What's cooking (2020/05 #06)

Junio C Hamano committed May 19, 2020 at 18:24 UTC e2a0456fe0c78c2d0894ad719bf30391a9fdd12b
1 file changed +255 -176
whats-cooking.txt
+255 -176
@@ -1,10 +1,10 @@
1 To: git@vger.kernel.org
2 Bcc: lwn@lwn.net
3 -Subject: What's cooking in git.git (May 2020, #05; Thu, 14)
3 +Subject: What's cooking in git.git (May 2020, #06; Tue, 19)
4 X-master-at: efcab5b7a3d2ce2ae4bf808b335938098b18d960
5 -X-next-at: 67701bce43065d3b2fbf3423f8cb13c0db322fee
5 +X-next-at: ede8c892b8d8194a9b64c2cfd62396141e5199d2
6
7 -What's cooking in git.git (May 2020, #05; Thu, 14)
7 +What's cooking in git.git (May 2020, #06; Tue, 19)
8 --------------------------------------------------
9
10 Here are the topics that have been cooking. Commits prefixed with
@@ -12,11 +12,10 @@ Here are the topics that have been cooking. Commits prefixed with
12 '+' are in 'next'. The ones marked with '.' do not appear in any of
13 the integration branches, but I am still holding onto them.
14
15 -2.27-rc0 has been tagged. I haven't seen much discussion on
16 -proposed topics for 2.26.3, so haven't merged anything (yet), but
17 -now the 'master' and 'next' front would be more quiescent, efforts
18 -focused on regressions, perhaps people can help pick the topics for
19 -the maintenance track.
15 +Git 2.27-rc1 tomorrow, with topics scheduled to be in 'master'
16 +(listed below). I am paying less attention than usual for topics
17 +that are not in 'next', but it would be good to have well-reviewed
18 +topics cooking in 'next' when the feature-freeze ends.
19
20 You can find the changes described here in the integration branches
21 of the repositories listed at
@@ -24,130 +23,240 @@ of the repositories listed at
23 http://git-blame.blogspot.com/p/git-public-repositories.html
24
25 --------------------------------------------------
27 -[Graduated to "master"]
26 +[New Topics]
27
29 -* ao/p4-d-f-conflict-recover (2020-05-10) 1 commit
30 - (merged to 'next' on 2020-05-11 at a83a2f6178)
31 - + git-p4: recover from inconsistent perforce history
28 +* ds/trace-log-progress-fix (2020-05-15) 1 commit
29 + (merged to 'next' on 2020-05-15 at d82c1fd03f)
30 + + progress: call trace2_region_leave() only after calling _enter()
31
33 - "git p4" learned to recover from a (broken) state where a directory
34 - and a file are recorded at the same path in the Perforce repository
35 - the same way as their clients do.
32 + Last-minute fix for our recent change to allow use of progress API
33 + as a traceable region.
34
35 + Will merge to 'master'.
36
38 -* cw/bisect-replay-with-dos (2020-05-08) 1 commit
39 - (merged to 'next' on 2020-05-11 at 2eb0edf56d)
40 - + bisect: allow CRLF line endings in "git bisect replay" input
37
42 - "git bisect replay" had trouble with input files when they used
43 - CRLF line ending, which has been corrected.
38 +* en/sparse-checkout (2020-05-15) 1 commit
39 + (merged to 'next' on 2020-05-15 at d4b3058787)
40 + + unpack-trees: also allow get_progress() to work on a different index
41
42 + Consistency fix to a topic already in 'master'.
43
46 -* ds/bloom-cleanup (2020-05-11) 7 commits
47 - (merged to 'next' on 2020-05-11 at 046d49d455)
48 - + completion: offer '--(no-)patch' among 'git log' options
49 - + bloom: use num_changes not nr for limit detection
50 - + bloom: de-duplicate directory entries
51 - + Documentation: changed-path Bloom filters use byte words
52 - + bloom: parse commit before computing filters
53 - + test-bloom: fix usage typo
54 - + bloom: fix whitespace around tab length
44 + Will merge to 'master'.
45
56 - Code cleanup and typofixes.
57 - This is an early part of ds/line-log-on-bloom that has been cooking
58 - in 'next'.
46
47 +* jc/fix-tap-output-under-bash (2020-05-15) 3 commits
48 + (merged to 'next' on 2020-05-15 at de8f92d652)
49 + + Revert "tests: when run in Bash, annotate test failures with file name/line number"
50 + + Revert "ci: add a problem matcher for GitHub Actions"
51 + + Revert "t/test_lib: avoid naked bash arrays in file_lineno"
52
61 -* es/bugreport-with-hooks (2020-05-07) 1 commit
62 - (merged to 'next' on 2020-05-11 at 2dd5d90a34)
63 - + bugreport: collect list of populated hooks
64 - (this branch is used by es/bugreport-shell.)
53 + A recent attempt to make the test output nicer to view on CI
54 + systems broke TAP output under bash. The effort has been reverted
55 + to be re-attempted in the next cycle.
56
66 - "git bugreport" learned to report enabled hooks in the repository.
57 + Will merge to 'master'.
58
59
69 -* es/trace-log-progress (2020-05-12) 1 commit
70 - (merged to 'next' on 2020-05-13 at a127540258)
71 - + trace2: log progress time and throughput
60 +* ak/slab-decl-cleanup (2020-05-18) 1 commit
61 + (merged to 'next' on 2020-05-19 at 7de41005de)
62 + + commit-slab-decl.h: update include guard
63
73 - Teach codepaths that show progress meter to also use the
74 - start_progress() and the stop_progress() calls as a "region" to be
75 - traced.
64 + Code clean-up.
65
66 + Will merge to 'master'.
67
78 -* jc/codingstyle-compare-with-null (2020-05-08) 1 commit
79 - (merged to 'next' on 2020-05-11 at d18f0d930e)
80 - + CodingGuidelines: do not ==/!= compare with 0 or '\0' or NULL
68
82 - Doc update.
69 +* an/merge-single-strategy-optim (2020-05-19) 1 commit
70 + - merge: optimization to skip evaluate_result for single strategy
71
72 + Code optimization for a common case.
73
85 -* js/rebase-autosquash-double-fixup-fix (2020-05-09) 1 commit
86 - (merged to 'next' on 2020-05-11 at 3cdf7f7ece)
87 - + rebase --autosquash: fix a potential segfault
74 + Will merge to 'next'.
75
89 - "rebase -i" segfaulted when rearranging a sequence that has a
90 - fix-up that applies another fix-up (which may or may not be a
91 - fix-up of yet another step).
76
77 +* cc/upload-pack-data (2020-05-18) 13 commits
78 + - upload-pack: use upload_pack_data fields in receive_needs()
79 + - upload-pack: pass upload_pack_data to create_pack_file()
80 + - upload-pack: remove static variable 'stateless_rpc'
81 + - upload-pack: pass upload_pack_data to check_non_tip()
82 + - upload-pack: pass upload_pack_data to send_ref()
83 + - upload-pack: move symref to upload_pack_data
84 + - upload-pack: use upload_pack_data writer in receive_needs()
85 + - upload-pack: pass upload_pack_data to receive_needs()
86 + - upload-pack: pass upload_pack_data to get_common_commits()
87 + - upload-pack: use 'struct upload_pack_data' in upload_pack()
88 + - upload-pack: move 'struct upload_pack_data' around
89 + - upload-pack: move {want,have}_obj to upload_pack_data
90 + - upload-pack: remove unused 'wants' from upload_pack_data
91 +
92 + Code clean-up.
93 +
94 +
95 +* dl/remote-curl-deadlock-fix (2020-05-19) 7 commits
96 + - stateless-connect: send response end packet
97 + - pkt-line: define PACKET_READ_RESPONSE_END
98 + - remote-curl: error on incomplete packet
99 + - pkt-line: extern packet_length()
100 + - transport: extract common fetch_pack() call
101 + - remote-curl: remove label indentation
102 + - remote-curl: fix typo
103 +
104 + On-the-wire protocol v2 easily falls into a deadlock between the
105 + remote-curl helper and the fetch-pack process when the server side
106 + prematurely throws an error and disconnects. The communication has
107 + been updated to make it more robust.
108
94 -* jt/t5500-unflake (2020-05-06) 1 commit
95 - (merged to 'next' on 2020-05-13 at 11450265ec)
96 - + t5500: count objects through stderr, not trace
109 + Will merge to 'next'.
110
98 - Test fix for a topic already in 'master' and meant for 'maint'.
111
112 +* es/bugreport (2020-05-18) 1 commit
113 + (merged to 'next' on 2020-05-19 at ede8c892b8)
114 + + git-bugreport.txt: adjust reference to strftime(3)
115
101 -* rs/fsck-duplicate-names-in-trees (2020-05-11) 1 commit
102 - (merged to 'next' on 2020-05-11 at f603ca2528)
103 - + fsck: report non-consecutive duplicate names in trees
116 + Doc fix.
117
105 - "git fsck" ensures that the paths recorded in tree objects are
106 - sorted and without duplicates, but it failed to notice a case where
107 - a blob is followed by entries that sort before a tree with the same
108 - name. This has been corrected.
118 + Will merge to 'master'.
119
120
111 -* sn/midx-repack-with-config (2020-05-10) 2 commits
112 - (merged to 'next' on 2020-05-11 at d73f8f51d6)
113 - + multi-pack-index: respect repack.packKeptObjects=false
114 - + midx: teach "git multi-pack-index repack" honor "git repack" configurations
121 +* gp/hppa-stack-test-fix (2020-05-18) 1 commit
122 + - tests: skip small-stack tests on hppa architecture
123
116 - "git multi-pack-index repack" has been taught to honor some
117 - repack.* configuration variables.
124 + Platform dependent tweak to a test for HP-PA.
125
119 ---------------------------------------------------
120 -[New Topics]
126 + Will merge to 'next'.
127
122 -* cb/no-more-gmtime (2020-05-14) 1 commit
123 - - compat: remove gmtime
128
125 - Code clean-up by removing a compatibility implementation of a
126 - function we no longer use.
129 +* jk/ci-only-on-selected-branches (2020-05-18) 1 commit
130 + - ci/config: correct instruction for CI preferences
131 +
132 + Dev support.
133
134 Will merge to 'next'.
135
136
131 -* cb/t4210-illseq-auto-detect (2020-05-14) 1 commit
132 - - t4210: detect REG_ILLSEQ dynamically
137 +* la/diff-relative-config (2020-05-19) 1 commit
138 + - diff: add config option relative
139
134 - As FreeBSD is not the only platform whose regexp library needs
135 - REG_ILLSEQ prerequisite, add a logic to detect the prerequisite
136 - automatically.
140 + The commands in the "diff" family learned to honor "diff.relative"
141 + configuration variable.
142
143 + Almost there.
144 + cf. <20200519230124.GA12509@danh.dev>
145
139 -* en/merge-rename-rename-worktree-fix (2020-05-14) 1 commit
140 - - merge-recursive: fix rename/rename(1to2) for working tree with a binary
146
142 - When a binary file gets renamed and then merged with a minor
143 - change, the copy on the working tree while conflict resolution
144 - sometimes got "their" version and not "ours".
147 +* ma/doc-fixes (2020-05-18) 5 commits
148 + - git-sparse-checkout.txt: add missing '
149 + - git-credential.txt: use list continuation
150 + - git-commit-graph.txt: fix list rendering
151 + - git-commit-graph.txt: fix grammo
152 + - date-formats.txt: fix list continuation
153 +
154 + Various doc fixes.
155 +
156 + Will merge to 'next'.
157 +
158 +
159 +* dd/t5703-grep-a-fix (2020-05-19) 1 commit
160 + - t5703: replace "grep -a" usage by perl
161 +
162 + Update an unconditional use of "grep -a" with a perl script in a test.
163 +
164 + Will merge to 'next'.
165 +
166 +
167 +* ds/multi-pack-verify (2020-05-19) 1 commit
168 + - fsck: use ERROR_MULTI_PACK_INDEX
169 +
170 + Fix for a copy-and-paste error introduced during 2.20 era.
171
172 Will merge to 'next'.
173
174 +
175 +* jt/avoid-prefetch-when-able-in-diff (2020-05-19) 1 commit
176 + - t4067: make rename detection test output raw diff
177 +
178 + Test-coverage enhancement.
179 +
180 + Will merge to 'next' and then to 'master'.
181 +
182 --------------------------------------------------
183 [Stalled]
184
185 +* dr/push-remoteref-fix (2020-04-23) 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 +
194 + Expecting a reroll.
195 + cf. <20200416152145.wp2zeibxmuyas6y6@feanor>
196 +
197 +
198 +* pw/rebase-i-more-options (2020-04-29) 5 commits
199 + - rebase: add --reset-author-date
200 + - rebase -i: support --ignore-date
201 + - sequencer: rename amend_author to author_to_free
202 + - rebase -i: support --committer-date-is-author-date
203 + - rebase -i: add --ignore-whitespace flag
204 +
205 + "git rebase -i" learns a bit more options.
206 +
207 + Needs review.
208 +
209 +
210 +* jk/complete-git-switch (2020-04-28) 11 commits
211 + - completion: complete remote branches for git switch --track
212 + - completion: recognize -c/-C when completing for git switch
213 + - completion: fix completion for git switch with no options
214 + - completion: perform DWIM logic directly in __git_complete_refs
215 + - completion: extract function __git_dwim_remote_heads
216 + - completion: rename --track option of __git_complete_refs
217 + - completion: stop completing refs for git switch --orphan
218 + - completion: add tests showing lack of support for git switch -c/-C
219 + - completion: add test highlighting subpar git switch --track completion
220 + - completion: add test showing subpar git switch completion
221 + - completion: add some simple test cases for git switch completion
222 +
223 + The command line completion (in contrib/) learned to complete
224 + options that the "git switch" command takes.
225 +
226 + Needs review.
227 +
228 +
229 +* dl/test-must-fail-fixes-5 (2020-05-05) 4 commits
230 + - lib-submodule-update: pass OVERWRITING_FAIL
231 + - lib-submodule-update: prepend "git" to $command
232 + - lib-submodule-update: consolidate --recurse-submodules
233 + - lib-submodule-update: add space after function name
234 +
235 + The effort to avoid using test_must_fail on non-git command continues.
236 +
237 + Needs review.
238 + cf. <cover.1588162842.git.liu.denton@gmail.com>
239 +
240 +
241 +* mr/bisect-in-c-2 (2020-04-23) 12 commits
242 + - bisect--helper: retire `--bisect-autostart` subcommand
243 + - bisect--helper: retire `--write-terms` subcommand
244 + - bisect--helper: retire `--check-expected-revs` subcommand
245 + - bisect--helper: reimplement `bisect_state` & `bisect_head` shell functions in C
246 + - bisect--helper: retire `--next-all` subcommand
247 + - bisect--helper: retire `--bisect-clean-state` subcommand
248 + - bisect--helper: finish porting `bisect_start()` to C
249 + - bisect--helper: reimplement `bisect_next` and `bisect_auto_next` shell functions in C
250 + - bisect--helper: reimplement `bisect_autostart` shell function in C
251 + - bisect--helper: introduce new `write_in_file()` function
252 + - bisect--helper: use '-res' in 'cmd_bisect__helper' return
253 + - bisect--helper: fix `cmd_*()` function switch default return
254 +
255 + Rewrite of the remainder of "git bisect" script in C continues.
256 +
257 + Needs review.
258 +
259 +
260 * mk/use-size-t-in-zlib (2018-10-15) 1 commit
261 - zlib.c: use size_t for size
262
@@ -158,13 +267,47 @@ of the repositories listed at
267 --------------------------------------------------
268 [Cooking]
269
270 +* cb/no-more-gmtime (2020-05-14) 1 commit
271 + (merged to 'next' on 2020-05-15 at 160e0ada09)
272 + + compat: remove gmtime
273 +
274 + Code clean-up by removing a compatibility implementation of a
275 + function we no longer use.
276 +
277 + Will merge to 'master'.
278 +
279 +
280 +* cb/t4210-illseq-auto-detect (2020-05-18) 2 commits
281 + - t4210: detect REG_ILLSEQ dynamically and skip affected tests
282 + - t/helper: teach test-regex to report pattern errors (like REG_ILLSEQ)
283 +
284 + As FreeBSD is not the only platform whose regexp library needs
285 + REG_ILLSEQ prerequisite, add a logic to detect the prerequisite
286 + automatically.
287 +
288 + Will merge to 'next'.
289 +
290 +
291 +* en/merge-rename-rename-worktree-fix (2020-05-14) 1 commit
292 + (merged to 'next' on 2020-05-15 at 7e1ddfa3c9)
293 + + merge-recursive: fix rename/rename(1to2) for working tree with a binary
294 +
295 + When a binary file gets modified and renamed on both sides of history
296 + to different locations, both files would be written to the working
297 + tree but both would have the contents from "ours". This has been
298 + corrected so that the path from each side gets their original content.
299 +
300 + Will merge to 'master'.
301 +
302 +
303 * bk/p4-prepare-p4-only-fix (2020-05-12) 1 commit
304 - git-p4.py: fix --prepare-p4-only error with multiple commits
305
306 The "--prepare-p4-only" option is supposed to stop after replaying
307 one changeset, but kept going (by mistake?)
308
167 - Needs review by some p4 folks.
309 + Will merge to 'next'.
310 + cf. <CAE5ih797YYxsR2H0TA65w9W-1jF4jQLayja_nGjQMGtc=PB6Jw@mail.gmail.com>
311
312
313 * jt/curl-verbose-on-trace-curl (2020-05-11) 2 commits
@@ -174,7 +317,7 @@ of the repositories listed at
317
318 Rewrite support for GIT_CURL_VERBOSE in terms of GIT_TRACE_CURL.
319
177 - Expecting further work on optionally disabling reacting authinfo
320 + Expecting further work on optionally disabling redacting authinfo
321
322
323 * mt/grep-sparse-checkout (2020-05-11) 4 commits
@@ -189,16 +332,16 @@ of the repositories listed at
332 Expecting a reroll.
333
334
192 -* js/ci-sdk-download-fix (2020-05-12) 1 commit
193 - - ci: avoid pounding on the poor ci-artifacts container
335 +* js/ci-sdk-download-fix (2020-05-15) 1 commit
336 + (merged to 'next' on 2020-05-15 at 9b28f7ec87)
337 + + ci: avoid pounding on the poor ci-artifacts container
338
339 Instead of downloading Windows SDK for CI jobs for windows builds
340 from an external site (wingit.blob.core.windows.net), use the one
341 created in the windows-build job, to work around quota issues at
342 the external site.
343
200 - Will merge to 'next'.
201 - Hopefully this can go away once cmake-for-windows-build topic lands?
344 + Will merge to 'master'.
345
346
347 * bc/sha-256-part-2 (2020-05-13) 44 commits
@@ -251,11 +394,12 @@ of the repositories listed at
394
395
396 * dd/t1509-i18n-fix (2020-05-13) 1 commit
254 - - t1509: correct i18n test
397 + (merged to 'next' on 2020-05-15 at a6342455c1)
398 + + t1509: correct i18n test
399
400 A few tests were not i18n clean.
401
258 - Will merge to 'next'.
402 + Will merge to 'master'.
403
404
405 * es/bugreport-shell (2020-05-12) 2 commits
@@ -264,6 +408,11 @@ of the repositories listed at
408
409 "git bugreport" learns to report what shell is in use.
410
411 + Will merge to 'next'.
412 + We may want to learn more details than just the path, but
413 + that can come later.
414 + cf. <20200512235924.GC6605@camp.crustytoothpaste.net>
415 +
416
417 * ds/line-log-on-bloom (2020-05-11) 5 commits
418 (merged to 'next' on 2020-05-11 at 046d49d455)
@@ -279,8 +428,7 @@ of the repositories listed at
428 Will cook in 'next'.
429
430
282 -* tb/commit-graph-no-check-oids (2020-05-14) 9 commits
283 - - fixup! builtin/commit-graph.c: extract 'read_one_commit()'
431 +* tb/commit-graph-no-check-oids (2020-05-18) 8 commits
432 - commit-graph: drop COMMIT_GRAPH_WRITE_CHECK_OIDS flag
433 - t5318: reorder test below 'graph_read_expect'
434 - commit-graph.c: simplify 'fill_oids_from_commits'
@@ -292,89 +440,20 @@ of the repositories listed at
440
441 Clean-up the commit-graph codepath.
442
295 -
296 -* dl/test-must-fail-fixes-5 (2020-05-05) 4 commits
297 - - lib-submodule-update: pass OVERWRITING_FAIL
298 - - lib-submodule-update: prepend "git" to $command
299 - - lib-submodule-update: consolidate --recurse-submodules
300 - - lib-submodule-update: add space after function name
301 -
302 - The effort to avoid using test_must_fail on non-git command continues.
303 -
304 - Needs review.
305 - cf. <cover.1588162842.git.liu.denton@gmail.com>
306 -
307 -
308 -* mr/bisect-in-c-2 (2020-04-23) 12 commits
309 - - bisect--helper: retire `--bisect-autostart` subcommand
310 - - bisect--helper: retire `--write-terms` subcommand
311 - - bisect--helper: retire `--check-expected-revs` subcommand
312 - - bisect--helper: reimplement `bisect_state` & `bisect_head` shell functions in C
313 - - bisect--helper: retire `--next-all` subcommand
314 - - bisect--helper: retire `--bisect-clean-state` subcommand
315 - - bisect--helper: finish porting `bisect_start()` to C
316 - - bisect--helper: reimplement `bisect_next` and `bisect_auto_next` shell functions in C
317 - - bisect--helper: reimplement `bisect_autostart` shell function in C
318 - - bisect--helper: introduce new `write_in_file()` function
319 - - bisect--helper: use '-res' in 'cmd_bisect__helper' return
320 - - bisect--helper: fix `cmd_*()` function switch default return
321 -
322 - Rewrite of the remainder of "git bisect" script in C continues.
323 -
324 - Needs review.
325 -
326 -
327 -* jk/complete-git-switch (2020-04-28) 11 commits
328 - - completion: complete remote branches for git switch --track
329 - - completion: recognize -c/-C when completing for git switch
330 - - completion: fix completion for git switch with no options
331 - - completion: perform DWIM logic directly in __git_complete_refs
332 - - completion: extract function __git_dwim_remote_heads
333 - - completion: rename --track option of __git_complete_refs
334 - - completion: stop completing refs for git switch --orphan
335 - - completion: add tests showing lack of support for git switch -c/-C
336 - - completion: add test highlighting subpar git switch --track completion
337 - - completion: add test showing subpar git switch completion
338 - - completion: add some simple test cases for git switch completion
339 -
340 - The command line completion (in contrib/) learned to complete
341 - options that the "git switch" command takes.
342 -
343 - Needs review.
344 -
345 -
346 -* dr/push-remoteref-fix (2020-04-23) 1 commit
347 - - remote.c: fix handling of %(push:remoteref)
348 -
349 - The "%(push:remoteref)" placeholder in the "--format=" argument of
350 - "git format-patch" (and friends) only showed what got explicitly
351 - configured, not what ref at the receiving end would be updated when
352 - "git push" was used, as it ignored the default behaviour (e.g. update
353 - the same ref as the source).
354 -
355 - Expecting a reroll.
356 - cf. <20200416152145.wp2zeibxmuyas6y6@feanor>
357 -
358 -
359 -* pw/rebase-i-more-options (2020-04-29) 5 commits
360 - - rebase: add --reset-author-date
361 - - rebase -i: support --ignore-date
362 - - sequencer: rename amend_author to author_to_free
363 - - rebase -i: support --committer-date-is-author-date
364 - - rebase -i: add --ignore-whitespace flag
365 -
366 - "git rebase -i" learns a bit more options.
367 -
368 - Needs review.
443 + Will merge to 'next'.
444
445
371 -* jx/proc-receive-hook (2020-05-07) 7 commits
446 +* jx/proc-receive-hook (2020-05-18) 11 commits
447 - doc: add documentation for the proc-receive hook
448 + - transport: parse report options for tracking refs
449 + - t5411: test updates of remote-tracking branches
450 - receive-pack: new config receive.procReceiveRefs
451 - refs.c: refactor to reuse ref_is_hidden()
452 - receive-pack: feed report options to post-receive
453 + - doc: add document for capability report-status-v2
454 - New capability "report-status-v2" for git-push
455 - receive-pack: add new proc-receive hook
456 + - t5411: add basic test cases for proc-receive hook
457 - transport: not report a non-head push as a branch
458
459 "git receive-pack" that accepts requests by "git push" learned to
@@ -395,18 +474,18 @@ of the repositories listed at
474 Preliminary clean-ups around refs API, plus file format
475 specification documentation for the reftable backend.
476
398 - I splitted these out of the hn/reftable topic, hoping that these
399 - should be easier to polish and merge quickly than the rest of the
400 - series.
477 + Almost there.
478
479
403 -* hn/reftable (2020-05-11) 7 commits
404 - - Add some reftable testing infrastructure
480 +* hn/reftable (2020-05-18) 9 commits
481 + - Add reftable testing infrastructure
482 - vcxproj: adjust for the reftable changes
483 + - Add GIT_DEBUG_REFS debugging mechanism
484 - Reftable support for git-core
485 - Add reftable library
486 - Add .gitattributes for the reftable/ directory
487 - Iterate over the "refs/" namespace in for_each_[raw]ref
488 + - Move REF_LOG_ONLY to refs-internal.h
489 - Write pseudorefs through ref backends.
490 (this branch uses hn/refs-cleanup.)
491