87f4cd26f7 · matheusbsbs/matheus-bagliano

@@ -1,18 +1,20 @@

1 To: git@vger.kernel.org

2 Bcc: lwn@lwn.net

3 -Subject: What's cooking in git.git (Jun 2013, #09; Thu, 27)

4 -X-master-at: ebeea528709abf5ee0a0f3e8ad706f7bacc36c05

5 -X-next-at: d7be87e40699fa4046df5713b92b993a842e1a38

3 +Subject: What's cooking in git.git (Jul 2013, #01; Mon, 1)

4 +X-master-at: 8c4e4ec3ff053331c845331645a437b01b1f0de6

5 +X-next-at: 27cfd277767dfcdf40973f837d71a02664df5d01

6

7 -What's cooking in git.git (Jun 2013, #09; Thu, 27)

7 +What's cooking in git.git (Jul 2013, #01; Mon, 1)

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

9

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

11 '-' are only in 'pu' (proposed updates) while commits prefixed with

12 '+' are in 'next'.

13

14 -There are quite a few fixes already on 'master' that needed merging

15 -down to 'maint'; I'll be tagging 1.8.3.2 shortly.

14 +We are in the middle of 5th week now in the 11-week releace cycle

15 +for 1.8.4, and quite a few topics have graduated to 'master'.

16 +Please help ensure the quality of the upcoming release by testing

17 +the tip of 'master' (and if you are so inclined, 'next') early.

18

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

20 of the repositories listed at

@@ -22,199 +24,256 @@ of the repositories listed at

24 --------------------------------------------------

25 [Graduated to "master"]

26

25 -* bc/checkout-tracking-name-plug-leak (2013-06-18) 2 commits

26 - (merged to 'next' on 2013-06-23 at 35f8f85)

27 - + t/t9802: explicitly name the upstream branch to use as a base

28 - + builtin/checkout.c: don't leak memory in check_tracking_name

29 -

30 - Plug a small leak in checkout.

31 -

32 -

33 -* bc/http-keep-memory-given-to-curl (2013-06-19) 1 commit

34 - (merged to 'next' on 2013-06-23 at cf91828)

35 - + http.c: don't rewrite the user:passwd string multiple times

36 -

37 - Older cURL wanted piece of memory we call it with to be stable, but

38 - we updated the auth material after handing it to a call.

27 +* ap/diff-ignore-blank-lines (2013-06-19) 1 commit

28 + (merged to 'next' on 2013-06-23 at 9cf8e03)

29 + + diff: add --ignore-blank-lines option

30

31 + "git diff" learned a mode that ignores hunks whose change consists

32 + only of additions and removals of blank lines, which is the same as

33 + "diff -B" (ignore blank lines) of GNU diff.

34

41 -* fg/submodule-non-ascii-path (2013-06-20) 2 commits

42 - (merged to 'next' on 2013-06-20 at 122c0b2)

43 - + t7400: test of UTF-8 submodule names pass under Mac OS

44 - (merged to 'next' on 2013-06-17 at a0b9fa3)

45 - + handle multibyte characters in name

35

47 - Many "git submodule" operations do not work on a submodule at a

48 - path whose name is not in ASCII.

36 +* ap/rebase-multiple-fixups (2013-06-27) 1 commit

37 + (merged to 'next' on 2013-06-28 at c3d1b1e)

38 + + rebase -i: handle fixup! fixup! in --autosquash

39

40 + Having multiple "fixup!" on a line in the rebase instruction sheet

41 + did not work very well with "git rebase -i --autosquash".

42

51 -* jk/add-i-custom-diff-algo (2013-06-23) 2 commits

52 - (merged to 'next' on 2013-06-24 at 42e7894)

53 - + add -i: add extra options at the right place in "diff" command line

54 - (merged to 'next' on 2013-06-17 at 48d5d4e)

55 - + add--interactive: respect diff.algorithm

56 -

43

58 -* jk/pull-into-dirty-unborn (2013-06-20) 2 commits

59 - (merged to 'next' on 2013-06-23 at ab4b111)

60 - + pull: merge into unborn by fast-forwarding from empty tree

61 - + pull: update unborn branch tip after index

44 +* ed/color-prompt (2013-06-26) 5 commits

45 + (merged to 'next' on 2013-06-28 at 334892c)

46 + + git-prompt.sh: add missing information in comments

47 + + git-prompt.sh: do not print duplicate clean color code

48 + + t9903: remove redundant tests

49 + + git-prompt.sh: refactor colored prompt code

50 + + t9903: add tests for git-prompt pcmode

51

63 - "git pull" into nothing trashed "local changes" that were in the

64 - index, and this avoids it.

52 + Code clean-up for in-prompt status script (in contrib/).

53

54

67 -* kb/diff-blob-blob-doc (2013-06-18) 1 commit

68 - (merged to 'next' on 2013-06-20 at 34e896a)

69 - + Documentation: Move "git diff <blob> <blob>"

55 +* ft/doc-git-transport (2013-06-26) 1 commit

56 + (merged to 'next' on 2013-06-27 at 4a7d248)

57 + + documentation: add git:// transport security notice

58

59

72 -* mm/merge-in-dirty-worktree-doc (2013-06-18) 1 commit

73 - (merged to 'next' on 2013-06-20 at 21b3267)

74 - + Documentation/git-merge.txt: weaken warning about uncommited changes

60 +* jc/topo-author-date-sort (2013-06-21) 9 commits

61 + (merged to 'next' on 2013-06-26 at 9283719)

62 + + t6003: add --author-date-order test

63 + + topology tests: teach a helper to set author dates as well

64 + + t6003: add --date-order test

65 + + topology tests: teach a helper to take abbreviated timestamps

66 + + t/lib-t6000: style fixes

67 + (merged to 'next' on 2013-06-15 at ad4fb48)

68 + + log: --author-date-order

69 + + sort-in-topological-order: use prio-queue

70 + + prio-queue: priority queue of pointers to structs

71 + + toposort: rename "lifo" field

72 + (this branch uses jk/commit-info-slab; is tangled with jc/show-branch.)

73

74 + "git log" learned the "--author-date-order" option, with which the

75 + output is topologically sorted and commits in parallel histories

76 + are shown intermixed together based on the author timestamp.

77

77 -* mm/push-force-is-dangerous (2013-06-18) 1 commit

78 - (merged to 'next' on 2013-06-20 at a6a8913)

79 - + Documentation/git-push.txt: explain better cases where --force is dangerous

78

79 +* jk/commit-info-slab (2013-06-07) 3 commits

80 + (merged to 'next' on 2013-06-15 at 908ab93)

81 + + commit-slab: introduce a macro to define a slab for new type

82 + + commit-slab: avoid large realloc

83 + + commit: allow associating auxiliary info on-demand

84 + (this branch is used by jc/show-branch and jc/topo-author-date-sort.)

85

82 -* mt/send-email-cc-match-fix (2013-06-20) 2 commits

83 - (merged to 'next' on 2013-06-23 at a2656b3)

84 - + send-email: sanitize author when writing From line

85 - + send-email: add test for duplicate utf8 name

86 + Allow adding custom information to commit objects in order to

87 + represent unbound number of flag bits etc.

88

87 - Logic used by git-send-email to suppress cc mishandled names that

88 - need RFC2047 quoting.

89

90 +* jk/submodule-subdirectory-ok (2013-06-17) 6 commits

91 + (merged to 'next' on 2013-06-23 at f17fb37)

92 + + submodule: drop the top-level requirement

93 + + rev-parse: add --prefix option

94 + + submodule: show full path in error message

95 + + t7403: add missing && chaining

96 + + t7403: modernize style

97 + + t7401: make indentation consistent

98

91 -* ph/builtin-srcs-are-in-subdir-these-days (2013-06-18) 1 commit

92 - (merged to 'next' on 2013-06-20 at 7507a48)

93 - + fix "builtin-*" references to be "builtin/*"

99 + Allow various subcommands of "git submodule" to be run not from the

100 + top of the working tree of the superproject.

101

102

96 -* rr/cherry-pick-fast-forward-reflog-message (2013-06-19) 1 commit

97 - (merged to 'next' on 2013-06-24 at 8b6e9d8)

98 - + sequencer: write useful reflog message for fast-forward

103 +* kb/am-deprecate-resolved (2013-06-27) 1 commit

104 + (merged to 'next' on 2013-06-28 at 177f491)

105 + + am: replace uses of --resolved with --continue

106

100 - The reflog message created when "git cherry-pick" fast-forwarded

101 - did not say anything but "cherry-pick", but it now says

102 - "cherry-pick: fast-forward".

107 + Promote "git am --continue" over "git am --resolved" for UI

108 + consistency.

109

110

105 -* rr/rebase-stash-store (2013-06-17) 5 commits

106 - (merged to 'next' on 2013-06-23 at 2398dad)

107 - + rebase: use 'git stash store' to simplify logic

108 - + stash: introduce 'git stash store'

109 - + stash: simplify option parser for create

110 - + stash doc: document short form -p in synopsis

111 - + stash doc: add a warning about using create

111 +* mh/loose-refs-race-with-pack-ref (2013-06-19) 3 commits

112 + (merged to 'next' on 2013-06-23 at 2099d0b)

113 + + resolve_ref_unsafe(): close race condition reading loose refs

114 + + resolve_ref_unsafe(): handle the case of an SHA-1 within loop

115 + + resolve_ref_unsafe(): extract function handle_missing_loose_ref()

116

113 - Finishing touches for the "git rebase --autostash" feature

114 - introduced earlier.

117 + We read loose and packed rerferences in two steps, but after

118 + deciding to read a loose ref but before actually opening it to read

119 + it, another process racing with us can unlink it, which would cause

120 + us to barf. Update the codepath to retry when such a race is

121 + detected.

122

123

117 -* tr/maint-apply-non-git-patch-parsefix (2013-06-21) 1 commit

118 - (merged to 'next' on 2013-06-24 at 50583a5)

119 - + apply: carefully strdup a possibly-NULL name

124 +* mh/ref-races (2013-06-20) 11 commits

125 + (merged to 'next' on 2013-06-23 at 71f1a18)

126 + + for_each_ref: load all loose refs before packed refs

127 + + get_packed_ref_cache: reload packed-refs file when it changes

128 + + add a stat_validity struct

129 + + Extract a struct stat_data from cache_entry

130 + + packed_ref_cache: increment refcount when locked

131 + + do_for_each_entry(): increment the packed refs cache refcount

132 + + refs: manage lifetime of packed refs cache via reference counting

133 + + refs: implement simple transactions for the packed-refs file

134 + + refs: wrap the packed refs cache in a level of indirection

135 + + pack_refs(): split creation of packed refs and entry writing

136 + + repack_without_ref(): split list curation and entry writing

137 + (this branch is used by mh/ref-races-optim-invalidate-cached.)

138

121 - Fix for the codepath to parse patches that add new files, generated

122 - by programs other than Git. THis is an old breakage in v1.7.11 and

123 - will need to be merged down to the maintanance tracks.

139 + "git pack-refs" that races with new ref creation or deletion have

140 + been susceptible to lossage of refs under right conditions, which

141 + has been tightened up.

142

143

126 -* vl/typofix (2013-06-19) 1 commit

127 - (merged to 'next' on 2013-06-23 at a057524)

128 - + random typofixes (committed missing a 't', successful missing an 's')

144 +* nk/name-rev-abbreviated-refs (2013-06-18) 1 commit

145 + (merged to 'next' on 2013-06-23 at 2ec9e05)

146 + + name-rev: allow to specify a subpath for --refs option

147

148 + "git name-rev --refs=tags/v*" were forbidden, which was a bit

149 + inconvenient (you had to give a pattern to match refs fully, like

150 + --refs=refs/tags/v*).

151

131 -* wk/doc-git-has-grown (2013-06-23) 1 commit

132 - (merged to 'next' on 2013-06-23 at 256263d)

133 - + user-manual: Update download size for Git and the kernel

152

153 +* rr/column-doc (2013-06-26) 1 commit

154 + (merged to 'next' on 2013-06-27 at 99497e1)

155 + + column doc: rewrite documentation for column.ui

156

136 -* ys/cygstart (2013-06-21) 1 commit

137 - (merged to 'next' on 2013-06-24 at 822d9ca)

138 - + web--browse: support /usr/bin/cygstart on Cygwin

157

140 - On "Cygwin", recognize "cygstart" as a possible way to start a web

141 - browser (used in "help -w" and "instaweb" among others).

158 +* rr/remote-branch-config-refresh (2013-06-23) 9 commits

159 + (merged to 'next' on 2013-06-26 at 48746e1)

160 + + t/t5505-remote: test multiple push/pull in remotes-file

161 + + ls-remote doc: don't encourage use of branches-file

162 + + ls-remote doc: rewrite <repository> paragraph

163 + + ls-remote doc: fix example invocation on git.git

164 + + t/t5505-remote: test url-with-# in branches-file

165 + + remote: remove dead code in read_branches_file()

166 + + t/t5505-remote: use test_path_is_missing

167 + + t/t5505-remote: test push-refspec in branches-file

168 + + t/t5505-remote: modernize style

169

143 ---------------------------------------------------

144 -[New Topics]

170 + The original way to specify remote repository using .git/branches/

171 + used to have a nifty feature. The code to support the feature was

172 + still in a function but the caller was changed not to call it 5

173 + years ago, breaking that feature and leaving the supporting code

174 + unreachable.

175

146 -* bp/mediawiki-preview (2013-06-26) 5 commits

147 - - git-remote-mediawiki: Add preview subcommand into git mw.

148 - - git-remote-mediawiki: Adding git-mw command

149 - - git-remote-mediawiki: factoring code between git-remote-mediawiki and Git::Mediawiki

150 - - git-remote-mediawiki: new git bin-wrapper for developement

151 - - git-remote-mediawiki: Introduction of Git::Mediawiki.pm

176

177 +* sb/mailmap-merijn-brand (2013-06-25) 1 commit

178 + (merged to 'next' on 2013-06-27 at 2dd247a)

179 + + .mailmap: Map "H.Merijn Brand" to "H. Merijn Brand"

180

154 -* ed/color-prompt (2013-06-26) 5 commits

155 - - git-prompt.sh: add missing information in comments

156 - - git-prompt.sh: do not print duplicate clean color code

157 - - t9903: remove redundant tests

158 - - git-prompt.sh: refactor colored prompt code

159 - - t9903: add tests for git-prompt pcmode

181

161 - Code clean-up for in-prompt status script (in contrib/).

182 +* sg/bash-prompt (2013-06-24) 16 commits

183 + (merged to 'next' on 2013-06-27 at 09fda6a)

184 + + bash prompt: mention that PROMPT_COMMAND mode is faster

185 + + bash prompt: avoid command substitution when finalizing gitstring

186 + + bash prompt: avoid command substitution when checking for untracked files

187 + + bash prompt: use bash builtins to check stash state

188 + + bash prompt: use bash builtins to check for unborn branch for dirty state

189 + + bash prompt: combine 'git rev-parse' for detached head

190 + + bash prompt: combine 'git rev-parse' executions in the main code path

191 + + bash prompt: use bash builtins to find out current branch

192 + + bash prompt: use bash builtins to find out rebase state

193 + + bash prompt: run 'git rev-parse --git-dir' directly instead of __gitdir()

194 + + bash prompt: return early from __git_ps1() when not in a git repository

195 + + bash prompt: print unique detached HEAD abbreviated object name

196 + + bash prompt: add a test for symbolic link symbolic refs

197 + + completion, bash prompt: move __gitdir() tests to completion test suite

198 + + bash prompt: use 'write_script' helper in interactive rebase test

199 + + bash prompt: fix redirection coding style in tests

200

163 - Will merge to 'next' after getting the merge sanity-checked.

201

202 +* wk/doc-in-linux-3.x-era (2013-06-22) 4 commits

203 + (merged to 'next' on 2013-06-26 at a24e9e6)

204 + + Documentation: Update 'linux-2.6.git' -> 'linux.git'

205 + + Documentation: Update the NFS remote examples to use the staging repo

206 + + doc/clone: Pick more compelling paths for the --reference example

207 + + doc/clone: Remove the '--bare -l -s' example

208

166 -* fg/submodule-clone-depth (2013-06-26) 1 commit

167 - - Add depth to submodule update

209 + Update documentation to match more recent realities.

210

169 - Allow shallow-cloning of submodules with "git submodule update".

211 +--------------------------------------------------

212 +[New Topics]

213

171 - Will merge to 'next'.

214 +* pb/stash-refuse-to-kill (2013-07-01) 2 commits

215 + - git stash: avoid data loss when "git stash save" kills a directory

216 + - treat_directory(): do not declare submodules to be untracked

217

218 + "git stash save" is not just about "saving" the local changes, but

219 + also is to restore the working tree state to that of HEAD. If you

220 + changed a non-directory into a directory in the local change, you

221 + may have untracked files in that directory, which have to be killed

222 + while doing so, unless you run it with --include-untracked. Teach

223 + the command to detect and error out before spreading the damage.

224

174 -* ft/doc-git-transport (2013-06-26) 1 commit

175 - (merged to 'next' on 2013-06-27 at 4a7d248)

176 - + documentation: add git:// transport security notice

225 + This needed a small fix to "ls-files --killed".

226

178 - Will merge to 'master'.

227 + Will merge to 'next' after waiting for a few days.

228

229

181 -* rr/column-doc (2013-06-26) 1 commit

182 - (merged to 'next' on 2013-06-27 at 99497e1)

183 - + column doc: rewrite documentation for column.ui

230 +* es/contacts (2013-06-30) 4 commits

231 + - contrib: contacts: interpret committish akin to format-patch

232 + - contrib: contacts: add ability to parse from committish

233 + - contrib: contacts: add support for multiple patches

234 + - contrib: add git-contacts helper

235

185 - Will merge to 'master'.

236 + Waiting for more discussion.

237

238

188 -* sb/mailmap-merijn-brand (2013-06-25) 1 commit

189 - (merged to 'next' on 2013-06-27 at 2dd247a)

190 - + .mailmap: Map "H.Merijn Brand" to "H. Merijn Brand"

239 +* jk/bash-completion (2013-06-30) 2 commits

240 + (merged to 'next' on 2013-07-01 at 6daca44)

241 + + completion: learn about --man-path

242 + + completion: handle unstuck form of base git options

243

244 Will merge to 'master'.

245

246

195 -* ap/rebase-multiple-fixups (2013-06-27) 1 commit

196 - - rebase -i: handle fixup! fixup! in --autosquash

247 +* cp/submodule-custom-update (2013-07-01) 1 commit

248 + - submodule update: allow custom command to update submodule working tree

249

198 - Having multiple "fixup!" on a line in the rebase instruction sheet

199 - did not work very well with "git rebase -i --autosquash".

250 + In addition to the choice from "rebase, merge, or checkout-detach",

251 + allow a custom command to be used in "submodule update" to update

252 + the working tree of submodules.

253

201 - Will merge to 'next'.

254 + Needs to be reviewed.

255

256

204 -* jc/pull-training-wheel (2013-06-27) 1 commit

205 - - pull: require choice between rebase/merge on non-fast-forward pull

257 +* mv/merge-ff-tristate (2013-07-01) 1 commit

258 + - merge: handle --ff/--no-ff/--ff-only as a tri-state option

259

207 - Make "git pull" refuse with "it does not fast forward; choose between

208 - 'pull --merge' and 'pull --rebase'".

260 + The configuration variable "merge.ff" was cleary a tri-state to

261 + choose one from "favor fast-forward when possible", "always create

262 + a merge even when the history could fast-forward" and "do not

263 + create any merge, only update when the history fast-forwards", but

264 + the command line parser did not implement the usual convention of

265 + "last one wins, and command line overrides the configuration"

266 + correctly.

267

268 + Needs to be reviewed.

269

211 -* kb/am-deprecate-resolved (2013-06-27) 1 commit

212 - - am: replace uses of --resolved with --continue

270

214 - Promote "git am --continue" over "git am --resolved" for UI

215 - consistency.

271 +* rr/rebase-reflog-message-reword (2013-06-23) 2 commits

272 + - rebase -i: use a better reflog message

273 + - rebase: use a better reflog message

274 + (this branch uses rr/rebase-checkout-reflog.)

275

217 - Will merge to 'next'.

276 + Split from the other topic.

277

278 --------------------------------------------------

279 [Stalled]

@@ -231,16 +290,6 @@ of the repositories listed at

290 $gmane/228294

291

292

234 -* fc/contrib-related (2013-06-09) 5 commits

235 - - contrib: related: print the amount of involvement

236 - - contrib: related: parse committish like format-patch

237 - - contrib: related: add option to parse from committish

238 - - contrib: related: add support for multiple patches

239 - - Add new git-related helper to contrib

240 -

241 - Will discard.

242 -

243 -

293 * hv/config-from-blob (2013-05-12) 5 commits

294 - do not die when error in config parsing of buf occurs

295 - teach config --blob option to parse config from database

@@ -352,55 +401,55 @@ of the repositories listed at

401 - show-branch.c: abstract out "flags" operation

402 - show-branch.c: lift all_mask/all_revs to a global static

403 - show-branch.c: update comment style

355 - (this branch uses jk/commit-info-slab; is tangled with jc/topo-author-date-sort.)

404

405 Waiting for the final step to lift the hard-limit before sending it out.

406

407 --------------------------------------------------

408 [Cooking]

409

362 -* as/log-output-encoding-in-user-format (2013-06-26) 5 commits

363 - - pretty: --format output should honor logOutputEncoding

364 - - pretty: Add failing tests: --format output should honor logOutputEncoding

365 - - t4205 (log-pretty-formats): don't hardcode SHA-1 in expected outputs

366 - - t7102 (reset): don't hardcode SHA-1 in expected outputs

367 - - t6006 (rev-list-format): don't hardcode SHA-1 in expected outputs

410 +* bp/mediawiki-preview (2013-06-26) 5 commits

411 + - git-remote-mediawiki: Add preview subcommand into git mw.

412 + - git-remote-mediawiki: Adding git-mw command

413 + - git-remote-mediawiki: factoring code between git-remote-mediawiki and Git::Mediawiki

414 + - git-remote-mediawiki: new git bin-wrapper for developement

415 + - git-remote-mediawiki: Introduction of Git::Mediawiki.pm

416

369 - "log --format=" did not honor i18n.logoutputencoding configuration

370 - and this is a beginning of the series that attempts to fix it.

417 + Still under discussion?

418

372 - Will merge to 'next'.

419

420 +* fg/submodule-clone-depth (2013-06-28) 1 commit

421 + - Add --depth to submodule update/add

422

375 -* rr/remote-branch-config-refresh (2013-06-23) 9 commits

376 - (merged to 'next' on 2013-06-26 at 48746e1)

377 - + t/t5505-remote: test multiple push/pull in remotes-file

378 - + ls-remote doc: don't encourage use of branches-file

379 - + ls-remote doc: rewrite <repository> paragraph

380 - + ls-remote doc: fix example invocation on git.git

381 - + t/t5505-remote: test url-with-# in branches-file

382 - + remote: remove dead code in read_branches_file()

383 - + t/t5505-remote: use test_path_is_missing

384 - + t/t5505-remote: test push-refspec in branches-file

385 - + t/t5505-remote: modernize style

423 + Allow shallow-cloning of submodules with "git submodule update".

424

387 - The original way to specify remote repository using .git/branches/

388 - used to have a nifty feature. The code to support the feature was

389 - still in a function but the caller was changed not to call it 5

390 - years ago, breaking that feature and leaving the supporting code

391 - unreachable.

425 + Waiting for a reroll.

426 + $gmane/229248

427

393 - Will merge to 'master'.

428

429 +* jc/pull-training-wheel (2013-06-27) 1 commit

430 + - pull: require choice between rebase/merge on non-fast-forward pull

431

396 -* wk/doc-in-linux-3.x-era (2013-06-22) 4 commits

397 - (merged to 'next' on 2013-06-26 at a24e9e6)

398 - + Documentation: Update 'linux-2.6.git' -> 'linux.git'

399 - + Documentation: Update the NFS remote examples to use the staging repo

400 - + doc/clone: Pick more compelling paths for the --reference example

401 - + doc/clone: Remove the '--bare -l -s' example

432 + Make "git pull" refuse with "it does not fast forward; choose between

433 + 'pull --merge' and 'pull --rebase'".

434

403 - Update documentation to match more recent realities.

435 + John Keeping seems to want to restrict this to "git pull" without

436 + any other argument. I personally do not have a strong opinion

437 + either way, but I'm inclined to take such a patch to loosen the new

438 + logic with other people's support. Hint, hint...

439 +

440 +

441 +* as/log-output-encoding-in-user-format (2013-07-01) 6 commits

442 + (merged to 'next' on 2013-07-01 at 3318aa8)

443 + + t4205: replace .\+ with ..* in sed commands

444 + (merged to 'next' on 2013-06-28 at 4063330)

445 + + pretty: --format output should honor logOutputEncoding

446 + + pretty: Add failing tests: --format output should honor logOutputEncoding

447 + + t4205 (log-pretty-formats): don't hardcode SHA-1 in expected outputs

448 + + t7102 (reset): don't hardcode SHA-1 in expected outputs

449 + + t6006 (rev-list-format): don't hardcode SHA-1 in expected outputs

450 +

451 + "log --format=" did not honor i18n.logoutputencoding configuration

452 + and this is a beginning of the series that attempts to fix it.

453

454 Will merge to 'master'.

455

@@ -412,8 +461,7 @@ of the repositories listed at

461 set to true (i.e. error out) and there are offending lines in the

462 working tree files.

463

415 - Will hold for a while to see if anybody can come up with better

416 - tests.

464 + Will merge to 'next'.

465

466

467 * jk/fast-import-empty-ls (2013-06-23) 4 commits

@@ -423,29 +471,7 @@ of the repositories listed at

471 - t9300: document fast-import empty path issues

472

473 Waiting for reviews.

426 -

427 -

428 -* sg/bash-prompt (2013-06-24) 16 commits

429 - (merged to 'next' on 2013-06-27 at 09fda6a)

430 - + bash prompt: mention that PROMPT_COMMAND mode is faster

431 - + bash prompt: avoid command substitution when finalizing gitstring

432 - + bash prompt: avoid command substitution when checking for untracked files

433 - + bash prompt: use bash builtins to check stash state

434 - + bash prompt: use bash builtins to check for unborn branch for dirty state

435 - + bash prompt: combine 'git rev-parse' for detached head

436 - + bash prompt: combine 'git rev-parse' executions in the main code path

437 - + bash prompt: use bash builtins to find out current branch

438 - + bash prompt: use bash builtins to find out rebase state

439 - + bash prompt: run 'git rev-parse --git-dir' directly instead of __gitdir()

440 - + bash prompt: return early from __git_ps1() when not in a git repository

441 - + bash prompt: print unique detached HEAD abbreviated object name

442 - + bash prompt: add a test for symbolic link symbolic refs

443 - + completion, bash prompt: move __gitdir() tests to completion test suite

444 - + bash prompt: use 'write_script' helper in interactive rebase test

445 - + bash prompt: fix redirection coding style in tests

446 - (this branch is used by sg/bash-prompt-premerge.)

447 -

448 - Will merge to 'master'.

474 + $gmane/228741

475

476

477 * af/rebase-i-merge-options (2013-06-25) 2 commits

@@ -455,6 +481,7 @@ of the repositories listed at

481 "git rebase -i" now honors --strategy and -X options.

482

483 Waiting for response to review comments.

484 + $gmane/228988

485

486

487 * rj/cygwin-remove-cheating-lstat (2013-06-25) 1 commit

@@ -463,6 +490,9 @@ of the repositories listed at

490 Seems that this gives different performance results to different

491 people.

492

493 + A replacement WIP has been posted.

494 + $gmane/229280

495 +

496

497 * jc/triangle-push-fixup (2013-06-24) 5 commits

498 (merged to 'next' on 2013-06-26 at 73cbb69)

@@ -512,10 +542,11 @@ of the repositories listed at

542

543 The bottom two has been graduated to 'master' but then reverted.

544 The tip two are quick attempts to fix the fallout. The one for

515 - status.short looks correct; the other one is still in flux and

516 - under discussion.

545 + status.short looks correct; the other one, while it is correct,

546 + is unfortunately overly complex in order not to introduce an

547 + unnecessary regression.

548

518 - Waiting for possible alternative cleaner patches for the tip one.

549 + Will merge to 'next'.

550

551

552 * jc/reflog-doc (2013-06-19) 1 commit

@@ -529,133 +560,28 @@ of the repositories listed at

560

561 * mh/ref-races-optim-invalidate-cached (2013-06-20) 1 commit

562 - refs: do not invalidate the packed-refs cache unnecessarily

532 - (this branch uses mh/ref-races.)

563

534 - Optional.

564 + This requires the platform lstat() to be correct to avoid false

565 + negatives.

566

567

537 -* rr/rebase-checkout-reflog (2013-06-23) 7 commits

538 - - rebase -i: use a better reflog message

539 - - rebase: use a better reflog message

540 - - checkout: respect GIT_REFLOG_ACTION

568 +* rr/rebase-checkout-reflog (2013-06-17) 5 commits

569 + (merged to 'next' on 2013-07-01 at 27cfd27)

570 + + checkout: respect GIT_REFLOG_ACTION

571 (merged to 'next' on 2013-06-27 at 4d99efa)

572 + status: do not depend on rebase reflog messages

573 + t/t2021-checkout-last: "checkout -" should work after a rebase finishes

574 + wt-status: remove unused field in grab_1st_switch_cbdata

575 + t7512: test "detached from" as well

576 + (this branch is used by rr/rebase-reflog-message-reword.)

577

578 Invocations of "git checkout" used internally by "git rebase" were

579 counted as "checkout", and affected later "git checkout -" to the

580 the user to an unexpected place.

581

551 - The first four are uncontroversial and ready to be merged.

552 - Will split and merge to 'next'.

553 -

554 -

555 -* nk/name-rev-abbreviated-refs (2013-06-18) 1 commit

556 - (merged to 'next' on 2013-06-23 at 2ec9e05)

557 - + name-rev: allow to specify a subpath for --refs option

558 -

559 - "git name-rev --refs=tags/v*" were forbidden, which was a bit

560 - inconvenient (you had to give a pattern to match refs fully, like

561 - --refs=refs/tags/v*).

562 -

563 - Will merge to 'master'.

564 -

565 -

566 -* jk/submodule-subdirectory-ok (2013-06-17) 6 commits

567 - (merged to 'next' on 2013-06-23 at f17fb37)

568 - + submodule: drop the top-level requirement

569 - + rev-parse: add --prefix option

570 - + submodule: show full path in error message

571 - + t7403: add missing && chaining

572 - + t7403: modernize style

573 - + t7401: make indentation consistent

574 -

575 - Allow various subcommands of "git submodule" to be run not from the

576 - top of the working tree of the superproject.

577 -

578 - Will merge to 'master'.

579 -

580 -

581 -* mh/ref-races (2013-06-20) 11 commits

582 - (merged to 'next' on 2013-06-23 at 71f1a18)

583 - + for_each_ref: load all loose refs before packed refs

584 - + get_packed_ref_cache: reload packed-refs file when it changes

585 - + add a stat_validity struct

586 - + Extract a struct stat_data from cache_entry

587 - + packed_ref_cache: increment refcount when locked

588 - + do_for_each_entry(): increment the packed refs cache refcount

589 - + refs: manage lifetime of packed refs cache via reference counting

590 - + refs: implement simple transactions for the packed-refs file

591 - + refs: wrap the packed refs cache in a level of indirection

592 - + pack_refs(): split creation of packed refs and entry writing

593 - + repack_without_ref(): split list curation and entry writing

594 - (this branch is used by mh/ref-races-optim-invalidate-cached.)

595 -

596 - "git pack-refs" that races with new ref creation or deletion have

597 - been susceptible to lossage of refs under right conditions, which

598 - has been tightened up.

599 -

600 - Will merge to 'master'.

601 -

602 -

603 -* jc/topo-author-date-sort (2013-06-21) 9 commits

604 - (merged to 'next' on 2013-06-26 at 9283719)

605 - + t6003: add --author-date-order test

606 - + topology tests: teach a helper to set author dates as well

607 - + t6003: add --date-order test

608 - + topology tests: teach a helper to take abbreviated timestamps

609 - + t/lib-t6000: style fixes

610 - (merged to 'next' on 2013-06-15 at ad4fb48)

611 - + log: --author-date-order

612 - + sort-in-topological-order: use prio-queue

613 - + prio-queue: priority queue of pointers to structs

614 - + toposort: rename "lifo" field

615 - (this branch uses jk/commit-info-slab; is tangled with jc/show-branch.)

616 -

617 - "git log" learned the "--author-date-order" option, with which the

618 - output is topologically sorted and commits in parallel histories

619 - are shown intermixed together based on the author timestamp.

620 -

621 - Will merge to 'master'.

622 -

623 -

624 -* mh/loose-refs-race-with-pack-ref (2013-06-19) 3 commits

625 - (merged to 'next' on 2013-06-23 at 2099d0b)

626 - + resolve_ref_unsafe(): close race condition reading loose refs

627 - + resolve_ref_unsafe(): handle the case of an SHA-1 within loop

628 - + resolve_ref_unsafe(): extract function handle_missing_loose_ref()

629 -

630 - We read loose and packed rerferences in two steps, but after

631 - deciding to read a loose ref but before actually opening it to read

632 - it, another process racing with us can unlink it, which would cause

633 - us to barf. Update the codepath to retry when such a race is

634 - detected.

635 -

636 - Will merge to 'master'.

637 -

638 -

639 -* jk/commit-info-slab (2013-06-07) 3 commits

640 - (merged to 'next' on 2013-06-15 at 908ab93)

641 - + commit-slab: introduce a macro to define a slab for new type

642 - + commit-slab: avoid large realloc

643 - + commit: allow associating auxiliary info on-demand

644 - (this branch is used by jc/show-branch and jc/topo-author-date-sort.)

645 -

646 - Allow adding custom information to commit objects in order to

647 - represent unbound number of flag bits etc.

648 -

649 - Will merge to 'master'.

650 -

651 -

652 -* ap/diff-ignore-blank-lines (2013-06-19) 1 commit

653 - (merged to 'next' on 2013-06-23 at 9cf8e03)

654 - + diff: add --ignore-blank-lines option

655 -

656 - "git diff" learned a mode that ignores hunks whose change consists

657 - only of additions and removals of blank lines, which is the same as

658 - "diff -B" (ignore blank lines) of GNU diff.

582 + The "bugfix" bits of the series is here. The follow-up part to

583 + reword reflog messages for steps in "rebase" are split to another

584 + topic.

585

586 Will merge to 'master'.

587

@@ -687,7 +613,9 @@ of the repositories listed at

613 merged to 'next'.

614

615

690 -* tr/test-v-and-v-subtest-only (2013-06-23) 8 commits

616 +* tr/test-v-and-v-subtest-only (2013-06-29) 9 commits

617 + (merged to 'next' on 2013-06-30 at 1c5fac1)

618 + + perf-lib: fix start/stop of perf tests

619 (merged to 'next' on 2013-06-26 at 8ff4d84)

620 + test-lib: support running tests under valgrind in parallel

621 + test-lib: allow prefixing a custom string before "ok N" etc.

@@ -701,8 +629,7 @@ of the repositories listed at

629 Allows N instances of tests run in parallel, each running 1/N parts

630 of the test suite under Valgrind, to speed things up.

631

704 - The tip two may be useful in practice but is a tad ugly ;-)

705 - Rerolled to fix breakage in non-prove tests.

632 + Will merge to 'master'.

633

634

635 * jn/add-2.0-u-A-sans-pathspec (2013-04-26) 1 commit

What's cooking (2013/07 #01)