f5e0f95eca · matheusbsbs/matheus-bagliano

@@ -1,10 +1,10 @@

1 To: git@vger.kernel.org

2 Bcc: lwn@lwn.net

3 -Subject: What's cooking in git.git (Feb 2020, #02; Mon, 10)

4 -X-master-at: de93cc14ab7e8db7645d8dbe4fd2603f76d5851f

5 -X-next-at: 5c85173df499e49e5e4dc041fb5d5e4a8594563e

3 +Subject: What's cooking in git.git (Feb 2020, #03; Wed, 12)

4 +X-master-at: d8437c57fa0752716dde2d3747e7c22bf7ce2e41

5 +X-next-at: cffb4d697ea3219afa675b81c08f1fe6006a74d1

6

7 -What's cooking in git.git (Feb 2020, #02; Mon, 10)

7 +What's cooking in git.git (Feb 2020, #03; Wed, 12)

8 --------------------------------------------------

9

10 Here are the topics that have been cooking. Commits prefixed with

@@ -20,318 +20,185 @@ of the repositories listed at

20 --------------------------------------------------

21 [Graduated to "master"]

22

23 -* am/checkout-file-and-ref-ref-ambiguity (2020-01-07) 2 commits

24 - (merged to 'next' on 2020-01-27 at daf1dadc78)

25 - + checkout: don't revert file on ambiguous tracking branches

26 - + parse_branchname_arg(): extract part as new function

27 -

28 - "git checkout X" did not correctly fail when X is not a local

29 - branch but could name more than one remote-tracking branches

30 - (i.e. to be dwimmed as the starting point to create a corresponding

31 - local branch), which has been corrected.

32 -

33 -

34 -* bc/hash-independent-tests-part-7 (2020-01-15) 20 commits

35 - (merged to 'next' on 2020-01-27 at a519c8fae4)

36 - + t5604: make hash independent

37 - + t5601: switch into repository to hash object

38 - + t5562: use $ZERO_OID

39 - + t5540: make hash size independent

40 - + t5537: make hash size independent

41 - + t5530: compute results based on object length

42 - + t5512: abstract away SHA-1-specific constants

43 - + t5510: make hash size independent

44 - + t5504: make hash algorithm independent

45 - + t5324: make hash size independent

46 - + t5319: make test work with SHA-256

47 - + t5319: change invalid offset for SHA-256 compatibility

48 - + t5318: update for SHA-256

49 - + t4300: abstract away SHA-1-specific constants

50 - + t4204: make hash size independent

51 - + t4202: abstract away SHA-1-specific constants

52 - + t4200: make hash size independent

53 - + t4134: compute appropriate length constant

54 - + t4066: compute index line in diffs

55 - + t4054: make hash-size independent

56 -

57 - Preparation of test scripts for the day when the object names will

58 - use SHA-256 continues.

59 -

60 -

61 -* dl/test-must-fail-fixes (2019-12-20) 15 commits

62 - (merged to 'next' on 2020-01-27 at ed382cd096)

63 - + t1507: inline full_name()

64 - + t1507: run commands within test_expect_success

65 - + t1507: stop losing return codes of git commands

66 - + t1501: remove use of `test_might_fail cp`

67 - + t1409: use test_path_is_missing()

68 - + t1409: let sed open its own input file

69 - + t1307: reorder `nongit test_must_fail`

70 - + t1306: convert `test_might_fail rm` to `rm -f`

71 - + t0020: use ! check_packed_refs_marked

72 - + t0020: don't use `test_must_fail has_cr`

73 - + t0003: don't use `test_must_fail attr_check`

74 - + t0003: use test_must_be_empty()

75 - + t0003: use named parameters in attr_check()

76 - + t0000: replace test_must_fail with run_sub_test_lib_test_err()

77 - + t/lib-git-p4: use test_path_is_missing()

78 -

79 - Test clean-up.

80 -

81 -

82 -* en/fill-directory-fixes-more (2020-01-16) 4 commits

83 - (merged to 'next' on 2020-01-27 at 9a27e1495b)

84 - + dir: point treat_leading_path() warning to the right place

85 - + dir: restructure in a way to avoid passing around a struct dirent

86 - + dir: treat_leading_path() and read_directory_recursive(), round 2

87 - + clean: demonstrate a bug with pathspecs

88 - (this branch is used by en/fill-directory-exponential.)

89 -

90 - Corner case bugs in "git clean" that stems from a (necessarily for

91 - performance reasons) awkward calling convention in the directory

92 - enumeration API has been corrected.

93 -

94 -

95 -* es/fetch-show-failed-submodules-atend (2020-01-17) 1 commit

96 - (merged to 'next' on 2020-01-27 at 7b41163881)

97 - + fetch: emphasize failure during submodule fetch

98 -

99 - A fetch that is told to recursively fetch updates in submodules

100 - inevitably produces reams of output, and it becomes hard to spot

101 - error messages. The command has been taught to enumerate

102 - submodules that had errors at the end of the operation.

103 -

104 -

105 -* js/add-p-leftover-bits (2020-01-15) 10 commits

106 - (merged to 'next' on 2020-01-27 at ec5021a1bd)

107 - + ci: include the built-in `git add -i` in the `linux-gcc` job

108 - + built-in add -p: handle Escape sequences more efficiently

109 - + built-in add -p: handle Escape sequences in interactive.singlekey mode

110 - + built-in add -p: respect the `interactive.singlekey` config setting

111 - + terminal: add a new function to read a single keystroke

112 - + terminal: accommodate Git for Windows' default terminal

113 - + terminal: make the code of disable_echo() reusable

114 - + built-in add -p: handle diff.algorithm

115 - + built-in add -p: support interactive.diffFilter

116 - + t3701: adjust difffilter test

117 - (this branch uses js/patch-mode-in-others-in-c.)

118 -

119 - The final leg of rewriting "add -i/-p" in C.

120 -

121 -

122 -* js/patch-mode-in-others-in-c (2019-12-21) 7 commits

123 - (merged to 'next' on 2020-01-27 at 36f6756aae)

124 - + commit --interactive: make it work with the built-in `add -i`

125 - + built-in add -p: implement the "worktree" patch modes

126 - + built-in add -p: implement the "checkout" patch modes

127 - + built-in stash: use the built-in `git add -p` if so configured

128 - + legacy stash -p: respect the add.interactive.usebuiltin setting

129 - + built-in add -p: implement the "stash" and "reset" patch modes

130 - + built-in add -p: prepare for patch modes other than "stage"

131 - (this branch is used by js/add-p-leftover-bits.)

132 -

133 - The effort to move "git-add--interactive" to C continues.

134 -

135 -

136 -* km/submodule-add-errmsg (2020-01-15) 1 commit

137 - (merged to 'next' on 2020-01-27 at 0ff8378907)

138 - + submodule add: show 'add --dry-run' stderr when aborting

139 -

140 - Improve error message generation for "git submodule add".

141 -

142 -

143 -* mt/sparse-checkout-doc-update (2020-01-23) 2 commits

144 - (merged to 'next' on 2020-01-27 at dbd5a83613)

145 - + completion: add support for sparse-checkout

146 - + doc: sparse-checkout: mention --cone option

23 +* dt/submodule-rm-with-stale-cache (2020-01-28) 1 commit

24 + (merged to 'next' on 2020-02-05 at 701ef6c41b)

25 + + git rm submodule: succeed if .gitmodules index stat info is zero

26

148 - Doc update.

27 + Running "git rm" on a submodule failed unnecessarily when

28 + .gitmodules is only cache-dirty, which has been corrected.

29

30

151 -* pb/recurse-submodule-in-worktree-fix (2020-01-22) 4 commits

152 - (merged to 'next' on 2020-01-27 at e4cb1e376a)

153 - + submodule.c: use get_git_dir() instead of get_git_common_dir()

154 - + t2405: clarify test descriptions and simplify test

155 - + t2405: use git -C and test_commit -C instead of subshells

156 - + t7410: rename to t2405-worktree-submodule.sh

31 +* es/submodule-fetch-message-fix (2020-02-07) 1 commit

32 + (merged to 'next' on 2020-02-11 at b8af267e1b)

33 + + submodule: add newline on invalid submodule error

34

158 - The "--recurse-submodules" option of various subcommands did not

159 - work well when run in an alternate worktree, which has been

160 - corrected.

35 + Error message fix.

36

162 ---------------------------------------------------

163 -[New Topics]

37

165 -* hn/reftable (2020-02-10) 5 commits

166 - - Reftable support for git-core

167 - - Add reftable library

168 - - refs: document how ref_iterator_advance_fn should handle symrefs

169 - - create .git/refs in files-backend.c

170 - - refs.h: clarify reflog iteration order

38 +* hw/doc-git-dir (2020-01-30) 1 commit

39 + (merged to 'next' on 2020-02-05 at 839ae05c00)

40 + + git: update documentation for --git-dir

41

172 - A new refs backend "reftable" to replace the traditional

173 - combination of packed-refs files and one-file-per-ref loose refs

174 - has been implemented and integrated for improved performance and

175 - atomicity.

42 + One effect of specifying where the GIT_DIR is (either with the

43 + environment variable, or with the "git --git-dir=<where> cmd"

44 + option) is to disable the repository discovery. This has been

45 + placed a bit more stress in the documentation, as new users often

46 + get confused.

47

177 - At v5.

48

49 +* jb/multi-pack-index-docfix (2020-02-10) 1 commit

50 + (merged to 'next' on 2020-02-11 at 79af4f7bda)

51 + + pack-format: correct multi-pack-index description

52

180 -* es/bugreport (2020-02-10) 15 commits

181 - - bugreport: summarize contents of alternates file

182 - - bugreport: list contents of $OBJDIR/info

183 - - bugreport: add packed object summary

184 - - bugreport: count loose objects

185 - - bugreport: collect list of populated hooks

186 - - bugreport: add config values from safelist

187 - - bugreport: generate config safelist based on docs

188 - - bugreport: include user interactive shell

189 - - bugreport: add git-remote-https version

190 - - bugreport: add compiler info

191 - - bugreport: add uname info

192 - - bugreport: gather git version and build info

193 - - bugreport: add tool to generate debugging info

194 - - help: add shell-path to --build-options

195 - - help: move list_config_help to builtin/help

196 - (this branch uses mr/show-config-scope; is tangled with bw/remote-rename-update-config.)

53 + Doc fix.

54

198 - The "bugreport" tool.

55

200 - at v6.

56 +* jb/parse-options-message-fix (2020-02-05) 1 commit

57 + (merged to 'next' on 2020-02-11 at 58058cc68c)

58 + + parse-options: lose an unnecessary space in an error message

59

60 + Error message fix.

61

203 -* gs/commit-graph-path-filter (2020-02-06) 11 commits

204 - - commit-graph: add GIT_TEST_COMMIT_GRAPH_CHANGED_PATHS test flag

205 - - revision.c: use Bloom filters to speed up path based revision walks

206 - - commit-graph: add --changed-paths option to write subcommand

207 - - commit-graph: reuse existing Bloom filters during write.

208 - - commit-graph: write Bloom filters to commit graph file

209 - - commit-graph: examine commits by generation number

210 - - commit-graph: examine changed-path objects in pack order

211 - - commit-graph: compute Bloom filters for changed paths

212 - - diff: halt tree-diff early after max_changes

213 - - bloom: core Bloom filter implementation for changed paths

214 - - commit-graph: use MAX_NUM_CHUNKS

62

216 - Introduce an extension to the commit-graph to make it efficient to

217 - check for the paths that were modified at each commit using Bloom

218 - filters.

63 +* jc/skip-prefix (2020-01-31) 1 commit

64 + (merged to 'next' on 2020-02-05 at f6893a42d0)

65 + + C: use skip_prefix() to avoid hardcoded string length

66

220 - Breakage due to byte-order dependency reported.

67 + Code simplification.

68

69

223 -* bc/hash-independent-tests-part-8 (2020-02-07) 21 commits

224 - - t6024: update for SHA-256

225 - - t6006: make hash size independent

226 - - t6000: abstract away SHA-1-specific constants

227 - - t5703: make test work with SHA-256

228 - - t5607: make hash size independent

229 - - t5318: update for SHA-256

230 - - t5515: make test hash independent

231 - - t5321: make test hash independent

232 - - t5313: make test hash independent

233 - - t5309: make test hash independent

234 - - t5302: make hash size independent

235 - - t4060: make test work with SHA-256

236 - - t4211: add test cases for SHA-256

237 - - t4211: move SHA-1-specific test cases into a directory

238 - - t4013: make test hash independent

239 - - t3311: make test work with SHA-256

240 - - t3310: make test work with SHA-256

241 - - t3309: make test work with SHA-256

242 - - t3308: make test work with SHA-256

243 - - t3206: make hash size independent

244 - - t/lib-pack: support SHA-256

70 +* jk/clang-sanitizer-fixes (2020-01-28) 5 commits

71 + (merged to 'next' on 2020-02-05 at 8ef84c2d96)

72 + + obstack: avoid computing offsets from NULL pointer

73 + + xdiff: avoid computing non-zero offset from NULL pointer

74 + + avoid computing zero offsets from NULL pointer

75 + + merge-recursive: use subtraction to flip stage

76 + + merge-recursive: silence -Wxor-used-as-pow warning

77

246 - Preparation for SHA-256 migration continues.

78 + C pedantry ;-) fix.

79

248 - Will merge to 'next'.

80

81 +* jk/escaped-wildcard-dwim (2020-01-27) 1 commit

82 + (merged to 'next' on 2020-02-05 at ebee5e92a7)

83 + + verify_filename(): handle backslashes in "wildcards are pathspecs" rule

84

251 -* es/submodule-fetch-message-fix (2020-02-07) 1 commit

252 - - submodule: add newline on invalid submodule error

85 + Disambiguation logic to tell revisions and pathspec apart has been

86 + tweaked so that backslash-escaped glob special characters do not

87 + count in the "wildcards are pathspec" rule.

88

254 - Error message fix.

89

256 - Will merge to 'next'.

90 +* jk/push-default-doc (2020-01-29) 1 commit

91 + (merged to 'next' on 2020-02-05 at 33167357a1)

92 + + doc: drop "explicitly given" from push.default description

93

94 + Doc update.

95

259 -* ma/diff-doc-clarify-regexp-example (2020-02-09) 1 commit

260 - - diff-options.txt: avoid "regex" overload in example

96

262 - Doc clarification.

97 +* jn/pretend-object-doc (2020-01-06) 1 commit

98 + (merged to 'next' on 2020-02-05 at 1d52a5c72d)

99 + + sha1-file: document how to use pretend_object_file

100

264 - Will merge to 'next'.

101 + Warn programmers about pretend_object_file() that allows the code

102 + to tentatively use in-core objects.

103

104

267 -* jb/multi-pack-index-docfix (2020-02-10) 1 commit

268 - - pack-format: correct multi-pack-index description

105 +* js/ci-squelch-doc-warning (2020-02-10) 1 commit

106 + (merged to 'next' on 2020-02-11 at 4fec29107e)

107 + + ci: ignore rubygems warning in the "Documentation" job

108

270 - Doc fix.

109 + Squelch unhelpful warning message during documentation build.

110

272 - Will merge to 'next'.

111

112 +* js/convert-typofix (2020-02-11) 1 commit

113 + (merged to 'next' on 2020-02-11 at 32ccd29dc5)

114 + + convert: fix typo

115

275 -* js/ci-squelch-doc-warning (2020-02-10) 1 commit

276 - - ci: ignore rubygems warning in the "Documentation" job

116 + Typofix.

117

278 - Squelch unhelpful warning message during documentation build.

118

280 - Will merge to 'next'.

119 +* jt/t5616-robustify (2020-01-27) 1 commit

120 + (merged to 'next' on 2020-02-05 at cdb02e8849)

121 + + t5616: make robust to delta base change

122

123 + Futureproofing a test not to depend on the current implementation

124 + detail.

125

283 -* mr/bisect-in-c-1 (2020-02-10) 13 commits

284 - - bisect: libify `bisect_next_all`

285 - - bisect: libify `handle_bad_merge_base` and its dependents

286 - - bisect: libify `check_good_are_ancestors_of_bad` and its dependents

287 - - bisect: libify `check_merge_bases` and its dependents

288 - - bisect: libify `bisect_checkout`

289 - - bisect: libify `exit_if_skipped_commits` to `error_if_skipped*` and its dependents

290 - - bisect--helper: return error codes from `cmd_bisect__helper()`

291 - - bisect: add enum to represent bisect returning codes

292 - - bisect--helper: introduce new `decide_next()` function

293 - - run-command: make `exists_in_PATH()` non-static

294 - - bisect: use the standard 'if (!var)' way to check for 0

295 - - bisect--helper: change `retval` to `res`

296 - - bisect--helper: convert `vocab_*` char pointers to char arrays

126

298 - Underlying machinery of "git bisect--helper" is being refactored

299 - into pieces that are more easily reused.

127 +* km/submodule-doc-use-sm-path (2020-01-31) 1 commit

128 + (merged to 'next' on 2020-02-05 at a73aa350b5)

129 + + submodule foreach: replace $path with $sm_path in example

130

131 + Docfix.

132

302 -* rs/parse-options-concat-dup (2020-02-10) 4 commits

303 - - parse-options: simplify parse_options_dup()

304 - - parse-options: const parse_options_concat() parameters

305 - - parse-options: factor out parse_options_count()

306 - - parse-options: use COPY_ARRAY in parse_options_concat()

133

308 - Code clean-up.

134 +* ma/diff-doc-clarify-regexp-example (2020-02-09) 1 commit

135 + (merged to 'next' on 2020-02-11 at 202d447e78)

136 + + diff-options.txt: avoid "regex" overload in example

137 +

138 + Doc clarification.

139 +

140 +

141 +* ma/filter-branch-doc-caret (2020-02-04) 1 commit

142 + (merged to 'next' on 2020-02-05 at 4d5dbd1dba)

143 + + git-filter-branch.txt: wrap "maths" notation in backticks

144 +

145 + Doc mark-up updates.

146 +

147 +

148 +* ms/doc-bundle-format (2020-02-07) 1 commit

149 + (merged to 'next' on 2020-02-11 at ca03a36134)

150 + + doc: describe Git bundle format

151 +

152 + Technical details of the bundle format has been documented.

153 + I think this is in a good enough shape.

154 +

155 +

156 +* pb/do-not-recurse-grep-no-index (2020-01-30) 1 commit

157 + (merged to 'next' on 2020-02-05 at 227bbe1e05)

158 + + grep: ignore --recurse-submodules if --no-index is given

159 +

160 + "git grep --no-index" should not get affected by the contents of

161 + the .gitmodules file but when "--recurse-submodules" is given or

162 + the "submodule.recurse" variable is set, it did. Now these

163 + settings are ignored in the "--no-index" mode.

164 +

165 +--------------------------------------------------

166 +[New Topics]

167 +

168 +* ds/sparse-add (2020-02-11) 4 commits

169 + - sparse-checkout: work with Windows paths

170 + - sparse-checkout: create 'add' subcommand

171 + - sparse-checkout: extract pattern update from 'set' subcommand

172 + - sparse-checkout: extract add_patterns_from_input()

173 + (this branch uses ds/sparse-checkout-harden.)

174 +

175 + "git sparse-checkout" learned a new "add" subcommand.

176

177 Will merge to 'next'.

178

179

313 -* rs/strbuf-insertstr (2020-02-10) 2 commits

314 - - mailinfo: don't insert header prefix for handle_content_type()

315 - - strbuf: add and use strbuf_insertstr()

180 +* jk/mailinfo-cleanup (2020-02-11) 4 commits

181 + (merged to 'next' on 2020-02-11 at ab9890fbcb)

182 + + mailinfo: factor out some repeated header handling

183 + + mailinfo: be more liberal with header whitespace

184 + + mailinfo: simplify parsing of header values

185 + + mailinfo: treat header values as C strings

186 + (this branch uses rs/strbuf-insertstr.)

187

188 Code clean-up.

189

319 - Will merge to 'next'.

190 + Will merge to 'master'.

191

321 ---------------------------------------------------

322 -[Stalled]

192

324 -* jc/format-patch-delay-message-id (2019-04-05) 1 commit

325 - - format-patch: move message-id and related headers to the end

193 +* js/test-write-junit-xml-fix (2020-02-12) 1 commit

194 + - tests: fix --write-junit-xml with subshells

195

327 - The location "git format-patch --thread" adds the Message-Id:

328 - header in the series of header fields has been moved down, which

329 - may help working around a suspected bug in GMail MSA, reported at

330 - <CAHk-=whP1stFZNAaJiMi5eZ9rj0MRt20Y_yHVczZPH+O01d+sA@mail.gmail.com>

196 + Testfix.

197

332 - Waiting for feedback to see if it truly helps.

333 - Needs tests.

198 + Will merge to 'next'.

199

200 +--------------------------------------------------

201 +[Stalled]

202

203 * pw/advise-rebase-skip (2019-12-06) 9 commits

204 - rebase -i: leave CHERRY_PICK_HEAD when there are conflicts

@@ -352,101 +219,167 @@ of the repositories listed at

219 The tip two are still RFC.

220

221

355 -* es/pathspec-f-f-grep (2020-01-13) 1 commit

356 - . grep: support the --pathspec-from-file option

222 +* vn/reset-deleted-ita (2019-07-26) 1 commit

223 + - reset: unstage empty deleted ita files

224

358 - "git grep" learned the "--pathspec-from-file" command line

359 - option.

225 + "git reset HEAD [<pathspec>]" did not reset an empty file that was

226 + added with the intent-to-add bit.

227

361 - Getting tired of waiting for review responses. Will discard.

362 - cf. <20191204203911.237056-1-emilyshaffer@google.com>

228 + Expecting a reroll.

229 +

230 +

231 +* mk/use-size-t-in-zlib (2018-10-15) 1 commit

232 + - zlib.c: use size_t for size

233 +

234 + The wrapper to call into zlib followed our long tradition to use

235 + "unsigned long" for sizes of regions in memory, which have been

236 + updated to use "size_t".

237

238 +--------------------------------------------------

239 +[Cooking]

240

365 -* at/rebase-fork-point-regression-fix (2019-12-09) 1 commit

366 - - rebase: fix --fork-point with short refname

241 +* at/rebase-fork-point-regression-fix (2020-02-11) 1 commit

242 + - rebase: --fork-point regression fix

243

244 The "--fork-point" mode of "git rebase" regressed when the command

245 was rewritten in C back in 2.20 era, which has been corrected.

246

247 Was waiting for discussion to settle.

248 cf. <CAPig+cQ-3Ds41hr91fRo_GvuFMTP7zNVJtaSqi-Yccq4Pk-8Qg@mail.gmail.com>

249 + cf. <xmqq36dtwcvw.fsf@gitster-ct.c.googlers.com>

250

251

375 -* ma/config-bool-valex (2019-11-14) 8 commits

376 - - builtin/config: die if "value_regex" doesn't canonicalize as boolean

377 - - builtin/config: warn if "value_regex" doesn't canonicalize as boolean

378 - - builtin/config: canonicalize "value_regex" with `--type=bool-or-int`

379 - - builtin/config: canonicalize "value_regex" with `--type=bool`

380 - - builtin/config: collect "value_regexp" data in a struct

381 - - builtin/config: extract `handle_value_regex()` from `get_value()`

382 - - t1300: modernize part of script

383 - - config: make `git_parse_maybe_bool_text()` public

252 +* hn/reftable (2020-02-10) 5 commits

253 + - Reftable support for git-core

254 + - Add reftable library

255 + - refs: document how ref_iterator_advance_fn should handle symrefs

256 + - create .git/refs in files-backend.c

257 + - refs.h: clarify reflog iteration order

258

385 - "git config" can be told to affect the existing entries that

386 - "match" the given value via its value_regex argument. It learned

387 - to normalize the value set in the configuration and the value given

388 - from the command line before computing they "match", e.g. "true" in

389 - the configuration file can now match with "yes" given from the

390 - command line.

259 + A new refs backend "reftable" to replace the traditional

260 + combination of packed-refs files and one-file-per-ref loose refs

261 + has been implemented and integrated for improved performance and

262 + atomicity.

263

392 - Retracted for now.

393 - cf. <CAN0heSpxhoAqSvN_TJHPntSagx3oA4s7iUebHcsYML8NSYwfUA@mail.gmail.com>

264 + At v5.

265

266

396 -* ds/fsmonitor-testing (2019-12-09) 8 commits

397 - - test-lib: clear watchman watches at test completion

398 - - t7519: disable external GIT_TEST_FSMONITOR variable

399 - - t7063: disable fsmonitor with status cache

400 - - tests: disable fsmonitor in submodule tests

401 - - t3030-merge-recursive.sh: disable fsmonitor when tweaking GIT_WORK_TREE

402 - - t1301-shared-repo.sh: disable FSMONITOR

403 - - fsmonitor: do not output to stderr for tests

404 - - fsmonitor: disable in a bare repo

267 +* es/bugreport (2020-02-12) 16 commits

268 + - SQUASH??? t0091 style and i18n fixes

269 + - bugreport: summarize contents of alternates file

270 + - bugreport: list contents of $OBJDIR/info

271 + - bugreport: add packed object summary

272 + - bugreport: count loose objects

273 + - bugreport: collect list of populated hooks

274 + - bugreport: add config values from safelist

275 + - bugreport: generate config safelist based on docs

276 + - bugreport: include user interactive shell

277 + - bugreport: add git-remote-https version

278 + - bugreport: add compiler info

279 + - bugreport: add uname info

280 + - bugreport: gather git version and build info

281 + - bugreport: add tool to generate debugging info

282 + - help: add shell-path to --build-options

283 + - help: move list_config_help to builtin/help

284 + (this branch uses mr/show-config-scope; is tangled with bw/remote-rename-update-config.)

285

406 - Updates around testing fsmoitor integration.

407 - cf. <pull.466.v2.git.1575907804.gitgitgadget@gmail.com>

286 + The "bugreport" tool.

287

288 + at v6. Breaks gettext-poison builds

289

410 -* vn/reset-deleted-ita (2019-07-26) 1 commit

411 - - reset: unstage empty deleted ita files

290

413 - "git reset HEAD [<pathspec>]" did not reset an empty file that was

414 - added with the intent-to-add bit.

291 +* gs/commit-graph-path-filter (2020-02-12) 12 commits

292 + - (bytesex breakage band-aid)

293 + - commit-graph: add GIT_TEST_COMMIT_GRAPH_CHANGED_PATHS test flag

294 + - revision.c: use Bloom filters to speed up path based revision walks

295 + - commit-graph: add --changed-paths option to write subcommand

296 + - commit-graph: reuse existing Bloom filters during write.

297 + - commit-graph: write Bloom filters to commit graph file

298 + - commit-graph: examine commits by generation number

299 + - commit-graph: examine changed-path objects in pack order

300 + - commit-graph: compute Bloom filters for changed paths

301 + - diff: halt tree-diff early after max_changes

302 + - bloom: core Bloom filter implementation for changed paths

303 + - commit-graph: use MAX_NUM_CHUNKS

304

416 - Expecting a reroll.

305 + Introduce an extension to the commit-graph to make it efficient to

306 + check for the paths that were modified at each commit using Bloom

307 + filters.

308

309 + Will be rerolled with bytesex fixes squashed in.

310 + Breakage due to byte-order dependency reported.

311

419 -* jn/unknown-index-extensions (2018-11-21) 2 commits

420 - . index: offer advice for unknown index extensions

421 - . index: do not warn about unrecognized extensions

312

423 - A bit too alarming warning given when unknown index extensions

424 - exist is getting revamped.

313 +* bc/hash-independent-tests-part-8 (2020-02-07) 21 commits

314 + (merged to 'next' on 2020-02-11 at 127748e61a)

315 + + t6024: update for SHA-256

316 + + t6006: make hash size independent

317 + + t6000: abstract away SHA-1-specific constants

318 + + t5703: make test work with SHA-256

319 + + t5607: make hash size independent

320 + + t5318: update for SHA-256

321 + + t5515: make test hash independent

322 + + t5321: make test hash independent

323 + + t5313: make test hash independent

324 + + t5309: make test hash independent

325 + + t5302: make hash size independent

326 + + t4060: make test work with SHA-256

327 + + t4211: add test cases for SHA-256

328 + + t4211: move SHA-1-specific test cases into a directory

329 + + t4013: make test hash independent

330 + + t3311: make test work with SHA-256

331 + + t3310: make test work with SHA-256

332 + + t3309: make test work with SHA-256

333 + + t3308: make test work with SHA-256

334 + + t3206: make hash size independent

335 + + t/lib-pack: support SHA-256

336

426 - Getting tired of waiting for a reroll. Will discard.

337 + Preparation for SHA-256 migration continues.

338

339 + Will merge to 'master'.

340

429 -* js/protocol-advertise-multi (2018-12-28) 1 commit

430 - . protocol: advertise multiple supported versions

341

432 - The transport layer has been updated so that the protocol version

433 - used can be negotiated between the parties, by the initiator

434 - listing the protocol versions it is willing to talk, and the other

435 - side choosing from one of them.

342 +* mr/bisect-in-c-1 (2020-02-10) 13 commits

343 + - bisect: libify `bisect_next_all`

344 + - bisect: libify `handle_bad_merge_base` and its dependents

345 + - bisect: libify `check_good_are_ancestors_of_bad` and its dependents

346 + - bisect: libify `check_merge_bases` and its dependents

347 + - bisect: libify `bisect_checkout`

348 + - bisect: libify `exit_if_skipped_commits` to `error_if_skipped*` and its dependents

349 + - bisect--helper: return error codes from `cmd_bisect__helper()`

350 + - bisect: add enum to represent bisect returning codes

351 + - bisect--helper: introduce new `decide_next()` function

352 + - run-command: make `exists_in_PATH()` non-static

353 + - bisect: use the standard 'if (!var)' way to check for 0

354 + - bisect--helper: change `retval` to `res`

355 + - bisect--helper: convert `vocab_*` char pointers to char arrays

356

437 - Getting tired of waiting for a reroll. Will discard.

438 - cf. <CANq=j3u-zdb_FvNJGPCmygNMScseav63GhVvBX3NcVS4f7TejA@mail.gmail.com>

357 + Underlying machinery of "git bisect--helper" is being refactored

358 + into pieces that are more easily reused.

359

360

441 -* mk/use-size-t-in-zlib (2018-10-15) 1 commit

442 - - zlib.c: use size_t for size

361 +* rs/parse-options-concat-dup (2020-02-10) 4 commits

362 + (merged to 'next' on 2020-02-11 at a1685687ab)

363 + + parse-options: simplify parse_options_dup()

364 + + parse-options: const parse_options_concat() parameters

365 + + parse-options: factor out parse_options_count()

366 + + parse-options: use COPY_ARRAY in parse_options_concat()

367

444 - The wrapper to call into zlib followed our long tradition to use

445 - "unsigned long" for sizes of regions in memory, which have been

446 - updated to use "size_t".

368 + Code clean-up.

369 +

370 + Will merge to 'master'.

371 +

372 +

373 +* rs/strbuf-insertstr (2020-02-10) 2 commits

374 + (merged to 'next' on 2020-02-11 at 5c89aa002f)

375 + + mailinfo: don't insert header prefix for handle_content_type()

376 + + strbuf: add and use strbuf_insertstr()

377 + (this branch is used by jk/mailinfo-cleanup.)

378 +

379 + Code clean-up.

380 +

381 + Will merge to 'master'.

382

448 ---------------------------------------------------

449 -[Cooking]

383

384 * bw/remote-rename-update-config (2020-02-10) 6 commits

385 - remote rename/remove: gently handle remote.pushDefault config

@@ -461,10 +394,10 @@ of the repositories listed at

394 (e.g. branch.<name>.remote) whose value used to be X to Y.

395 branch.<name>.pushRemote is now also updated.

396

464 - Waiting for mr/show-config-scope to stabilize.

397 + Will merge to 'next'?

398

399

467 -* es/bright-colors (2020-01-23) 3 commits

400 +* es/bright-colors (2020-02-11) 3 commits

401 - color.c: alias RGB colors 8-15 to aixterm colors

402 - color.c: support bright aixterm colors

403 - color.c: refactor color_output arguments

@@ -472,9 +405,7 @@ of the repositories listed at

405 The basic 7 colors learned the brighter counterparts

406 (e.g. "brightred").

407

475 - Expecting a reroll.

476 - cf. <xmqqmuagbsyw.fsf@gitster-ct.c.googlers.com>

477 - cf. <xmqqo8ut3hsu.fsf@gitster-ct.c.googlers.com>

408 + Will merge to 'next'.

409

410

411 * pk/status-of-uncloned-submodule (2020-02-04) 3 commits

@@ -518,6 +449,7 @@ of the repositories listed at

449 + sparse-checkout: create leading directories

450 + t1091: improve here-docs

451 + t1091: use check_files to reduce boilerplate

452 + (this branch is used by ds/sparse-add.)

453

454 Some rough edges in the sparse-checkout feature, especially around

455 the cone mode, have been cleaned up.

@@ -536,28 +468,31 @@ of the repositories listed at

468

469

470 * mr/show-config-scope (2020-02-10) 10 commits

539 - - config: add '--show-scope' to print the scope of a config value

540 - - submodule-config: add subomdule config scope

541 - - config: teach git_config_source to remember its scope

542 - - config: preserve scope in do_git_config_sequence

543 - - config: clarify meaning of command line scoping

544 - - config: split repo scope to local and worktree

545 - - config: make scope_name non-static and rename it

546 - - t1300: create custom config file without special characters

547 - - t1300: fix over-indented HERE-DOCs

548 - - config: fix typo in variable name

471 + (merged to 'next' on 2020-02-11 at 904bca086e)

472 + + config: add '--show-scope' to print the scope of a config value

473 + + submodule-config: add subomdule config scope

474 + + config: teach git_config_source to remember its scope

475 + + config: preserve scope in do_git_config_sequence

476 + + config: clarify meaning of command line scoping

477 + + config: split repo scope to local and worktree

478 + + config: make scope_name non-static and rename it

479 + + t1300: create custom config file without special characters

480 + + t1300: fix over-indented HERE-DOCs

481 + + config: fix typo in variable name

482 (this branch is used by bw/remote-rename-update-config and es/bugreport.)

483

484 "git config" learned to show in which "scope", in addition to in

485 which file, each config setting comes from.

486

554 - Will merge to 'next'.

487 + Will merge to 'master'.

488

489

557 -* bk/p4-pre-edit-changelist (2020-02-06) 5 commits

490 +* bk/p4-pre-edit-changelist (2020-02-11) 7 commits

491 + - git-p4: add RCS keyword status message

492 - git-p4: add p4 submit hooks

493 - git-p4: restructure code in submit

494 - git-p4: add --no-verify option

495 + - git-p4: add p4-pre-submit exit text

496 - git-p4: create new function run_git_hook

497 - git-p4: rewrite prompt to be Windows compatible

498

@@ -565,16 +500,6 @@ of the repositories listed at

500 bypass them (and the existing "p4-pre-submit" hook).

501

502

568 -* dt/submodule-rm-with-stale-cache (2020-01-28) 1 commit

569 - (merged to 'next' on 2020-02-05 at 701ef6c41b)

570 - + git rm submodule: succeed if .gitmodules index stat info is zero

571 -

572 - Running "git rm" on a submodule failed unnecessarily when

573 - .gitmodules is only cache-dirty, which has been corrected.

574 -

575 - Will merge to 'master'.

576 -

577 -

503 * en/fill-directory-exponential (2020-01-31) 6 commits

504 - t7063: blindly accept diffs

505 - dir: replace exponential algorithm with a linear one

@@ -592,36 +517,6 @@ of the repositories listed at

517 cf. <CABPp-BF0SO92wxY-dM14Vxv6BGA5=T=CicdwbiaZ7c9w6-j2jQ@mail.gmail.com>

518

519

595 -* hw/doc-git-dir (2020-01-30) 1 commit

596 - (merged to 'next' on 2020-02-05 at 839ae05c00)

597 - + git: update documentation for --git-dir

598 -

599 - One effect of specifying where the GIT_DIR is (either with the

600 - environment variable, or with the "git --git-dir=<where> cmd"

601 - option) is to disable the repository discovery. This has been

602 - placed a bit more stress in the documentation, as new users often

603 - get confused.

604 -

605 - Will merge to 'master'.

606 -

607 -

608 -* jb/parse-options-message-fix (2020-02-05) 1 commit

609 - - parse-options: lose an unnecessary space in an error message

610 -

611 - Error message fix.

612 -

613 - Will merge to 'next'.

614 -

615 -

616 -* jc/skip-prefix (2020-01-31) 1 commit

617 - (merged to 'next' on 2020-02-05 at f6893a42d0)

618 - + C: use skip_prefix() to avoid hardcoded string length

619 -

620 - Code simplification.

621 -

622 - Will merge to 'master'.

623 -

624 -

520 * jh/notes-fanout-fix (2020-02-04) 2 commits

521 (merged to 'next' on 2020-02-07 at a4ec94aee0)

522 + notes.c: fix off-by-one error when decreasing notes fanout

@@ -645,19 +540,6 @@ of the repositories listed at

540 Will merge to 'master'.

541

542

648 -* jk/clang-sanitizer-fixes (2020-01-28) 5 commits

649 - (merged to 'next' on 2020-02-05 at 8ef84c2d96)

650 - + obstack: avoid computing offsets from NULL pointer

651 - + xdiff: avoid computing non-zero offset from NULL pointer

652 - + avoid computing zero offsets from NULL pointer

653 - + merge-recursive: use subtraction to flip stage

654 - + merge-recursive: silence -Wxor-used-as-pow warning

655 -

656 - C pedantry ;-) fix.

657 -

658 - Will merge to 'master'.

659 -

660 -

543 * jk/diff-honor-wserrhighlight-in-plumbing (2020-01-31) 1 commit

544 (merged to 'next' on 2020-02-07 at acb077fdcb)

545 + diff: move diff.wsErrorHighlight to "basic" config

@@ -670,17 +552,6 @@ of the repositories listed at

552 Will merge to 'master'.

553

554

673 -* jk/escaped-wildcard-dwim (2020-01-27) 1 commit

674 - (merged to 'next' on 2020-02-05 at ebee5e92a7)

675 - + verify_filename(): handle backslashes in "wildcards are pathspecs" rule

676 -

677 - Disambiguation logic to tell revisions and pathspec apart has been

678 - tweaked so that backslash-escaped glob special characters do not

679 - count in the "wildcards are pathspec" rule.

680 -

681 - Will merge to 'master'.

682 -

683 -

555 * jk/get-oid-error-message-i18n (2020-01-30) 3 commits

556 (merged to 'next' on 2020-02-07 at 6bfe9235e5)

557 + sha1-name: mark get_oid() error messages for translation

@@ -706,15 +577,6 @@ of the repositories listed at

577 Will merge to 'master'.

578

579

709 -* jk/push-default-doc (2020-01-29) 1 commit

710 - (merged to 'next' on 2020-02-05 at 33167357a1)

711 - + doc: drop "explicitly given" from push.default description

712 -

713 - Doc update.

714 -

715 - Will merge to 'master'.

716 -

717 -

580 * jt/connectivity-check-optim-in-partial-clone (2020-01-30) 2 commits

581 (merged to 'next' on 2020-02-07 at ca5102fa58)

582 + fetch: forgo full connectivity check if --filter

@@ -726,25 +588,6 @@ of the repositories listed at

588 Will merge to 'master'.

589

590

729 -* jt/t5616-robustify (2020-01-27) 1 commit

730 - (merged to 'next' on 2020-02-05 at cdb02e8849)

731 - + t5616: make robust to delta base change

732 -

733 - Futureproofing a test not to depend on the current implementation

734 - detail.

735 -

736 - Will merge to 'master'.

737 -

738 -

739 -* km/submodule-doc-use-sm-path (2020-01-31) 1 commit

740 - (merged to 'next' on 2020-02-05 at a73aa350b5)

741 - + submodule foreach: replace $path with $sm_path in example

742 -

743 - Docfix.

744 -

745 - Will merge to 'master'.

746 -

747 -

591 * ld/p4-cleanup-processes (2020-01-30) 7 commits

592 (merged to 'next' on 2020-02-07 at e8c573f01c)

593 + git-p4: avoid leak of file handle when cloning

@@ -760,24 +603,6 @@ of the repositories listed at

603 Will merge to 'master'.

604

605

763 -* ma/filter-branch-doc-caret (2020-02-04) 1 commit

764 - (merged to 'next' on 2020-02-05 at 4d5dbd1dba)

765 - + git-filter-branch.txt: wrap "maths" notation in backticks

766 -

767 - Doc mark-up updates.

768 -

769 - Will merge to 'master'.

770 -

771 -

772 -* ms/doc-bundle-format (2020-02-07) 1 commit

773 - - doc: describe Git bundle format

774 -

775 - Technical details of the bundle format has been documented.

776 -

777 - Will merge to 'next'.

778 - I think this is in a good enough shape.

779 -

780 -

606 * mt/use-passed-repo-more-in-funcs (2020-01-31) 7 commits

607 (merged to 'next' on 2020-02-07 at 536709ab4b)

608 + sha1-file: allow check_object_signature() to handle any repo

@@ -795,35 +620,24 @@ of the repositories listed at

620 Will merge to 'master'.

621

622

798 -* pb/do-not-recurse-grep-no-index (2020-01-30) 1 commit

799 - (merged to 'next' on 2020-02-05 at 227bbe1e05)

800 - + grep: ignore --recurse-submodules if --no-index is given

801 -

802 - "git grep --no-index" should not get affected by the contents of

803 - the .gitmodules file but when "--recurse-submodules" is given or

804 - the "submodule.recurse" variable is set, it did. Now these

805 - settings are ignored in the "--no-index" mode.

806 -

807 - Will merge to 'master'.

808 -

809 -

623 * rs/name-rev-memsave (2020-02-05) 11 commits

811 - - name-rev: sort tip names before applying

812 - - name-rev: release unused name strings

813 - - name-rev: generate name strings only if they are better

814 - - name-rev: pre-size buffer in get_parent_name()

815 - - name-rev: factor out get_parent_name()

816 - - name-rev: put struct rev_name into commit slab

817 - - name-rev: don't _peek() in create_or_update_name()

818 - - name-rev: don't leak path copy in name_ref()

819 - - name-rev: respect const qualifier

820 - - name-rev: remove unused typedef

821 - - name-rev: rewrite create_or_update_name()

624 + (merged to 'next' on 2020-02-11 at 17616cb9b1)

625 + + name-rev: sort tip names before applying

626 + + name-rev: release unused name strings

627 + + name-rev: generate name strings only if they are better

628 + + name-rev: pre-size buffer in get_parent_name()

629 + + name-rev: factor out get_parent_name()

630 + + name-rev: put struct rev_name into commit slab

631 + + name-rev: don't _peek() in create_or_update_name()

632 + + name-rev: don't leak path copy in name_ref()

633 + + name-rev: respect const qualifier

634 + + name-rev: remove unused typedef

635 + + name-rev: rewrite create_or_update_name()

636

637 Memory footprint and performance of "git name-rev" has been

638 improved.

639

826 - Will merge to 'next'.

640 + Will merge to 'master'.

641

642

643 * tb/commit-graph-object-dir (2020-02-04) 5 commits

@@ -842,7 +656,7 @@ of the repositories listed at

656 Will merge to 'master'.

657

658

845 -* tb/commit-graph-split-merge (2020-02-05) 3 commits

659 +* tb/commit-graph-split-merge (2020-02-12) 3 commits

660 - builtin/commit-graph.c: support '--input=none'

661 - builtin/commit-graph.c: introduce '--input=<source>'

662 - builtin/commit-graph.c: support '--split[=<strategy>]'

@@ -853,7 +667,7 @@ of the repositories listed at

667 or a single new incremental graph is created.

668

669 Waiting for the discussion to settle.

856 - cf. <20200206025130.GA22748@syl.local>

670 + cf. <20200212205028.GE4364@syl.local>

671

672

673 * ag/edit-todo-drop-check (2020-01-28) 2 commits

@@ -946,16 +760,6 @@ of the repositories listed at

760 Will merge to 'master'.

761

762

949 -* jn/pretend-object-doc (2020-01-06) 1 commit

950 - (merged to 'next' on 2020-02-05 at 1d52a5c72d)

951 - + sha1-file: document how to use pretend_object_file

952 -

953 - Warn programmers about pretend_object_file() that allows the code

954 - to tentatively use in-core objects.

955 -

956 - Will merge to 'master'.

957 -

958 -

763 * dl/merge-autostash (2020-01-13) 17 commits

764 - pull: pass --autostash to merge

765 - t5520: make test_pull_autostash() accept expect_parent_num

@@ -1090,3 +894,85 @@ of the repositories listed at

894 to generate its result has been improved.

895

896 Will merge to 'master'.

897 +

898 +--------------------------------------------------

899 +[Discarded]

900 +

901 +* jc/format-patch-delay-message-id (2019-04-05) 1 commit

902 + . format-patch: move message-id and related headers to the end

903 +

904 + The location "git format-patch --thread" adds the Message-Id:

905 + header in the series of header fields has been moved down, which

906 + may help working around a suspected bug in GMail MSA, reported at

907 + <CAHk-=whP1stFZNAaJiMi5eZ9rj0MRt20Y_yHVczZPH+O01d+sA@mail.gmail.com>

908 +

909 + Was waiting for feedback to see if it truly helps.

910 +

911 +

912 +* es/pathspec-f-f-grep (2020-01-13) 1 commit

913 + . grep: support the --pathspec-from-file option

914 +

915 + "git grep" learned the "--pathspec-from-file" command line

916 + option.

917 +

918 + Getting tired of waiting for review responses.

919 + cf. <20191204203911.237056-1-emilyshaffer@google.com>

920 +

921 +

922 +* ma/config-bool-valex (2019-11-14) 8 commits

923 + . builtin/config: die if "value_regex" doesn't canonicalize as boolean

924 + . builtin/config: warn if "value_regex" doesn't canonicalize as boolean

925 + . builtin/config: canonicalize "value_regex" with `--type=bool-or-int`

926 + . builtin/config: canonicalize "value_regex" with `--type=bool`

927 + . builtin/config: collect "value_regexp" data in a struct

928 + . builtin/config: extract `handle_value_regex()` from `get_value()`

929 + . t1300: modernize part of script

930 + . config: make `git_parse_maybe_bool_text()` public

931 +

932 + "git config" can be told to affect the existing entries that

933 + "match" the given value via its value_regex argument. It learned

934 + to normalize the value set in the configuration and the value given

935 + from the command line before computing they "match", e.g. "true" in

936 + the configuration file can now match with "yes" given from the

937 + command line.

938 +

939 + Retracted for now.

940 + cf. <CAN0heSpxhoAqSvN_TJHPntSagx3oA4s7iUebHcsYML8NSYwfUA@mail.gmail.com>

941 +

942 +

943 +* ds/fsmonitor-testing (2019-12-09) 8 commits

944 + . test-lib: clear watchman watches at test completion

945 + . t7519: disable external GIT_TEST_FSMONITOR variable

946 + . t7063: disable fsmonitor with status cache

947 + . tests: disable fsmonitor in submodule tests

948 + . t3030-merge-recursive.sh: disable fsmonitor when tweaking GIT_WORK_TREE

949 + . t1301-shared-repo.sh: disable FSMONITOR

950 + . fsmonitor: do not output to stderr for tests

951 + . fsmonitor: disable in a bare repo

952 +

953 + Updates around testing fsmoitor integration.

954 +

955 + Retracted for now.

956 + cf. <a780fe83-c82c-c8fa-3f8d-5db9a7886875@gmail.com>

957 +

958 +

959 +* jn/unknown-index-extensions (2018-11-21) 2 commits

960 + . index: offer advice for unknown index extensions

961 + . index: do not warn about unrecognized extensions

962 +

963 + A bit too alarming warning given when unknown index extensions

964 + exist is getting revamped.

965 +

966 + Getting tired of waiting for a reroll.

967 +

968 +

969 +* js/protocol-advertise-multi (2018-12-28) 1 commit

970 + . protocol: advertise multiple supported versions

971 +

972 + The transport layer has been updated so that the protocol version

973 + used can be negotiated between the parties, by the initiator

974 + listing the protocol versions it is willing to talk, and the other

975 + side choosing from one of them.

976 +

977 + Getting tired of waiting for a reroll.

978 + cf. <CANq=j3u-zdb_FvNJGPCmygNMScseav63GhVvBX3NcVS4f7TejA@mail.gmail.com>

What's cooking (2020/02 #03)