstage empty deleted ita files
283
-
284
- "git reset HEAD [<pathspec>]" did not reset an empty file that was
285
- added with the intent-to-add bit.
286
-
287
- Will discard.
288
- Getting tired of expecting a reroll.
289
-
290
-
291
-* mk/use-size-t-in-zlib (2018-10-15) 1 commit
292
- - zlib.c: use size_t for size
293
-
294
- The wrapper to call into zlib followed our long tradition to use
295
- "unsigned long" for sizes of regions in memory, which have been
296
- updated to use "size_t".
297
-
298
---------------------------------------------------
299
-[Cooking]
510
511
* dl/merge-autostash (2020-03-24) 19 commits
512
- pull: pass --autostash to merge
@@ -321,16 +531,7 @@ of the repositories listed at
531
532
"git merge" learns the "--autostash" option.
533
324
-
325
-* ds/default-pack-use-sparse-to-true (2020-03-20) 2 commits
326
- (merged to 'next' on 2020-03-21 at 25cc87784d)
327
- + pack-objects: flip the use of GIT_TEST_PACK_SPARSE
328
- + config: set pack.useSparse=true by default
329
-
330
- The 'pack.useSparse' configuration variable now defaults to 'true',
331
- enabling an optimization that has been experimental since Git 2.21.
332
-
333
- Will cook in 'next'.
534
+ Will merge to 'next'.
535
536
537
* js/trace2-env-vars (2020-03-23) 1 commit
@@ -338,6 +539,8 @@ of the repositories listed at
539
540
Trace2 enhancement to allow logging of the environment variables.
541
542
+ Will merge to 'next'.
543
+
544
545
* ar/test-style-fixes (2020-03-22) 2 commits
546
- t: fix whitespace around &&
@@ -364,13 +567,16 @@ of the repositories listed at
567
Will merge to 'next'.
568
569
367
-* jt/connectivity-check-optim-in-partial-clone (2020-03-22) 1 commit
570
+* jt/connectivity-check-optim-in-partial-clone (2020-03-29) 1 commit
571
- connected: always use partial clone optimization
572
573
Simplify the commit ancestry connectedness check in a partial clone
574
repository in which "promised" objects are assumed to be obtainable
575
lazily on-demand from promisor remote repositories.
576
577
+ Looking good.
578
+ cf. <20200326231416.GC12694@google.com>
579
+
580
581
* mt/test-lib-bundled-short-options (2020-03-25) 1 commit
582
- test-lib: allow short options to be bundled
@@ -397,7 +603,8 @@ of the repositories listed at
603
cf. <pull.698.v6.git.git.1581691486.gitgitgadget@gmail.com>
604
605
400
-* jt/rebase-allow-duplicate (2020-03-18) 1 commit
606
+* jt/rebase-allow-duplicate (2020-03-29) 2 commits
607
+ - t3402: use POSIX compliant regex(7)
608
- rebase --merge: optionally skip upstreamed commits
609
610
Allow "git rebase" to reapply all local commits, even if the may be
@@ -407,29 +614,14 @@ of the repositories listed at
614
cf. <20200318192821.43808-1-jonathantanmy@google.com>
615
616
410
-* bc/filter-process (2020-03-16) 8 commits
411
- (merged to 'next' on 2020-03-17 at 2cd9dbf794)
412
- + t0021: test filter metadata for additional cases
413
- + builtin/reset: compute checkout metadata for reset
414
- + builtin/rebase: compute checkout metadata for rebases
415
- + builtin/clone: compute checkout metadata for clones
416
- + builtin/checkout: compute checkout metadata for checkouts
417
- + convert: provide additional metadata to filters
418
- + convert: permit passing additional metadata to filter processes
419
- + builtin/checkout: pass branch info down to checkout_worktree
420
-
421
- Provide more information (e.g. the object of the tree-ish in which
422
- the blob being converted appears, in addition to its path, which
423
- has already been given) to smudge/clean conversion filters.
424
-
425
- Will merge to 'master'.
426
-
427
-
617
* bc/faq (2020-03-24) 1 commit
618
- docs: add a FAQ
619
620
Doc update.
621
622
+ Expecting a reroll.
623
+ cf. <20200325110328.GK6499@camp.crustytoothpaste.net>
624
+
625
626
* jc/log-no-mailmap (2020-03-16) 3 commits
627
- log: give --[no-]use-mailmap a more sensible synonym --[no-]mailmap
@@ -438,15 +630,7 @@ of the repositories listed at
630
631
"git log" learns "--[no-]mailmap" as a synonym to "--[no-]use-mailmap"
632
441
-
442
-* hi/gpg-prefer-check-signature (2020-03-15) 2 commits
443
- (merged to 'next' on 2020-03-17 at 2def2d9a7e)
444
- + gpg-interface: prefer check_signature() for GPG verification
445
- + t: increase test coverage of signature verification output
446
-
447
- The code to interface with GnuPG has been refactored.
448
-
449
- Will merge to 'master'.
633
+ Will merge to 'next'.
634
635
636
* tb/commit-graph-split-merge (2020-03-24) 3 commits
@@ -461,100 +645,6 @@ of the repositories listed at
645
Will merge to 'next'.
646
647
464
-* ah/force-pull-rebase-configuration (2020-03-10) 1 commit
465
- (merged to 'next' on 2020-03-11 at c79dbbaf9c)
466
- + pull: warn if the user didn't say whether to rebase or to merge
467
-
468
- "git pull" learned to warn when no pull.rebase configuration
469
- exists, and neither --[no-]rebase nor --ff-only is given (which
470
- would result a merge).
471
-
472
- Will cook in 'next'.
473
-
474
-
475
-* tg/retire-scripted-stash (2020-03-05) 2 commits
476
- (merged to 'next' on 2020-03-05 at 8e82eb9dec)
477
- + stash: remove the stash.useBuiltin setting
478
- + stash: get git_stash_config at the top level
479
-
480
- "git stash" has kept an escape hatch to use the scripted version
481
- for a few releases, which got stale. It has been removed.
482
-
483
- Will cook in 'next'.
484
- cf. <20200306172913.GF1571684@cat>
485
-
486
-
487
-* at/rebase-fork-point-regression-fix (2020-02-11) 1 commit
488
- (merged to 'next' on 2020-03-02 at a1a84d37a7)
489
- + rebase: --fork-point regression fix
490
-
491
- The "--fork-point" mode of "git rebase" regressed when the command
492
- was rewritten in C back in 2.20 era, which has been corrected.
493
-
494
- Will cook in 'next'.
495
-
496
-
497
-* jc/describe-misnamed-annotated-tag (2020-02-20) 1 commit
498
- (merged to 'next' on 2020-03-02 at b4e2ca6a46)
499
- + describe: force long format for a name based on a mislocated tag
500
-
501
- When "git describe C" finds an annotated tag with tagname A to be
502
- the best name to explain commit C, and the tag is stored in a
503
- "wrong" place in the refs/tags hierarchy, e.g. refs/tags/B, the
504
- command gave a warning message but used A (not B) to describe C.
505
- If C is exactly at the tag, the describe output would be "A", but
506
- "git rev-parse A^0" would not be equal as "git rev-parse C^0". The
507
- behavior of the command has been changed to use the "long" form
508
- i.e. A-0-gOBJECTNAME, which is correctly interpreted by rev-parse.
509
-
510
- Will cook in 'next'.
511
-
512
-
513
-* pb/recurse-submodules-fix (2020-02-19) 6 commits
514
- (merged to 'next' on 2020-03-17 at b46922ddd1)
515
- + t/lib-submodule-update: add test removing nested submodules
516
- + unpack-trees: check for missing submodule directory in merged_entry
517
- + unpack-trees: remove outdated description for verify_clean_submodule
518
- + t/lib-submodule-update: move a test to the right section
519
- + t/lib-submodule-update: remove outdated test description
520
- + t7112: remove mention of KNOWN_FAILURE_SUBMODULE_RECURSIVE_NESTED
521
-
522
- Fix "git checkout --recurse-submodules" of a nested submodule
523
- hierarchy.
524
-
525
- Will merge to 'master'.
526
-
527
-
528
-* bc/sha-256-part-1-of-4 (2020-02-28) 22 commits
529
- (merged to 'next' on 2020-03-17 at 436c4e64a7)
530
- + fast-import: add options for rewriting submodules
531
- + fast-import: add a generic function to iterate over marks
532
- + fast-import: make find_marks work on any mark set
533
- + fast-import: add helper function for inserting mark object entries
534
- + fast-import: permit reading multiple marks files
535
- + commit: use expected signature header for SHA-256
536
- + worktree: allow repository version 1
537
- + init-db: move writing repo version into a function
538
- + builtin/init-db: add environment variable for new repo hash
539
- + builtin/init-db: allow specifying hash algorithm on command line
540
- + setup: allow check_repository_format to read repository format
541
- + t/helper: make repository tests hash independent
542
- + t/helper: initialize repository if necessary
543
- + t/helper/test-dump-split-index: initialize git repository
544
- + t6300: make hash algorithm independent
545
- + t6300: abstract away SHA-1-specific constants
546
- + t: use hash-specific lookup tables to define test constants
547
- + repository: require a build flag to use SHA-256
548
- + hex: add functions to parse hex object IDs in any algorithm
549
- + hex: introduce parsing variants taking hash algorithms
550
- + hash: implement and use a context cloning function
551
- + builtin/pack-objects: make hash agnostic
552
-
553
- SHA-256 transition continues.
554
-
555
- Will merge to 'master'.
556
-
557
-
648
* hn/reftable (2020-02-26) 6 commits
649
. Reftable support for git-core
650
. Add reftable library
@@ -572,7 +662,7 @@ of the repositories listed at
662
SHA-256 topic.
663
664
575
-* es/bugreport (2020-03-23) 5 commits
665
+* es/bugreport (2020-03-26) 5 commits
666
- bugreport: add compiler info
667
- bugreport: add uname info
668
- bugreport: gather git version and build info
@@ -581,9 +671,9 @@ of the repositories listed at
671
672
The "bugreport" tool.
673
584
- New iteration (v11) exists, which needs to be picked up.
585
- As the scope of the topic got trimmed, hopefully these early parts
586
- can be polished quickly enough to be merged down.
674
+ Ready for 'next' to be improved/enhanced on top? As the scope of
675
+ the topic got trimmed, hopefully these early parts can be polished
676
+ quickly enough to be merged down.
677
678
--------------------------------------------------
679
[Discarded]
@@ -595,3 +685,10 @@ of the repositories listed at
685
unexpected behaviour that need to be ironed out before it becomes
686
the default. Let's switch the default back to the "apply" backend
687
for now.
688
+
689
+
690
+* vn/reset-deleted-ita (2019-07-26) 1 commit
691
+ . reset: unstage empty deleted ita files
692
+
693
+ "git reset HEAD [<pathspec>]" did not reset an empty file that was
694
+ added with the intent-to-add bit.