27fe017671 · 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 (Dec 2016, #03; Tue, 13)

4 -X-master-at: de2efebf7ce2b308ea77d8b06f971e935238cd2f

5 -X-next-at: 3074f9477c394cb6e81d163f738566535afd22ed

3 +Subject: What's cooking in git.git (Dec 2016, #04; Fri, 16)

4 +X-master-at: eff96d7e16a769dca9b0319ccf460a83f514676e

5 +X-next-at: f8bf8f2a7b2b08001fdf3e1c554eef9249e47c19

6

7 -What's cooking in git.git (Dec 2016, #03; Tue, 13)

7 +What's cooking in git.git (Dec 2016, #04; Fri, 16)

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

9

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

@@ -12,9 +12,6 @@ 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 -A few topics were merged to 'master' and are meant for 'maint' to be

16 -in the first maintenance release for 2.11.x series.

17 -

15 You can find the changes described here in the integration branches

16 of the repositories listed at

17

@@ -23,65 +20,254 @@ of the repositories listed at

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

21 [Graduated to "master"]

22

26 -* ew/svn-fixes (2016-12-12) 2 commits

27 - (merged to 'next' on 2016-12-12 at 91ed5b3cf3)

28 - + git-svn: document useLogAuthor and addAuthorFrom config keys

29 - + git-svn: allow "0" in SVN path components

23 +* ak/lazy-prereq-mktemp (2016-11-29) 1 commit

24 + (merged to 'next' on 2016-12-12 at f346d1f053)

25 + + t7610: clean up foo.XXXXXX tmpdir

26 +

27 + Test code clean-up.

28 +

29 +

30 +* bw/push-dry-run (2016-11-23) 2 commits

31 + (merged to 'next' on 2016-12-12 at bde7a0f9ae)

32 + + push: fix --dry-run to not push submodules

33 + + push: --dry-run updates submodules when --recurse-submodules=on-demand

34 + (this branch uses hv/submodule-not-yet-pushed-fix; is tangled with sb/push-make-submodule-check-the-default.)

35 +

36 + "git push --dry-run --recurse-submodule=on-demand" wasn't

37 + "--dry-run" in the submodules.

38 +

39 +

40 +* da/mergetool-trust-exit-code (2016-11-29) 2 commits

41 + (merged to 'next' on 2016-12-12 at 28ae202868)

42 + + mergetools/vimdiff: trust Vim's exit code

43 + + mergetool: honor mergetool.$tool.trustExitCode for built-in tools

44 +

45 + mergetool.<tool>.trustExitCode configuration variable did not apply

46 + to built-in tools, but now it does.

47 +

48 +

49 +* dt/empty-submodule-in-merge (2016-11-17) 1 commit

50 + (merged to 'next' on 2016-12-12 at 6de2350b2b)

51 + + submodules: allow empty working-tree dirs in merge/cherry-pick

52 +

53 + An empty directory in a working tree that can simply be nuked used

54 + to interfere while merging or cherry-picking a change to create a

55 + submodule directory there, which has been fixed..

56 +

57 +

58 +* hv/submodule-not-yet-pushed-fix (2016-11-16) 4 commits

59 + (merged to 'next' on 2016-12-05 at c9d729fca2)

60 + + submodule_needs_pushing(): explain the behaviour when we cannot answer

61 + + batch check whether submodule needs pushing into one call

62 + + serialize collection of refs that contain submodule changes

63 + + serialize collection of changed submodules

64 + (this branch is used by bw/push-dry-run and sb/push-make-submodule-check-the-default.)

65 +

66 + Originally merged to 'next' on 2016-11-21

67 +

68 + The code in "git push" to compute if any commit being pushed in the

69 + superproject binds a commit in a submodule that hasn't been pushed

70 + out was overly inefficient, making it unusable even for a small

71 + project that does not have any submodule but have a reasonable

72 + number of refs.

73 +

74 +

75 +* jk/rev-parse-symbolic-parents-fix (2016-11-16) 1 commit

76 + (merged to 'next' on 2016-12-12 at 6839c1ea28)

77 + + rev-parse: fix parent shorthands with --symbolic

78 +

79 + "git rev-parse --symbolic" failed with a more recent notation like

80 + "HEAD^-1" and "HEAD^!".

81 +

82 +

83 +* ld/p4-update-shelve (2016-12-05) 1 commit

84 + (merged to 'next' on 2016-12-12 at 22f6bec94c)

85 + + git-p4: support updating an existing shelved changelist

86 + (this branch uses vk/p4-submit-shelve.)

87 +

88 + Will merge to 'master'.

89 +

90 +

91 +* ls/p4-empty-file-on-lfs (2016-12-05) 1 commit

92 + (merged to 'next' on 2016-12-12 at 1fce8e037a)

93 + + git-p4: fix empty file processing for large file system backend GitLFS

94 +

95 + "git p4" LFS support was broken when LFS stores an empty blob.

96 +

97 +

98 +* ls/p4-retry-thrice (2016-12-05) 1 commit

99 + (merged to 'next' on 2016-12-12 at 9462e660a8)

100 + + git-p4: add config to retry p4 commands; retry 3 times by default

101 +

102 + Will merge to 'master'.

103 +

104 +

105 +* nd/qsort-in-merge-recursive (2016-11-28) 1 commit

106 + (merged to 'next' on 2016-12-12 at e9700f5b93)

107 + + merge-recursive.c: use string_list_sort instead of qsort

108 +

109 + Code simplification.

110

31 - Meant eventually for 'maint'.

111

112 +* nd/worktree-list-fixup (2016-11-28) 5 commits

113 + (merged to 'next' on 2016-12-12 at 1f46421a59)

114 + + worktree list: keep the list sorted

115 + + worktree.c: get_worktrees() takes a new flag argument

116 + + get_worktrees() must return main worktree as first item even on error

117 + + worktree: reorder an if statement

118 + + worktree.c: zero new 'struct worktree' on allocation

119 + (this branch is used by nd/worktree-move and sb/submodule-embed-gitdir.)

120 +

121 + The output from "git worktree list" was made in readdir() order,

122 + and was unstable.

123

34 -* js/mingw-isatty (2016-12-11) 1 commit

35 - (merged to 'next' on 2016-12-12 at 60c1da6676)

36 - + mingw: intercept isatty() to handle /dev/null as Git expects it

124

38 - We often decide if a session is interactive by checking if the

39 - standard I/O streams are connected to a TTY, but isatty() that

40 - comes with Windows incorrectly returned true if it is used on NUL

41 - (i.e. an equivalent to /dev/null). This has been fixed.

125 +* vk/p4-submit-shelve (2016-11-29) 1 commit

126 + (merged to 'next' on 2016-12-12 at 3fce6df117)

127 + + git-p4: allow submit to create shelved changelists.

128 + (this branch is used by ld/p4-update-shelve.)

129

43 - Meant eventually for 'maint'.

130 + Will merge to 'master'.

131

132 --------------------------------------------------

133 [New Topics]

134

48 -* bw/realpath-wo-chdir (2016-12-12) 4 commits

49 - - real_path: have callers use real_pathdup and strbuf_realpath

50 - - real_path: create real_pathdup

51 - - real_path: convert real_path_internal to strbuf_realpath

52 - - real_path: resolve symlinks by hand

135 +* bw/pathspec-cleanup (2016-12-14) 16 commits

136 + - pathspec: rename prefix_pathspec to init_pathspec_item

137 + - pathspec: small readability changes

138 + - pathspec: create strip submodule slash helpers

139 + - pathspec: create parse_element_magic helper

140 + - pathspec: create parse_long_magic function

141 + - pathspec: create parse_short_magic function

142 + - pathspec: factor global magic into its own function

143 + - pathspec: simpler logic to prefix original pathspec elements

144 + - pathspec: always show mnemonic and name in unsupported_magic

145 + - pathspec: remove unused variable from unsupported_magic

146 + - pathspec: copy and free owned memory

147 + - pathspec: remove the deprecated get_pathspec function

148 + - ls-tree: convert show_recursive to use the pathspec struct interface

149 + - dir: convert fill_directory to use the pathspec struct interface

150 + - dir: remove struct path_simplify

151 + - mv: remove use of deprecated 'get_pathspec()'

152 +

153 + Code clean-up in the pathspec API.

154 +

155 + Waiting for the (hopefully) final round of review before 'next'.

156 +

157 +

158 +* cp/merge-continue (2016-12-15) 4 commits

159 + - merge: mark usage error strings for translation

160 + - merge: ensure '--abort' option takes no arguments

161 + - completion: add --continue option for merge

162 + - merge: add '--continue' option as a synonym for 'git commit'

163 +

164 + "git merge --continue" has been added as a synonym to "git commit"

165 + to conclude a merge that has stopped due to conflicts.

166

54 - The implementation of "real_path()" was to go there with chdir(2)

55 - and call getcwd(3), but this obviously wouldn't be usable in a

56 - threaded environment. Rewrite it to manually resolve relative

57 - paths including symbolic links in path components.

167 + Will merge to 'next'.

168

59 - What's the donness of the topic? Is this ready for 'next'?

169

170 +* jk/make-tags-find-sources-tweak (2016-12-14) 4 commits

171 + (merged to 'next' on 2016-12-16 at 06d0b0fbfd)

172 + + Makefile: exclude contrib from FIND_SOURCE_FILES

173 + + Makefile: match shell scripts in FIND_SOURCE_FILES

174 + + Makefile: exclude test cruft from FIND_SOURCE_FILES

175 + + Makefile: reformat FIND_SOURCE_FILES

176

62 -* jk/quote-env-path-list-component (2016-12-13) 4 commits

63 - - t5547-push-quarantine: run the path separator test on Windows, too

64 - - tmp-objdir: quote paths we add to alternates

65 - - alternates: accept double-quoted paths

66 - - Merge branch 'jk/alt-odb-cleanup' into jk/quote-env-path-list-component

177 + Update the procedure to generate "tags" for developer support.

178

68 - A recent update to receive-pack to make it easier to drop garbage

69 - objects made it clear that GIT_ALTERNATE_OBJECT_DIRECTORIES cannot

70 - have a pathname with a colon in it (no surprise!), and this in turn

71 - made it impossible to push into a repository at such a path. This

72 - has been fixed by introducing a quoting mechanism used when

73 - appending such a path to the colon-separated list.

179 + Will merge to 'master'.

180 +

181 +

182 +* jk/parseopt-usage-msg-opt (2016-12-14) 1 commit

183 + - parse-options: print "fatal:" before usage_msg_opt()

184 +

185 + The function usage_msg_opt() has been updated to say "fatal:"

186 + before the custom message programs give, when they want to die

187 + with a message about wrong command line options followed by the

188 + standard usage string.

189

190 Will merge to 'next'.

191

192

78 -* js/normalize-path-copy-ceil (2016-12-13) 1 commit

79 - - normalize_path_copy(): fix pushing to //server/share/dir paths on Windows

193 +* ld/p4-compare-dir-vs-symlink (2016-12-14) 1 commit

194 + - git-p4: avoid crash adding symlinked directory

195

81 - A pathname that begins with "//" or "\\" on Windows is special but

82 - path normalization logic was unaware of it.

196 + "git p4" misbehaved when swapping a directory and a symbolic link.

197 +

198 + Will merge to 'next'.

199 +

200 +

201 +* js/prepare-sequencer-more (2016-12-14) 34 commits

202 + - sequencer (rebase -i): write out the final message

203 + - sequencer (rebase -i): write the progress into files

204 + - sequencer (rebase -i): show the progress

205 + - sequencer (rebase -i): suggest --edit-todo upon unknown command

206 + - sequencer (rebase -i): show only failed cherry-picks' output

207 + - sequencer (rebase -i): show only failed `git commit`'s output

208 + - run_command_opt(): optionally hide stderr when the command succeeds

209 + - sequencer (rebase -i): differentiate between comments and 'noop'

210 + - sequencer (rebase -i): implement the 'drop' command

211 + - sequencer (rebase -i): allow rescheduling commands

212 + - sequencer (rebase -i): respect strategy/strategy_opts settings

213 + - sequencer (rebase -i): respect the rebase.autostash setting

214 + - sequencer (rebase -i): run the post-rewrite hook, if needed

215 + - sequencer (rebase -i): record interrupted commits in rewritten, too

216 + - sequencer (rebase -i): copy commit notes at end

217 + - sequencer (rebase -i): set the reflog message consistently

218 + - sequencer (rebase -i): refactor setting the reflog message

219 + - sequencer (rebase -i): allow fast-forwarding for edit/reword

220 + - sequencer (rebase -i): implement the 'reword' command

221 + - sequencer (rebase -i): leave a patch upon error

222 + - sequencer (rebase -i): update refs after a successful rebase

223 + - sequencer (rebase -i): the todo can be empty when continuing

224 + - sequencer (rebase -i): skip some revert/cherry-pick specific code path

225 + - sequencer (rebase -i): remove CHERRY_PICK_HEAD when no longer needed

226 + - sequencer (rebase -i): allow continuing with staged changes

227 + - sequencer (rebase -i): write an author-script file

228 + - sequencer (rebase -i): implement the short commands

229 + - sequencer (rebase -i): add support for the 'fixup' and 'squash' commands

230 + - sequencer (rebase -i): write the 'done' file

231 + - sequencer (rebase -i): learn about the 'verbose' mode

232 + - sequencer (rebase -i): implement the 'exec' command

233 + - sequencer (rebase -i): implement the 'edit' command

234 + - sequencer (rebase -i): implement the 'noop' command

235 + - sequencer: support a new action: 'interactive rebase'

236 +

237 + The sequencer has further been extended in preparation to act as a

238 + back-end for "rebase -i".

239 +

240 + Waiting for review.

241 +

242 +

243 +* jk/index-pack-wo-repo-from-stdin (2016-12-16) 4 commits

244 + - index-pack: skip collision check when not in repository

245 + - t: use nongit() function where applicable

246 + - index-pack: complain when --stdin is used outside of a repo

247 + - t5000: extract nongit function to test-lib-functions.sh

248 +

249 + "git index-pack --stdin" needs an access to an existing repository,

250 + but "git index-pack file.pack" to generate an .idx file that

251 + corresponds to a packfile does not.

252 +

253 + Will merge to 'next'.

254 +

255 +

256 +* jk/readme-gmane-is-no-more (2016-12-15) 1 commit

257 + (merged to 'next' on 2016-12-16 at 44ad5c5205)

258 + + README: replace gmane link with public-inbox

259

84 - Will merge to 'next', but I'd appreciate a second set of eyes on this.

260 + Will merge to 'master'.

261 +

262 +

263 +* lt/shortlog-by-committer (2016-12-16) 2 commits

264 + - shortlog: test and document --committer option

265 + - shortlog: group by committer information

266 +

267 + "git shortlog" learned "--committer" option to group commits by

268 + committer, instead of author.

269 +

270 + Will merge to 'next'.

271

272 --------------------------------------------------

273 [Stalled]

@@ -129,7 +315,6 @@ of the repositories listed at

315 . copy.c: convert bb_(p)error_msg to error(_errno)

316 . copy.c: delete unused code in copy_file()

317 . copy.c: import copy_file() from busybox

132 - (this branch uses nd/worktree-list-fixup; is tangled with sb/submodule-embed-gitdir.)

318

319 "git worktree" learned move and remove subcommands.

320

@@ -223,6 +408,48 @@ of the repositories listed at

408 --------------------------------------------------

409 [Cooking]

410

411 +* bw/realpath-wo-chdir (2016-12-12) 4 commits

412 + - real_path: have callers use real_pathdup and strbuf_realpath

413 + - real_path: create real_pathdup

414 + - real_path: convert real_path_internal to strbuf_realpath

415 + - real_path: resolve symlinks by hand

416 + (this branch is used by bw/grep-recurse-submodules.)

417 +

418 + The implementation of "real_path()" was to go there with chdir(2)

419 + and call getcwd(3), but this obviously wouldn't be usable in a

420 + threaded environment. Rewrite it to manually resolve relative

421 + paths including symbolic links in path components.

422 +

423 + Will merge to 'next'.

424 +

425 +

426 +* jk/quote-env-path-list-component (2016-12-13) 4 commits

427 + (merged to 'next' on 2016-12-16 at d2cd6008b9)

428 + + t5547-push-quarantine: run the path separator test on Windows, too

429 + + tmp-objdir: quote paths we add to alternates

430 + + alternates: accept double-quoted paths

431 + + Merge branch 'jk/alt-odb-cleanup' into jk/quote-env-path-list-component

432 +

433 + A recent update to receive-pack to make it easier to drop garbage

434 + objects made it clear that GIT_ALTERNATE_OBJECT_DIRECTORIES cannot

435 + have a pathname with a colon in it (no surprise!), and this in turn

436 + made it impossible to push into a repository at such a path. This

437 + has been fixed by introducing a quoting mechanism used when

438 + appending such a path to the colon-separated list.

439 +

440 + Will merge to 'master'.

441 +

442 +

443 +* js/normalize-path-copy-ceil (2016-12-16) 1 commit

444 + (merged to 'next' on 2016-12-16 at 634ba4debc)

445 + + normalize_path_copy(): fix pushing to //server/share/dir on Windows

446 +

447 + A pathname that begins with "//" or "\\" on Windows is special but

448 + path normalization logic was unaware of it.

449 +

450 + Will merge to 'master'.

451 +

452 +

453 * jc/lock-report-on-error (2016-12-07) 3 commits

454 (merged to 'next' on 2016-12-13 at cb6c07ee92)

455 + lockfile: LOCK_REPORT_ON_ERROR

@@ -252,7 +479,9 @@ of the repositories listed at

479 Will merge to 'master'.

480

481

255 -* sb/sequencer-abort-safety (2016-12-09) 5 commits

482 +* sb/sequencer-abort-safety (2016-12-14) 6 commits

483 + (merged to 'next' on 2016-12-16 at ec71fb1217)

484 + + Revert "sequencer: remove useless get_dir() function"

485 (merged to 'next' on 2016-12-13 at 6107e43d65)

486 + sequencer: remove useless get_dir() function

487 + sequencer: make sequencer abort safer

@@ -324,26 +553,6 @@ of the repositories listed at

553 Will merge to 'master'.

554

555

327 -* ak/lazy-prereq-mktemp (2016-11-29) 1 commit

328 - (merged to 'next' on 2016-12-12 at f346d1f053)

329 - + t7610: clean up foo.XXXXXX tmpdir

330 -

331 - Test code clean-up.

332 -

333 - Will merge to 'master'.

334 -

335 -

336 -* da/mergetool-trust-exit-code (2016-11-29) 2 commits

337 - (merged to 'next' on 2016-12-12 at 28ae202868)

338 - + mergetools/vimdiff: trust Vim's exit code

339 - + mergetool: honor mergetool.$tool.trustExitCode for built-in tools

340 -

341 - mergetool.<tool>.trustExitCode configuration variable did not apply

342 - to built-in tools, but now it does.

343 -

344 - Will merge to 'master'.

345 -

346 -

556 * jb/diff-no-index-no-abbrev (2016-12-08) 1 commit

557 (merged to 'next' on 2016-12-12 at 959981ef50)

558 + diff: handle --no-abbrev in no-index case

@@ -353,14 +562,6 @@ of the repositories listed at

562 Will merge to 'master'.

563

564

356 -* vk/p4-submit-shelve (2016-11-29) 1 commit

357 - (merged to 'next' on 2016-12-12 at 3fce6df117)

358 - + git-p4: allow submit to create shelved changelists.

359 - (this branch is used by ld/p4-update-shelve.)

360 -

361 - Will merge to 'master'.

362 -

363 -

565 * jk/http-walker-limit-redirect-2.9 (2016-12-06) 5 commits

566 (merged to 'next' on 2016-12-12 at 3e4bcd7bca)

567 + http: treat http-alternates like redirects

@@ -368,7 +569,7 @@ of the repositories listed at

569 + remote-curl: rename shadowed options variable

570 + http: always update the base URL for redirects

571 + http: simplify update_url_from_redirect

371 - (this branch is used by jk/http-walker-limit-redirect.)

572 + (this branch is used by bw/transport-protocol-policy and jk/http-walker-limit-redirect.)

573

574 Transport with dumb http can be fooled into following foreign URLs

575 that the end user does not intend to, especially with the server

@@ -383,7 +584,7 @@ of the repositories listed at

584 (merged to 'next' on 2016-12-12 at 8b58025e3a)

585 + http-walker: complain about non-404 loose object errors

586 + Merge branch 'ew/http-walker' into jk/http-walker-limit-redirect

386 - (this branch uses jk/http-walker-limit-redirect-2.9.)

587 + (this branch is used by bw/transport-protocol-policy; uses jk/http-walker-limit-redirect-2.9.)

588

589 Update the error messages from the dumb-http client when it fails

590 to obtain loose objects; we used to give sensible error message

@@ -417,50 +618,27 @@ of the repositories listed at

618 Will merge to 'master'.

619

620

420 -* bb/unicode-9.0 (2016-12-13) 6 commits

421 - - unicode_width.h: update the width tables to Unicode 9.0

422 - - update_unicode.sh: remove the plane filter

423 - - update-unicode.sh: automatically download newer definition files

424 - - update_unicode.sh: pin the uniset repo to a known good commit

425 - - update_unicode.sh: remove an unnecessary subshell level

426 - - update_unicode.sh: move it into contrib/update-unicode

621 +* bb/unicode-9.0 (2016-12-14) 6 commits

622 + (merged to 'next' on 2016-12-16 at be2531431a)

623 + + unicode_width.h: update the width tables to Unicode 9.0

624 + + update_unicode.sh: remove the plane filter

625 + + update_unicode.sh: automatically download newer definition files

626 + + update_unicode.sh: pin the uniset repo to a known good commit

627 + + update_unicode.sh: remove an unnecessary subshell level

628 + + update_unicode.sh: move it into contrib/update-unicode

629

630 The character width table has been updated to match Unicode 9.0

631

430 - Will merge to 'next'.

431 -

432 -

433 -* ld/p4-update-shelve (2016-12-05) 1 commit

434 - (merged to 'next' on 2016-12-12 at 22f6bec94c)

435 - + git-p4: support updating an existing shelved changelist

436 - (this branch uses vk/p4-submit-shelve.)

437 -

632 Will merge to 'master'.

633

634

635 * ld/p4-worktree (2016-12-13) 1 commit

442 - - git-p4: support git worktrees

636 + (merged to 'next' on 2016-12-16 at 5210ab9973)

637 + + git-p4: support git worktrees

638

639 "git p4" didn't interact with the internal of .git directory

640 correctly in the modern "git-worktree"-enabled world.

641

447 - Will merge to 'next'.

448 - Rerolled. Looking good.

449 -

450 -

451 -* ls/p4-empty-file-on-lfs (2016-12-05) 1 commit

452 - (merged to 'next' on 2016-12-12 at 1fce8e037a)

453 - + git-p4: fix empty file processing for large file system backend GitLFS

454 -

455 - "git p4" LFS support was broken when LFS stores an empty blob.

456 -

457 - Will merge to 'master'.

458 -

459 -

460 -* ls/p4-retry-thrice (2016-12-05) 1 commit

461 - (merged to 'next' on 2016-12-12 at 9462e660a8)

462 - + git-p4: add config to retry p4 commands; retry 3 times by default

463 -

642 Will merge to 'master'.

643

644

@@ -580,62 +758,15 @@ of the repositories listed at

758 What's the doneness of this topic?

759

760

583 -* nd/qsort-in-merge-recursive (2016-11-28) 1 commit

584 - (merged to 'next' on 2016-12-12 at e9700f5b93)

585 - + merge-recursive.c: use string_list_sort instead of qsort

586 -

587 - Code simplification.

588 -

589 - Will merge to 'master'.

590 -

591 -

592 -* bw/push-dry-run (2016-11-23) 2 commits

593 - (merged to 'next' on 2016-12-12 at bde7a0f9ae)

594 - + push: fix --dry-run to not push submodules

595 - + push: --dry-run updates submodules when --recurse-submodules=on-demand

596 - (this branch uses hv/submodule-not-yet-pushed-fix; is tangled with sb/push-make-submodule-check-the-default.)

597 -

598 - "git push --dry-run --recurse-submodule=on-demand" wasn't

599 - "--dry-run" in the submodules.

600 -

601 - Will merge to 'master'.

602 -

603 -

761 * sb/push-make-submodule-check-the-default (2016-11-29) 2 commits

762 (merged to 'next' on 2016-12-12 at 1863e05af5)

763 + push: change submodule default to check when submodules exist

764 + submodule add: extend force flag to add existing repos

608 - (this branch uses hv/submodule-not-yet-pushed-fix; is tangled with bw/push-dry-run.)

765

766 Turn the default of "push.recurseSubmodules" to "check" when

767 submodules seem to be in use.

768

613 - Will merge to 'master'.

614 -

615 -

616 -* jk/rev-parse-symbolic-parents-fix (2016-11-16) 1 commit

617 - (merged to 'next' on 2016-12-12 at 6839c1ea28)

618 - + rev-parse: fix parent shorthands with --symbolic

619 -

620 - "git rev-parse --symbolic" failed with a more recent notation like

621 - "HEAD^-1" and "HEAD^!".

622 -

623 - Will merge to 'master'.

624 -

625 -

626 -* nd/worktree-list-fixup (2016-11-28) 5 commits

627 - (merged to 'next' on 2016-12-12 at 1f46421a59)

628 - + worktree list: keep the list sorted

629 - + worktree.c: get_worktrees() takes a new flag argument

630 - + get_worktrees() must return main worktree as first item even on error

631 - + worktree: reorder an if statement

632 - + worktree.c: zero new 'struct worktree' on allocation

633 - (this branch is used by nd/worktree-move and sb/submodule-embed-gitdir.)

634 -

635 - The output from "git worktree list" was made in readdir() order,

636 - and was unstable.

637 -

638 - Will merge to 'master'.

769 + Will cook in 'next'.

770

771

772 * jk/trailers-placeholder-in-pretty (2016-12-11) 2 commits

@@ -657,38 +788,28 @@ of the repositories listed at

788 - test-lib-functions.sh: teach test_commit -C <dir>

789 - submodule helper: support super prefix

790 - submodule: use absolute path for computing relative path connecting

660 - (this branch uses nd/worktree-list-fixup; is tangled with nd/worktree-move.)

791

792 A new submodule helper "git submodule embedgitdirs" to make it

793 easier to move embedded .git/ directory for submodules in a

794 superproject to .git/modules/ (and point the latter with the former

795 that is turned into a "gitdir:" file) has been added.

796

667 - Is this now pretty much done and ready for 'next'?

668 -

669 -

670 -* dt/empty-submodule-in-merge (2016-11-17) 1 commit

671 - (merged to 'next' on 2016-12-12 at 6de2350b2b)

672 - + submodules: allow empty working-tree dirs in merge/cherry-pick

673 -

674 - An empty directory in a working tree that can simply be nuked used

675 - to interfere while merging or cherry-picking a change to create a

676 - submodule directory there, which has been fixed..

677 -

678 - Will merge to 'master'.

797 + Will merge to 'next'.

798

799

681 -* bw/grep-recurse-submodules (2016-11-22) 6 commits

682 - . grep: search history of moved submodules

683 - . grep: enable recurse-submodules to work on <tree> objects

684 - . grep: optionally recurse into submodules

685 - . grep: add submodules as a grep source type

686 - . submodules: load gitmodules file from commit sha1

687 - . submodules: add helper functions to determine presence of submodules

800 +* bw/grep-recurse-submodules (2016-12-16) 7 commits

801 + - grep: search history of moved submodules

802 + - grep: enable recurse-submodules to work on <tree> objects

803 + - grep: optionally recurse into submodules

804 + - grep: add submodules as a grep source type

805 + - submodules: load gitmodules file from commit sha1

806 + - submodules: add helper to determine if a submodule is initialized

807 + - submodules: add helper to determine if a submodule is populated

808 + (this branch uses bw/realpath-wo-chdir.)

809

810 "git grep" learns to optionally recurse into submodules

811

691 - Will be redone on top of the bw/realpath-wo-chdir.

812 + Will merge to 'next'.

813

814

815 * dt/smart-http-detect-server-going-away (2016-11-18) 2 commits

@@ -745,25 +866,6 @@ of the repositories listed at

866 Will cook in 'next'.

867

868

748 -* hv/submodule-not-yet-pushed-fix (2016-11-16) 4 commits

749 - (merged to 'next' on 2016-12-05 at c9d729fca2)

750 - + submodule_needs_pushing(): explain the behaviour when we cannot answer

751 - + batch check whether submodule needs pushing into one call

752 - + serialize collection of refs that contain submodule changes

753 - + serialize collection of changed submodules

754 - (this branch is used by bw/push-dry-run and sb/push-make-submodule-check-the-default.)

755 -

756 - Originally merged to 'next' on 2016-11-21

757 -

758 - The code in "git push" to compute if any commit being pushed in the

759 - superproject binds a commit in a submodule that hasn't been pushed

760 - out was overly inefficient, making it unusable even for a small

761 - project that does not have any submodule but have a reasonable

762 - number of refs.

763 -

764 - Will cook in 'next'.

765 -

766 -

869 * kn/ref-filter-branch-list (2016-12-08) 20 commits

870 - branch: implement '--format' option

871 - branch: use ref-filter printing APIs

@@ -793,19 +895,20 @@ of the repositories listed at

895 and also saw a "squash this in when you reroll", but I lost track.

896

897

796 -* bw/transport-protocol-policy (2016-12-05) 5 commits

898 +* bw/transport-protocol-policy (2016-12-15) 6 commits

899 + - http: respect protocol.*.allow=user for http-alternates

900 - transport: add from_user parameter to is_transport_allowed

901 - http: create function to get curl allowed protocols

799 - - http: always warn if libcurl version is too old

902 - transport: add protocol policy config option

903 + - http: always warn if libcurl version is too old

904 - lib-proto-disable: variable name fix

905 + (this branch uses jk/http-walker-limit-redirect and jk/http-walker-limit-redirect-2.9.)

906

907 Finer-grained control of what protocols are allowed for transports

908 during clone/fetch/push have been enabled via a new configuration

909 mechanism.

910

807 - What's the doneness of this topic? Did we agree that it should be

808 - rebased on top of Peff's http-walker-limit-redirect series?

911 + Will merge to 'next'.

912

913

914 * jt/fetch-no-redundant-tag-fetch-map (2016-11-11) 1 commit

@@ -997,13 +1100,13 @@ of the repositories listed at

1100

1101

1102 * sb/attr (2016-11-11) 35 commits

1000 - - completion: clone can initialize specific submodules

1001 - - clone: add --init-submodule=<pathspec> switch

1002 - - submodule update: add `--init-default-path` switch

1003 - - pathspec: allow escaped query values

1004 - - pathspec: allow querying for attributes

1005 - - pathspec: move prefix check out of the inner loop

1006 - - pathspec: move long magic parsing out of prefix_pathspec

1103 + . completion: clone can initialize specific submodules

1104 + . clone: add --init-submodule=<pathspec> switch

1105 + . submodule update: add `--init-default-path` switch

1106 + . pathspec: allow escaped query values

1107 + . pathspec: allow querying for attributes

1108 + . pathspec: move prefix check out of the inner loop

1109 + . pathspec: move long magic parsing out of prefix_pathspec

1110 - Documentation: fix a typo

1111 - attr: keep attr stack for each check

1112 - attr: convert to new threadsafe API

@@ -1038,10 +1141,11 @@ of the repositories listed at

1141 Building on top of the updated API, the pathspec machinery learned

1142 to select only paths with given attributes set.

1143

1041 - Waiting for review.

1144 + The parts near the tip about pathspec would need to work better

1145 + with bw/pathspec-cleanup topic and has been dropped for now.

1146

1147

1044 -* va/i18n-perl-scripts (2016-11-11) 16 commits

1148 +* va/i18n-perl-scripts (2016-12-14) 16 commits

1149 - i18n: difftool: mark warnings for translation

1150 - i18n: send-email: mark composing message for translation

1151 - i18n: send-email: mark string with interpolation for translation

@@ -1061,7 +1165,7 @@ of the repositories listed at

1165

1166 Porcelain scripts written in Perl are getting internationalized.

1167

1064 - Waiting for review.

1168 + Will merge to 'next'.

1169

1170

1171 * jc/latin-1 (2016-09-26) 2 commits

@@ -1098,13 +1202,14 @@ of the repositories listed at

1202

1203

1204 * jc/pull-rebase-ff (2016-11-29) 1 commit

1101 - - pull: fast-forward "pull --rebase=true"

1205 + (merged to 'next' on 2016-12-16 at c1a0cedd9e)

1206 + + pull: fast-forward "pull --rebase=true"

1207

1208 "git pull --rebase", when there is no new commits on our side since

1209 we forked from the upstream, should be able to fast-forward without

1210 invoking "git rebase", but it didn't.

1211

1107 - Will merge to 'next'.

1212 + Will merge to 'master'.

1213

1214

1215 * jc/merge-drop-old-syntax (2015-04-29) 1 commit

What's cooking (2016/12 #04)