What's cooking (2020/06 #04)
Junio C Hamano committed
Jun 22, 2020 at 17:20 UTC
a4a549d7cbd23a401f0c6db9602c6fbacb6ad8c2
1 file changed
+341
-276
whats-cooking.txt
+341
-276
@@ -1,117 +1,281 @@
1
To: git@vger.kernel.org
2
Bcc: lwn@lwn.net
3
-Subject: What's cooking in git.git (Jun 2020, #03; Wed, 17)
4
-X-master-at: 101b3204f37606972b40fc17dec84560c22f69f6
5
-X-next-at: f2d1464379e7567b01b55fbf8432e75d72aa537f
3
+Subject: What's cooking in git.git (Jun 2020, #04; Mon, 22)
4
+X-master-at: c9c318d6bf26bcecdca5b6f31683b9d5887a83ee
5
+X-next-at: 4b34aa94c75220f6f4cd334b08a3fb053128a32d
6
7
-What's cooking in git.git (Jun 2020, #03; Wed, 17)
7
+What's cooking in git.git (Jun 2020, #04; Mon, 22)
8
--------------------------------------------------
9
10
-Here are the topics that have been cooking. Commits prefixed with
11
-'-' are only in 'pu' (proposed updates) while commits prefixed with
12
-'+' are in 'next'. The ones marked with '.' do not appear in any of
10
+Here are the topics that have been cooking. Commits prefixed with '-' are
11
+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
-You can find the changes described here in the integration branches
16
-of the repositories listed at
15
+A few topiocs have graduated to 'master'. As I expect that the next
16
+week would be slower than usual due to being summer (and folks in US
17
+having holidays), I am hoping that I can merge down a handful more
18
+topics from 'next', perhaps with shorter gestation period than
19
+usual, by the end of this week.
20
+
21
+The 'pu' (proposed updates) branch has been renamed to 'seen' to use
22
+a more meaningful name (after all, the purpose of it is not to
23
+contain all proposals, but to merely serve as a place to park what
24
+the maintainer happened to have seen) and more importantly, to make
25
+room for topics from those contributors whose two-letter name
26
+abbreviation needs to be 'pu' under 'pu/$topicName'.
27
+
28
+You can find the changes described here in the integration branches of the
29
+repositories listed at
30
31
http://git-blame.blogspot.com/p/git-public-repositories.html
32
33
--------------------------------------------------
34
[Graduated to "master"]
35
23
-* dl/python-2.7-is-the-floor-version (2020-06-08) 1 commit
24
- (merged to 'next' on 2020-06-10 at ad3269ff40)
25
- + CodingGuidelines: specify Python 2.7 is the oldest version
36
+* cc/upload-pack-data-2 (2020-06-04) 13 commits
37
+ (merged to 'next' on 2020-06-12 at ee63b70dc2)
38
+ + upload-pack: move pack_objects_hook to upload_pack_data
39
+ + upload-pack: move allow_sideband_all to upload_pack_data
40
+ + upload-pack: move allow_ref_in_want to upload_pack_data
41
+ + upload-pack: move allow_filter to upload_pack_data
42
+ + upload-pack: move keepalive to upload_pack_data
43
+ + upload-pack: pass upload_pack_data to upload_pack_config()
44
+ + upload-pack: change multi_ack to an enum
45
+ + upload-pack: move multi_ack to upload_pack_data
46
+ + upload-pack: move filter_capability_requested to upload_pack_data
47
+ + upload-pack: move use_sideband to upload_pack_data
48
+ + upload-pack: move static vars to upload_pack_data
49
+ + upload-pack: annotate upload_pack_data fields
50
+ + upload-pack: actually use some upload_pack_data bitfields
51
+ (this branch is used by cc/upload-pack-data-3 and jt/cdn-offload.)
52
+
53
+ Further code clean-up.
54
+
55
+
56
+* en/sparse-with-submodule-doc (2020-06-12) 1 commit
57
+ (merged to 'next' on 2020-06-17 at 3521be3834)
58
+ + git-sparse-checkout: clarify interactions with submodules
59
+
60
+ The effect of sparse checkout settings on submodules is documented.
61
+
62
+
63
+* es/worktree-duplicate-paths (2020-06-10) 7 commits
64
+ (merged to 'next' on 2020-06-12 at 5f7c822a9d)
65
+ + worktree: make "move" refuse to move atop missing registered worktree
66
+ + worktree: generalize candidate worktree path validation
67
+ + worktree: prune linked worktree referencing main worktree path
68
+ + worktree: prune duplicate entries referencing same worktree path
69
+ + worktree: make high-level pruning re-usable
70
+ + worktree: give "should be pruned?" function more meaningful name
71
+ + worktree: factor out repeated string literal
72
+ (this branch is used by es/get-worktrees-unsort.)
73
+
74
+ The same worktree directory must be registered only once, but
75
+ "git worktree move" allowed this invariant to be violated, which
76
+ has been corrected.
77
27
- Document that we do not support Python 2.6 or older.
78
79
+* jt/redact-all-cookies (2020-06-05) 1 commit
80
+ (merged to 'next' on 2020-06-12 at 912c337d63)
81
+ + http: redact all cookies, teach GIT_TRACE_REDACT=0
82
30
-* dl/t-readme-spell-git-correctly (2020-06-08) 1 commit
31
- (merged to 'next' on 2020-06-10 at 60144a59ed)
32
- + t/README: avoid poor-man's small caps GIT
83
+ The interface to redact sensitive information in the trace output
84
+ has been simplified.
85
34
- Doc updates.
86
+--------------------------------------------------
87
+[New Topics]
88
89
+* en/sparse-status (2020-06-22) 3 commits
90
+ - git-prompt: include sparsity state as well
91
+ - git-prompt: document how in-progress operations affect the prompt
92
+ - wt-status: show sparse checkout status as well
93
37
-* en/do-match-pathspec-fix (2020-06-05) 1 commit
38
- (merged to 'next' on 2020-06-10 at 2e63c8c0fc)
39
- + dir: fix treatment of negated pathspecs
94
+ "git status" learned to report the status of sparse checkout.
95
41
- Use of negative pathspec, while collecting paths including
42
- untracked ones in the working tree, was broken.
96
+ Will merge to 'next'.
97
98
45
-* en/sparse-checkout (2020-06-05) 1 commit
46
- (merged to 'next' on 2020-06-10 at 697a10b031)
47
- + sparse-checkout: avoid staging deletions of all files
99
+* dl/diff-usage-comment-update (2020-06-18) 1 commit
100
+ (merged to 'next' on 2020-06-19 at b53a12bb66)
101
+ + builtin/diff: update usage comment
102
49
- The behaviour of "sparse-checkout" in the state "git clone
50
- --no-checkout" left was changed accidentally in 2.27, which has
51
- been corrected.
103
+ An in-code comment in "git diff" has been updated.
104
105
+ Will merge to 'master'.
106
+
107
+
108
+* ps/ref-transaction-hook (2020-06-19) 1 commit
109
+ (merged to 'next' on 2020-06-22 at 3a23dcdbdc)
110
+ + refs: implement reference transaction hook
111
+
112
+ A new hook.
113
+
114
+ Will merge to 'master'.
115
+
116
+
117
+* sk/diff-files-show-i-t-a-as-new (2020-06-22) 1 commit
118
+ (merged to 'next' on 2020-06-22 at 825a823416)
119
+ + diff-files: treat "i-t-a" files as "not-in-index"
120
+
121
+ "git diff-files" has been taught to say paths that are marked as
122
+ intent-to-add are new files, not modified from an empty blob.
123
+
124
+ Will merge to 'master'.
125
+
126
+
127
+* jk/fast-export-anonym (2020-06-22) 4 commits
128
+ (merged to 'next' on 2020-06-22 at b517b2f707)
129
+ + fast-export: allow dumping the path mapping
130
+ + fast-export: refactor path printing to not rely on stdout
131
+ + fast-export: anonymize "master" refname
132
+ + fast-export: allow dumping the refname mapping
133
+
134
+ The way refnames are anonymized has been updated and a way to help
135
+ debugging using the anonymized output hsa been added.
136
+
137
+ Will merge to 'master'.
138
54
-* es/advertise-contribution-doc (2020-06-08) 1 commit
55
- (merged to 'next' on 2020-06-10 at c72c7da667)
56
- + docs: mention MyFirstContribution in more places
139
58
- Doc updates.
140
+* pb/t4014-unslave (2020-06-19) 1 commit
141
+ (merged to 'next' on 2020-06-19 at 4d328a12d9)
142
+ + t4014: do not use "slave branch" nomenclature
143
144
+ A branch name used in a test has been clarified to match what is
145
+ going on.
146
61
-* hn/refs-cleanup (2020-06-09) 6 commits
62
- (merged to 'next' on 2020-06-09 at abb59f83a2)
63
- + reftable: define version 2 of the spec to accomodate SHA256
64
- + reftable: clarify how empty tables should be written
65
- + reftable: file format documentation
66
- + refs: improve documentation for ref iterator
67
- + t: use update-ref and show-ref to reading/writing refs
68
- + refs.h: clarify reflog iteration order
69
- (this branch is used by hn/reftable.)
147
+ Will merge to 'master'.
148
71
- Preliminary clean-ups around refs API, plus file format
72
- specification documentation for the reftable backend.
149
150
+* rs/commit-reach-leakfix (2020-06-19) 1 commit
151
+ (merged to 'next' on 2020-06-22 at fcc1e385f0)
152
+ + commit-reach: plug minor memory leak after using is_descendant_of()
153
75
-* jk/diff-memuse-optim-with-stat-unmatch (2020-06-02) 1 commit
76
- (merged to 'next' on 2020-06-10 at 4cea3bfaac)
77
- + diff: discard blob data from stat-unmatched pairs
154
+ Leakfix.
155
79
- Reduce memory usage during "diff --quiet" in a worktree with too
80
- many stat-unmatched paths.
156
+ Will merge to 'master'.
157
158
83
-* js/fuzz-commit-graph-leakfix (2020-06-08) 1 commit
84
- (merged to 'next' on 2020-06-10 at 92944ed7f6)
85
- + fuzz-commit-graph: properly free graph struct
159
+* rs/pull-leakfix (2020-06-19) 1 commit
160
+ (merged to 'next' on 2020-06-22 at d80a6b9ca3)
161
+ + pull: plug minor memory leak after using is_descendant_of()
162
163
Leakfix.
164
165
+ Will merge to 'master'.
166
90
-* js/msvc-build-fix (2020-06-04) 1 commit
91
- (merged to 'next' on 2020-06-10 at 167a629359)
92
- + msvc: fix "REG_STARTEND" issue
167
94
- Workaround breakage in MSVC build, where "curl-config --cflags"
95
- gives settings appropriate for GCC build.
168
+* rs/retire-strbuf-write-fd (2020-06-19) 2 commits
169
+ (merged to 'next' on 2020-06-22 at c175b54f2a)
170
+ + strbuf: remove unreferenced strbuf_write_fd method.
171
+ + bugreport.c: replace strbuf_write_fd with write_in_full
172
173
+ A misdesigned strbuf_write_fd() function has been retired.
174
98
-* js/reflog-anonymize-for-clone-and-fetch (2020-06-04) 1 commit
99
- (merged to 'next' on 2020-06-10 at 2901ff943e)
100
- + clone/fetch: anonymize URLs in the reflog
175
+ Will merge to 'master'.
176
+
177
+
178
+* bc/sha-256-cvs-svn-updates (2020-06-22) 14 commits
179
+ - git-cvsexportcommit: port to SHA-256
180
+ - git-cvsimport: port to SHA-256
181
+ - git-cvsserver: port to SHA-256
182
+ - git-svn: set the OID length based on hash algorithm
183
+ - perl: make SVN code hash independent
184
+ - perl: make Git::IndexInfo work with SHA-256
185
+ - perl: create and switch variables for hash constants
186
+ - t/lib-git-svn: make hash size independent
187
+ - t9101: make hash independent
188
+ - t9104: make hash size independent
189
+ - t9100: make test work with SHA-256
190
+ - t9108: make test hash independent
191
+ - t9168: make test hash independent
192
+ - t9109: make test hash independent
193
+ (this branch uses bc/sha-256-part-2.)
194
+
195
+ CVS/SVN interface have been prepared for SHA-256 transition
196
+
197
+ Will merge to 'next'.
198
102
- The reflog entries for "git clone" and "git fetch" did not
103
- anonymize the URL they operated on.
199
200
+* ds/commit-graph-bloom-updates (2020-06-17) 8 commits
201
+ - commit-graph: persist existence of changed-paths
202
+ - commit-graph: change test to die on parse, not load
203
+ - bloom: enforce a minimum size of 8 bytes
204
+ - commit-graph: check all leading directories in changed path Bloom filters
205
+ - commit-graph: check chunk sizes after writing
206
+ - commit-graph: simplify chunk writes into loop
207
+ - commit-graph: unify the signatures of all write_graph_chunk_*() functions
208
+ - commit-graph: place bloom_settings in context
209
+ (this branch uses sg/commit-graph-cleanups.)
210
106
-* tb/t5318-cleanup (2020-06-04) 2 commits
107
- (merged to 'next' on 2020-06-10 at 5a72620cbe)
108
- + t5318: test that '--stdin-commits' respects '--[no-]progress'
109
- + t5318: use 'test_must_be_empty'
211
+ Updates to the changed-paths bloom filter.
212
111
- Code cleanup.
213
+ Being discussed.
214
+ cf. <20200619171922.GA57283@syl.local>
215
+
216
+
217
+* es/get-worktrees-unsort (2020-06-22) 2 commits
218
+ - worktree: drop get_worktrees() unused 'flags' argument
219
+ - worktree: drop get_worktrees() special-purpose sorting option
220
+
221
+ API cleanup for get_worktrees()
222
+
223
+ Will merge to 'next'.
224
+
225
+
226
+* jl/complete-git-prune (2020-06-22) 1 commit
227
+ - bash-completion: add git-prune into bash completion
228
+
229
+ Add "git prune" to the completion (in contrib/), which could be
230
+ typed by end-users from the command line.
231
+
232
+ Will merge to 'next'.
233
234
--------------------------------------------------
114
-[New Topics]
235
+[Stalled]
236
+
237
+* dr/push-remoteref-fix (2020-04-23) 1 commit
238
+ - remote.c: fix handling of %(push:remoteref)
239
+
240
+ The "%(push:remoteref)" placeholder in the "--format=" argument of
241
+ "git format-patch" (and friends) only showed what got explicitly
242
+ configured, not what ref at the receiving end would be updated when
243
+ "git push" was used, as it ignored the default behaviour (e.g. update
244
+ the same ref as the source).
245
+
246
+ Expecting a reroll.
247
+ cf. <20200416152145.wp2zeibxmuyas6y6@feanor>
248
+
249
+
250
+* mr/bisect-in-c-2 (2020-04-23) 12 commits
251
+ - bisect--helper: retire `--bisect-autostart` subcommand
252
+ - bisect--helper: retire `--write-terms` subcommand
253
+ - bisect--helper: retire `--check-expected-revs` subcommand
254
+ - bisect--helper: reimplement `bisect_state` & `bisect_head` shell functions in C
255
+ - bisect--helper: retire `--next-all` subcommand
256
+ - bisect--helper: retire `--bisect-clean-state` subcommand
257
+ - bisect--helper: finish porting `bisect_start()` to C
258
+ - bisect--helper: reimplement `bisect_next` and `bisect_auto_next` shell functions in C
259
+ - bisect--helper: reimplement `bisect_autostart` shell function in C
260
+ - bisect--helper: introduce new `write_in_file()` function
261
+ - bisect--helper: use '-res' in 'cmd_bisect__helper' return
262
+ - bisect--helper: fix `cmd_*()` function switch default return
263
+
264
+ Rewrite of the remainder of "git bisect" script in C continues.
265
+
266
+ Expecting a response to reviews.
267
+ cf. <nycvar.QRO.7.76.6.2005230007260.56@tvgsbejvaqbjf.bet>
268
+
269
+
270
+* mk/use-size-t-in-zlib (2018-10-15) 1 commit
271
+ - zlib.c: use size_t for size
272
+
273
+ The wrapper to call into zlib followed our long tradition to use
274
+ "unsigned long" for sizes of regions in memory, which have been
275
+ updated to use "size_t".
276
+
277
+--------------------------------------------------
278
+[Cooking]
279
280
* cc/upload-pack-data-3 (2020-06-11) 14 commits
281
(merged to 'next' on 2020-06-17 at 393eff577a)
@@ -129,7 +293,6 @@ of the repositories listed at
293
+ upload-pack: pass upload_pack_data to deepen_by_rev_list()
294
+ upload-pack: pass upload_pack_data to deepen()
295
+ upload-pack: pass upload_pack_data to send_shallow_list()
132
- (this branch uses cc/upload-pack-data-2)
296
297
Code clean-up in the codepath that serves "git fetch" continues.
298
@@ -149,9 +312,8 @@ of the repositories listed at
312
Will merge to 'master'.
313
314
152
-* js/default-branch-name (2020-06-10) 9 commits
315
+* js/default-branch-name (2020-06-19) 8 commits
316
- Document how the default branch name can be overridden
154
- - fast-export: respect the possibly-overridden default branch name
317
- fmt-merge-msg: learn about the possibly-configured default branch name
318
- clone: learn about the possibly-configured default branch name
319
- submodule: use the (possibly overridden) default branch name
@@ -169,23 +331,23 @@ of the repositories listed at
331
332
333
* jt/cdn-offload (2020-06-17) 10 commits
172
- - upload-pack: fix a sparse '0 as NULL pointer' warning
173
- - upload-pack: send part of packfile response as uri
174
- - fetch-pack: support more than one pack lockfile
175
- - upload-pack: refactor reading of pack-objects out
176
- - Documentation: add Packfile URIs design doc
177
- - Documentation: order protocol v2 sections
178
- - http-fetch: support fetching packfiles by URL
179
- - http-fetch: refactor into function
180
- - http: refactor finish_http_pack_request()
181
- - http: use --stdin when indexing dumb HTTP pack
182
- (this branch uses cc/upload-pack-data-2)
334
+ (merged to 'next' on 2020-06-18 at e8ba1cc988)
335
+ + upload-pack: fix a sparse '0 as NULL pointer' warning
336
+ + upload-pack: send part of packfile response as uri
337
+ + fetch-pack: support more than one pack lockfile
338
+ + upload-pack: refactor reading of pack-objects out
339
+ + Documentation: add Packfile URIs design doc
340
+ + Documentation: order protocol v2 sections
341
+ + http-fetch: support fetching packfiles by URL
342
+ + http-fetch: refactor into function
343
+ + http: refactor finish_http_pack_request()
344
+ + http: use --stdin when indexing dumb HTTP pack
345
346
The "fetch/clone" protocol has been updated to allow the server to
347
instruct the clients to grab pre-packaged packfile(s) in addition
348
to the packed object data coming over the wire.
349
188
- Will merge to 'next'.
350
+ Will merge to 'master'.
351
352
353
* ss/cmake-build (2020-06-12) 8 commits
@@ -227,71 +389,28 @@ of the repositories listed at
389
390
391
* dl/branch-cleanup (2020-06-17) 3 commits
230
- - branch: don't mix --edit-description
231
- - t3200: test for specific errors
232
- - t3200: rename "expected" to "expect"
392
+ (merged to 'next' on 2020-06-18 at 76e151342c)
393
+ + branch: don't mix --edit-description
394
+ + t3200: test for specific errors
395
+ + t3200: rename "expected" to "expect"
396
397
Code clean-up around "git branch" with a minor bugfix.
398
236
- Will merge to 'next'.
399
+ Will merge to 'master'.
400
401
402
* ds/merge-base-is-ancestor-optim (2020-06-17) 2 commits
240
- - commit-reach: use fast logic in repo_in_merge_base
241
- - commit-reach: create repo_is_descendant_of()
403
+ (merged to 'next' on 2020-06-18 at 86b34a3688)
404
+ + commit-reach: use fast logic in repo_in_merge_base
405
+ + commit-reach: create repo_is_descendant_of()
406
407
"git merge-base --is-ancestor" is taught to take advantage of the
408
commit graph.
409
246
- Will merge to 'next'.
247
-
248
---------------------------------------------------
249
-[Stalled]
250
-
251
-* dr/push-remoteref-fix (2020-04-23) 1 commit
252
- - remote.c: fix handling of %(push:remoteref)
253
-
254
- The "%(push:remoteref)" placeholder in the "--format=" argument of
255
- "git format-patch" (and friends) only showed what got explicitly
256
- configured, not what ref at the receiving end would be updated when
257
- "git push" was used, as it ignored the default behaviour (e.g. update
258
- the same ref as the source).
259
-
260
- Expecting a reroll.
261
- cf. <20200416152145.wp2zeibxmuyas6y6@feanor>
262
-
263
-
264
-* mr/bisect-in-c-2 (2020-04-23) 12 commits
265
- - bisect--helper: retire `--bisect-autostart` subcommand
266
- - bisect--helper: retire `--write-terms` subcommand
267
- - bisect--helper: retire `--check-expected-revs` subcommand
268
- - bisect--helper: reimplement `bisect_state` & `bisect_head` shell functions in C
269
- - bisect--helper: retire `--next-all` subcommand
270
- - bisect--helper: retire `--bisect-clean-state` subcommand
271
- - bisect--helper: finish porting `bisect_start()` to C
272
- - bisect--helper: reimplement `bisect_next` and `bisect_auto_next` shell functions in C
273
- - bisect--helper: reimplement `bisect_autostart` shell function in C
274
- - bisect--helper: introduce new `write_in_file()` function
275
- - bisect--helper: use '-res' in 'cmd_bisect__helper' return
276
- - bisect--helper: fix `cmd_*()` function switch default return
277
-
278
- Rewrite of the remainder of "git bisect" script in C continues.
279
-
280
- Expecting a response to reviews.
281
- cf. <nycvar.QRO.7.76.6.2005230007260.56@tvgsbejvaqbjf.bet>
282
-
283
-
284
-* mk/use-size-t-in-zlib (2018-10-15) 1 commit
285
- - zlib.c: use size_t for size
286
-
287
- The wrapper to call into zlib followed our long tradition to use
288
- "unsigned long" for sizes of regions in memory, which have been
289
- updated to use "size_t".
410
+ Will merge to 'master'.
411
291
---------------------------------------------------
292
-[Cooking]
412
294
-* dl/test-must-fail-fixes-5 (2020-06-12) 4 commits
413
+* dl/test-must-fail-fixes-5 (2020-06-18) 4 commits
414
- lib-submodule-update: use callbacks in test_submodule_switch_common()
415
- lib-submodule-update: prepend "git" to $command
416
- lib-submodule-update: consolidate --recurse-submodules
@@ -299,45 +418,11 @@ of the repositories listed at
418
419
The effort to avoid using test_must_fail on non-git command continues.
420
302
- Ready???
303
-
304
-
305
-* jt/redact-all-cookies (2020-06-05) 1 commit
306
- (merged to 'next' on 2020-06-12 at 912c337d63)
307
- + http: redact all cookies, teach GIT_TRACE_REDACT=0
421
+ Getting very close.
422
+ cf. <20200618181533.GA633383@coredump.intra.peff.net>
423
309
- The interface to redact sensitive information in the trace output
310
- has been simplified.
424
312
- Will merge to 'master'.
313
-
314
-
315
-* es/worktree-duplicate-paths (2020-06-10) 7 commits
316
- (merged to 'next' on 2020-06-12 at 5f7c822a9d)
317
- + worktree: make "move" refuse to move atop missing registered worktree
318
- + worktree: generalize candidate worktree path validation
319
- + worktree: prune linked worktree referencing main worktree path
320
- + worktree: prune duplicate entries referencing same worktree path
321
- + worktree: make high-level pruning re-usable
322
- + worktree: give "should be pruned?" function more meaningful name
323
- + worktree: factor out repeated string literal
324
-
325
- The same worktree directory must be registered only once, but
326
- "git worktree move" allowed this invariant to be violated, which
327
- has been corrected.
328
-
329
- Will merge to 'master'.
330
-
331
-
332
-* sg/commit-graph-cleanups (2020-06-17) 18 commits
333
- - commit-graph: persist existence of changed-paths
334
- - commit-graph: change test to die on parse, not load
335
- - bloom: enforce a minimum size of 8 bytes
336
- - commit-graph: check all leading directories in changed path Bloom filters
337
- - commit-graph: check chunk sizes after writing
338
- - commit-graph: simplify chunk writes into loop
339
- - commit-graph: unify the signatures of all write_graph_chunk_*() functions
340
- - commit-graph: place bloom_settings in context
425
+* sg/commit-graph-cleanups (2020-06-08) 10 commits
426
- commit-graph: simplify write_commit_graph_file() #2
427
- commit-graph: simplify write_commit_graph_file() #1
428
- commit-graph: simplify parse_commit_graph() #2
@@ -348,37 +433,18 @@ of the repositories listed at
433
- commit-graph-format.txt: all multi-byte numbers are in network byte order
434
- commit-graph: fix parsing the Chunk Lookup table
435
- tree-walk.c: don't match submodule entries for 'submod/anything'
436
+ (this branch is used by ds/commit-graph-bloom-updates.)
437
352
- Code cleanup.
353
-
354
-
355
-* cc/upload-pack-data-2 (2020-06-04) 13 commits
356
- (merged to 'next' on 2020-06-12 at ee63b70dc2)
357
- + upload-pack: move pack_objects_hook to upload_pack_data
358
- + upload-pack: move allow_sideband_all to upload_pack_data
359
- + upload-pack: move allow_ref_in_want to upload_pack_data
360
- + upload-pack: move allow_filter to upload_pack_data
361
- + upload-pack: move keepalive to upload_pack_data
362
- + upload-pack: pass upload_pack_data to upload_pack_config()
363
- + upload-pack: change multi_ack to an enum
364
- + upload-pack: move multi_ack to upload_pack_data
365
- + upload-pack: move filter_capability_requested to upload_pack_data
366
- + upload-pack: move use_sideband to upload_pack_data
367
- + upload-pack: move static vars to upload_pack_data
368
- + upload-pack: annotate upload_pack_data fields
369
- + upload-pack: actually use some upload_pack_data bitfields
370
- (this branch is used by cc/upload-pack-data-3 and jt/cdn-offload.)
371
-
372
- Further code clean-up.
373
-
374
- Will merge to 'master'.
438
+ The changed-path Bloom filter is improved using ideas from an
439
+ independent implementation.
440
441
442
* xl/upgrade-repo-format (2020-06-05) 4 commits
378
- - check_repository_format_gently(): refuse extensions for old repositories
379
- - sparse-checkout: upgrade repository to version 1 when enabling extension
380
- - fetch: allow adding a filter after initial clone
381
- - repository: add a helper function to perform repository format upgrade
443
+ (merged to 'next' on 2020-06-19 at 02bf7a9d8c)
444
+ + check_repository_format_gently(): refuse extensions for old repositories
445
+ + sparse-checkout: upgrade repository to version 1 when enabling extension
446
+ + fetch: allow adding a filter after initial clone
447
+ + repository: add a helper function to perform repository format upgrade
448
449
Allow runtime upgrade of the repository format version, which needs
450
to be done carefully.
@@ -387,7 +453,7 @@ of the repositories listed at
453
last step of this series, but it is the right thing to do in the
454
longer term.
455
390
- Will merge to 'next'.
456
+ Will merge to 'master'.
457
458
459
* jk/complete-git-switch (2020-05-28) 16 commits
@@ -415,23 +481,15 @@ of the repositories listed at
481
Will merge to 'master'.
482
483
418
-* en/sparse-with-submodule-doc (2020-06-12) 1 commit
419
- (merged to 'next' on 2020-06-17 at 3521be3834)
420
- + git-sparse-checkout: clarify interactions with submodules
421
-
422
- The effect of sparse checkout settings on submodules is documented.
423
-
424
- Will merge to 'master'.
425
-
426
-
484
* ss/submodule-set-branch-in-c (2020-06-02) 1 commit
428
- - submodule: port subcommand 'set-branch' from shell to C
485
+ (merged to 'next' on 2020-06-18 at 8880b35c5a)
486
+ + submodule: port subcommand 'set-branch' from shell to C
487
488
Rewrite of parts of the scripted "git submodule" Porcelain command
489
continues; this time it is "git submodule set-branch" subcommand's
490
turn.
491
434
- Will merge to 'next'.
492
+ Will merge to 'master'.
493
494
495
* es/config-hooks (2020-05-21) 4 commits
@@ -473,55 +531,57 @@ of the repositories listed at
531
cf. <CABPp-BGdEyEeajYZj_rdxp=MyEQdszuyjVTax=hhYj3fOtRQUQ@mail.gmail.com>
532
533
476
-* bc/sha-256-part-2 (2020-05-27) 44 commits
477
- - remote-testgit: adapt for object-format
478
- - bundle: detect hash algorithm when reading refs
479
- - t5300: pass --object-format to git index-pack
480
- - t5704: send object-format capability with SHA-256
481
- - t5703: use object-format serve option
482
- - t5702: offer an object-format capability in the test
483
- - t/helper: initialize the repository for test-sha1-array
484
- - remote-curl: avoid truncating refs with ls-remote
485
- - t1050: pass algorithm to index-pack when outside repo
486
- - builtin/index-pack: add option to specify hash algorithm
487
- - remote-curl: detect algorithm for dumb HTTP by size
488
- - builtin/ls-remote: initialize repository based on fetch
489
- - t5500: make hash independent
490
- - serve: advertise object-format capability for protocol v2
491
- - connect: parse v2 refs with correct hash algorithm
492
- - connect: pass full packet reader when parsing v2 refs
493
- - Documentation/technical: document object-format for protocol v2
494
- - t1302: expect repo format version 1 for SHA-256
495
- - builtin/show-index: provide options to determine hash algo
496
- - t5302: modernize test formatting
497
- - packfile: compute and use the index CRC offset
498
- - t3200: mark assertion with SHA1 prerequisite
499
- - setup: set the_repository's hash algo when checking format
500
- - fetch-pack: parse and advertise the object-format capability
501
- - t5562: pass object-format in synthesized test data
502
- - builtin/clone: initialize hash algorithm properly
503
- - remote-curl: implement object-format extensions
504
- - transport-helper: implement object-format extensions
505
- - docs: update remote helper docs for object-format extensions
506
- - builtin/receive-pack: detect when the server doesn't support our hash
507
- - connect: detect algorithm when fetching refs
508
- - fetch-pack: detect when the server doesn't support our hash
509
- - connect: make parse_feature_value extern
510
- - send-pack: detect when the server doesn't support our hash
511
- - connect: add function to detect supported v1 hash functions
512
- - transport: add a hash algorithm member
513
- - pkt-line: add a member for hash algorithm
514
- - connect: add function to fetch value of a v2 server capability
515
- - connect: add function to parse multiple v1 capability values
516
- - remote: advertise the object-format capability on the server side
517
- - wrapper: add function to compare strings with different NUL termination
518
- - connect: have ref processing code take struct packet_reader
519
- - Documentation: document v1 protocol object-format capability
520
- - t1050: match object ID paths in a hash-insensitive way
534
+* bc/sha-256-part-2 (2020-06-19) 44 commits
535
+ (merged to 'next' on 2020-06-22 at dc153f366d)
536
+ + remote-testgit: adapt for object-format
537
+ + bundle: detect hash algorithm when reading refs
538
+ + t5300: pass --object-format to git index-pack
539
+ + t5704: send object-format capability with SHA-256
540
+ + t5703: use object-format serve option
541
+ + t5702: offer an object-format capability in the test
542
+ + t/helper: initialize the repository for test-sha1-array
543
+ + remote-curl: avoid truncating refs with ls-remote
544
+ + t1050: pass algorithm to index-pack when outside repo
545
+ + builtin/index-pack: add option to specify hash algorithm
546
+ + remote-curl: detect algorithm for dumb HTTP by size
547
+ + builtin/ls-remote: initialize repository based on fetch
548
+ + t5500: make hash independent
549
+ + serve: advertise object-format capability for protocol v2
550
+ + connect: parse v2 refs with correct hash algorithm
551
+ + connect: pass full packet reader when parsing v2 refs
552
+ + Documentation/technical: document object-format for protocol v2
553
+ + t1302: expect repo format version 1 for SHA-256
554
+ + builtin/show-index: provide options to determine hash algo
555
+ + t5302: modernize test formatting
556
+ + packfile: compute and use the index CRC offset
557
+ + t3200: mark assertion with SHA1 prerequisite
558
+ + setup: set the_repository's hash algo when checking format
559
+ + fetch-pack: parse and advertise the object-format capability
560
+ + t5562: pass object-format in synthesized test data
561
+ + builtin/clone: initialize hash algorithm properly
562
+ + remote-curl: implement object-format extensions
563
+ + transport-helper: implement object-format extensions
564
+ + docs: update remote helper docs for object-format extensions
565
+ + builtin/receive-pack: detect when the server doesn't support our hash
566
+ + connect: detect algorithm when fetching refs
567
+ + fetch-pack: detect when the server doesn't support our hash
568
+ + connect: make parse_feature_value extern
569
+ + send-pack: detect when the server doesn't support our hash
570
+ + connect: add function to detect supported v1 hash functions
571
+ + transport: add a hash algorithm member
572
+ + pkt-line: add a member for hash algorithm
573
+ + connect: add function to fetch value of a v2 server capability
574
+ + connect: add function to parse multiple v1 capability values
575
+ + remote: advertise the object-format capability on the server side
576
+ + wrapper: add function to compare strings with different NUL termination
577
+ + connect: have ref processing code take struct packet_reader
578
+ + Documentation: document v1 protocol object-format capability
579
+ + t1050: match object ID paths in a hash-insensitive way
580
+ (this branch is used by bc/sha-256-cvs-svn-updates.)
581
582
SHA-256 migration work continues.
583
524
- Ready???
584
+ Will merge to 'master'.
585
586
587
* jx/proc-receive-hook (2020-05-18) 11 commits
@@ -543,31 +603,36 @@ of the repositories listed at
603
Needs review.
604
605
546
-* hn/reftable (2020-06-17) 18 commits
547
- . SQUASH??? whitespace breakage fix
548
- . Add "test-tool dump-reftable" command.
549
- . Add reftable testing infrastructure
550
- . vcxproj: adjust for the reftable changes
551
- . Add GIT_DEBUG_REFS debugging mechanism
552
- . Hookup unittests for the reftable library.
553
- . Reftable support for git-core
554
- . Add reftable library
555
- . Add .gitattributes for the reftable/ directory
556
- . Iterate over the "refs/" namespace in for_each_[raw]ref
557
- . Move REF_LOG_ONLY to refs-internal.h
558
- . Treat REVERT_HEAD as a pseudo ref
559
- . Treat CHERRY_PICK_HEAD as a pseudo ref
560
- . Treat BISECT_HEAD as a pseudo ref
561
- . Make refs_ref_exists public
562
- . Write pseudorefs through ref backends.
563
- . checkout: add '\n' to reflog message
564
- . lib-t6000.sh: write tag using git-update-ref
606
+* hn/reftable (2020-06-22) 20 commits
607
+ - SQUASH??? fix new blank line at EOF
608
+ - Add "test-tool dump-reftable" command.
609
+ - Add reftable testing infrastructure
610
+ - git-prompt: prepare for reftable refs backend
611
+ - vcxproj: adjust for the reftable changes
612
+ - Add GIT_DEBUG_REFS debugging mechanism
613
+ - Hookup unittests for the reftable library.
614
+ - Reftable support for git-core
615
+ - Add standalone build infrastructure for reftable
616
+ - Add reftable library
617
+ - Add .gitattributes for the reftable/ directory
618
+ - Iterate over the "refs/" namespace in for_each_[raw]ref
619
+ - Move REF_LOG_ONLY to refs-internal.h
620
+ - Treat REVERT_HEAD as a pseudo ref
621
+ - Treat CHERRY_PICK_HEAD as a pseudo ref
622
+ - Treat BISECT_HEAD as a pseudo ref
623
+ - Make refs_ref_exists public
624
+ - Write pseudorefs through ref backends.
625
+ - checkout: add '\n' to reflog message
626
+ - lib-t6000.sh: write tag using git-update-ref
627
628
A new refs backend "reftable" to replace the traditional
629
combination of packed-refs files and one-file-per-ref loose refs
630
has been implemented and integrated for improved performance and
631
atomicity.
632
633
+ Does not seem to play well with the xl/upgrade-repo-format topic;
634
+ perhaps rework on top of 'master' after that topic graduates?
635
+
636
--------------------------------------------------
637
[Discarded]
638