764fdd9904 · matheusbsbs/matheus-bagliano

@@ -1,119 +1,204 @@

1 To: git@vger.kernel.org

2 Bcc: lwn@lwn.net

3 -Subject: What's cooking in git.git (May 2013, #03; Thu, 9)

4 -X-master-at: b387c77b12953d543bf5efc3825b5be26d753f9c

5 -X-next-at: 747d353d98dc660fb24ce70ad72ad9d9553c40ae

3 +Subject: What's cooking in git.git (May 2013, #04; Wed, 15)

4 +X-master-at: 0c2b1cf812be6e7e35bed0844b37e412cd232b6b

5 +X-next-at: 2fb82f5641542fc2b7c7ba84210b67d8255297c8

6

7 -What's cooking in git.git (May 2013, #03; Thu, 9)

7 +What's cooking in git.git (May 2013, #04; Wed, 15)

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 -I'll tag 1.8.3-rc2 by the end of the week, and hopefully we can have

15 -a solid final release late next week. One embarrassing regression

16 -was found and had to be fixed today.

14 +We are past -rc2 and haven't seen any regression reported since the

15 +feature freeze started, which may be a good sign (the coming release

16 +is perfect) or a bad sign (nobody is testing). We'll see soon when

17 +we tag the 1.8.3 final sometime next week. We may see subsystem

18 +(git-svn? gitk? git-gui?) and l10n updates before that happens.

19

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

21 of the repositories listed at

22

23 http://git-blame.blogspot.com/p/git-public-repositories.html

24

23 ---------------------------------------------------

24 -[Graduated to "master"]

25 -

26 -* fc/sha1-name-at-home (2013-05-07) 14 commits

27 - - sha1_name: compare variable with constant, not constant with variable

28 - - Add new @ shortcut for HEAD

29 - - sha1_name: refactor reinterpret()

30 - - sha1_name: check @{-N} errors sooner

31 - - sha1_name: reorganize get_sha1_basic()

32 - - sha1_name: don't waste cycles in the @-parsing loop

33 - - sha1_name: remove unnecessary braces

34 - - sha1_name: remove no-op

35 - - tests: at-combinations: @{N} versus HEAD@{N}

36 - - tests: at-combinations: increase coverage

37 - - tests: at-combinations: improve nonsense()

38 - - SQUASH??? style fixups

39 - - tests: at-combinations: check ref names directly

40 - - tests: at-combinations: simplify setup

41 -

25 --------------------------------------------------

26 [New Topics]

27

45 -* jh/shorten-refname (2013-05-07) 4 commits

46 - - t1514: refname shortening is done after dereferencing symbolic refs

47 - - shorten_unambiguous_ref(): Fix shortening refs/remotes/origin/HEAD to origin

48 - - t1514: Demonstrate failure to correctly shorten "refs/remotes/origin/HEAD"

49 - - t1514: Add tests of shortening refnames in strict/loose mode

50 -

51 - When remotes/origin/HEAD is not a symbolic ref, "rev-parse

52 - --abbrev-ref remotes/origin/HEAD" ought to show "origin", not

53 - "origin/HEAD", which is fixed with this series (if it is a symbolic

54 - ref that points at remotes/origin/something, then it should show

55 - "origin/something" and it already does).

56 -

28 +* dw/asciidoc-sources-are-dot-txt-files (2013-05-10) 1 commit

29 + - CodingGuidelines: Documentation/*.txt are the sources

30 +

31 +

32 +* fc/doc-style (2013-05-09) 2 commits

33 + - SQUASH??? more consistently update docs

34 + - documentation: trivial style cleanups

35 +

36 +

37 +* da/darwin (2013-05-15) 3 commits

38 + - imap-send: eliminate HMAC deprecation warnings on Mac OS X

39 + - cache.h: eliminate SHA-1 deprecation warnings on Mac OS X

40 + - Makefile: fix default regex settings on Darwin

41 +

42 + Waiting for polishing discussion to finish.

43 +

44 +

45 +* fc/macos-x-clipped-write (2013-05-10) 2 commits

46 + - SQUASH???

47 + - compate/clipped-write.c: large write(2) fails on Mac OS X/XNU

48 +

49 + The tip needs to be squashed, as it was reported to have tested OK.

50 +

51 +

52 +* fc/remote-hg (2013-05-15) 39 commits

53 + - remote-hg: remove files before modifications

54 + - remote-hg: improve lightweight tag author

55 + - remote-hg: use remote 'default' not local one

56 + - remote-hg: improve branch listing

57 + - remote-hg: simplify branch_tip()

58 + - remote-hg: check diverged bookmarks

59 + - remote-hg: pass around revision refs

60 + - remote-hg: implement custom checkheads()

61 + - remote-hg: implement custom push()

62 + - remote-hg: only update necessary revisions

63 + - remote-hg: force remote bookmark push selectively

64 + - remote-hg: reorganize bookmark handling

65 + - remote-hg: add test for failed double push

66 + - remote-hg: add test for big push

67 + - remote-hg: add test for new bookmark special

68 + - remote-hg: add test for bookmark diverge

69 + - remote-hg: add test for diverged push

70 + - remote-hg: add test to push new bookmark

71 + - remote-hg: add remote tests

72 + - remote-hg: add check_bookmark() test helper

73 + - remote-bzr: simplify test checks

74 + - remote-hg: always point HEAD to master

75 + - remote-hg: improve progress calculation

76 + - remote-hg: trivial cleanups

77 + - remote-hg: ensure remote rebasing works

78 + - remote-hg: upgrade version 1 marks

79 + - remote-hg: switch from revisions to SHA-1 noteids

80 + - remote-hg: add version checks to the marks

81 + - remote-hg: improve node traversing

82 + - remote-hg: shuffle some code

83 + - remote-hg: use a shared repository store

84 + - remote-hg: load all extensions

85 + - remote-hg: test: simplify previous branch checkout

86 + - remote-helpers: test: simplify remote URLs

87 + - remote-helpers: tests: general improvements

88 + - remote-helpers: test: cleanup style

89 + - remote-helpers: test: cleanup white-spaces

90 + - remote-hg: trivial reorganization

91 + - remote-hg: test: be a little more quiet

92 +

93 + Probably will be one of the early topics to graduate post 1.8.3.

94 +

95 +

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

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

98 + - teach config --blob option to parse config from database

99 + - config: make parsing stack struct independent from actual data source

100 + - config: drop cf validity check in get_next_char()

101 + - config: factor out config file stack management

102 +

103 +

104 +* jc/t5551-posix-sed-bre (2013-05-12) 1 commit

105 + - t5551: do not use unportable sed '\+'

106 +

107 + Not a regression fix and not urgent.

108 +

109 +

110 +* jk/fetch-always-update-tracking (2013-05-12) 4 commits

111 + - fetch: opportunistically update tracking refs

112 + - refactor "ref->merge" flag

113 + - fetch/pull doc: untangle meaning of bare <ref>

114 + - t5510: start tracking-ref tests from a known state

115 +

116 + "git fetch origin master" unlike "git fetch origin" or "git fetch"

117 + does not update "refs/remotes/origin/master" and it was an early

118 + design decision to keep the update of remote tracking branches

119 + predictable, but in practice it turns out that people find it more

120 + convenient to opportunisticly update them whenever we have a

121 + chance, and we have been updating them when we run "git push" which

122 + already breaks the original "predictability" anyway.

123 +

124 +

125 +* nd/clone-connectivity-shortcut (2013-05-11) 4 commits

126 + - clone: open a shortcut for connectivity check

127 + - index-pack: remove dead code (it should never happen)

128 + - fetch-pack: prepare updated shallow file before fetching the pack

129 + - clone: let the user know when check_everything_connected is run

130 +

131 +

132 +* rr/rebase-autostash (2013-05-12) 7 commits

133 + - rebase: implement --[no-]autostash and rebase.autostash

134 + - rebase --merge: return control to caller, for housekeeping

135 + - rebase -i: return control to caller, for housekeeping

136 + - am: return control to caller, for housekeeping

137 + - rebase: prepare to do generic housekeeping

138 + - rebase -i: don't error out if $state_dir already exists

139 + - am: tighten a conditional that checks for $dotest

140 +

141 +

142 +* kb/ancestry-path-threedots (2013-05-13) 1 commit

143 + - t6019: demonstrate --ancestry-path A...B breakage

144 + (this branch is used by kb/full-history-compute-treesame-carefully-2.)

145 +

146 +

147 +* kb/full-history-compute-treesame-carefully-2 (2013-05-13) 11 commits

148 + - revision.c: treat A...B merge bases as if manually specified

149 + - revision.c: discount side branches when computing TREESAME

150 + - simplify-merges: drop merge from irrelevant side branch

151 + - simplify-merges: never remove all TREESAME parents

152 + - t6012: update test for tweaked full-history traversal

153 + - revision.c: Make --full-history consider more merges

154 + - rev-list-options.txt: correct TREESAME for P

155 + - t6111: allow checking the parents as well

156 + - t6111: new TREESAME test set

157 + - t6019: test file dropped in -s ours merge

158 + - decorate.c: compact table when growing

159 + (this branch uses kb/ancestry-path-threedots.)

160

58 -* nd/clone-local-with-colon (2013-05-07) 1 commit

59 - - clone: allow cloning local paths with colons in them

161 + Major update to a very core part of the system to improve culling

162 + of irrelevant parents while traversing a mergy history.

163

61 - "git clone foo/bar:baz" cannot be a request to clone from a remote

62 - over git-over-ssh specified in the scp style. Detect this case and

63 - clone from a local repository at "foo/bar:baz".

164 + Will not be a 1.8.3 material, but is an important topic.

165

166

66 -* nd/warn-ambiguous-object-name (2013-05-07) 1 commit

67 - - get_sha1: improve ambiguity warning regarding SHA-1 and ref names

167 +* jc/strbuf-branchname-fix (2013-05-15) 1 commit

168 + - strbuf_branchname(): do not double-expand @{-1}~22

169

69 - "git cmd <name>", when <name> happens to be a 40-hex string,

70 - directly uses the 40-hex string as an object name, even if a ref

71 - "refs/<some hierarchy>/<name>" exists. This disambiguation order

72 - is unlikely to change, but we should warn about the ambiguity just

73 - like we warn when more than one refs/ hierachies share the same

74 - name.

170 + "git merge @{-1}~22" was rewritten to "git merge frotz@{1}~22"

171 + incorrectly when your previous branch was "frotz" (it should be

172 + rewritten to "git merge frotz~22" instead).

173

76 - The message needs to be fixed up, as this is not "refname is

77 - ambiguous".

174

175 +* mk/combine-diff-context-horizon-fix (2013-05-15) 1 commit

176 + - combine-diff.c: Fix output when changes are exactly 3 lines apart

177

80 -* fc/zsh-leftover-bits (2013-05-08) 2 commits

81 - - completion: synchronize zsh wrapper

82 - - completion: cleanup zsh wrapper

178 + "git diff -c -p" was not showing a deleted line from a hunk when

179 + another hunk immediately begins where the earlier one ends.

180

181

85 -* rs/empty-archive (2013-05-09) 2 commits

86 - - t5004: resurrect original empty tar archive test

87 - - t5004: avoid using tar for checking emptiness of archive

182 +* mm/color-auto-default (2013-05-15) 2 commits

183 + - make color.ui default to 'auto'

184 + - config: refactor management of color.ui's default value

185

89 - Fixes tests added in 1.8.2 era that are broken on BSDs,

90 - but it is not all that urgent.

186 + Flip the default for color.ui to 'auto', which is what many

187 + tutorials recommend new users to do.

188

189

93 -* vv/help-unknown-ref (2013-05-08) 2 commits

94 - - merge: use help_unknown_ref()

95 - - help: add help_unknown_ref()

190 +* tr/coverage (2013-05-13) 4 commits

191 + - coverage: build coverage-untested-functions by default

192 + - coverage: set DEFAULT_TEST_TARGET to avoid using prove

193 + - coverage: do not delete .gcno files before building

194 + - coverage: split build target into compile and test

195

97 - Detect "git merge foo" that might have meant "git merge origin/foo"

98 - and give an error message that is more specific than "foo is not

99 - something we can merge".

196 + Update the test coverage support that was left to bitrot for some

197 + time.

198

199 --------------------------------------------------

200 [Stalled]

201

104 -* mg/more-textconv (2013-04-23) 7 commits

105 - - git grep: honor textconv by default

106 - - grep: honor --textconv for the case rev:path

107 - - grep: allow to use textconv filters

108 - - t7008: demonstrate behavior of grep with textconv

109 - - cat-file: do not die on --textconv without textconv filters

110 - - show: honor --textconv for blobs

111 - - t4030: demonstrate behavior of show with textconv

112 -

113 - Rerolled. I am not sure if I like "show <blob>" and "grep" that use

114 - textconv by default, though.

115 -

116 -

202 * mh/multimail (2013-04-21) 1 commit

203 - git-multimail: a replacement for post-receive-email

204

@@ -165,6 +250,77 @@ of the repositories listed at

250 --------------------------------------------------

251 [Cooking]

252

253 +* mg/more-textconv (2013-05-10) 7 commits

254 + - grep: honor --textconv for the case rev:path

255 + - grep: allow to use textconv filters

256 + - t7008: demonstrate behavior of grep with textconv

257 + - cat-file: do not die on --textconv without textconv filters

258 + - show: honor --textconv for blobs

259 + - diff_opt: track whether flags have been set explicitly

260 + - t4030: demonstrate behavior of show with textconv

261 +

262 + Looked reasonable, I think. Comments?

263 +

264 +

265 +* jh/shorten-refname (2013-05-07) 4 commits

266 + - t1514: refname shortening is done after dereferencing symbolic refs

267 + - shorten_unambiguous_ref(): Fix shortening refs/remotes/origin/HEAD to origin

268 + - t1514: Demonstrate failure to correctly shorten "refs/remotes/origin/HEAD"

269 + - t1514: Add tests of shortening refnames in strict/loose mode

270 +

271 + When remotes/origin/HEAD is not a symbolic ref, "rev-parse

272 + --abbrev-ref remotes/origin/HEAD" ought to show "origin", not

273 + "origin/HEAD", which is fixed with this series (if it is a symbolic

274 + ref that points at remotes/origin/something, then it should show

275 + "origin/something" and it already does).

276 +

277 + I think this is being rerolled using strbuf_expand().

278 +

279 +

280 +* nd/clone-local-with-colon (2013-05-07) 1 commit

281 + - clone: allow cloning local paths with colons in them

282 +

283 + "git clone foo/bar:baz" cannot be a request to clone from a remote

284 + over git-over-ssh specified in the scp style. Detect this case and

285 + clone from a local repository at "foo/bar:baz".

286 +

287 +

288 +* nd/warn-ambiguous-object-name (2013-05-07) 1 commit

289 + - get_sha1: improve ambiguity warning regarding SHA-1 and ref names

290 +

291 + "git cmd <name>", when <name> happens to be a 40-hex string,

292 + directly uses the 40-hex string as an object name, even if a ref

293 + "refs/<some hierarchy>/<name>" exists. This disambiguation order

294 + is unlikely to change, but we should warn about the ambiguity just

295 + like we warn when more than one refs/ hierachies share the same

296 + name.

297 +

298 + The message needs to be fixed up, as this is not "refname is

299 + ambiguous".

300 +

301 +

302 +* fc/zsh-leftover-bits (2013-05-08) 2 commits

303 + - completion: synchronize zsh wrapper

304 + - completion: cleanup zsh wrapper

305 +

306 +

307 +* rs/empty-archive (2013-05-09) 2 commits

308 + - t5004: resurrect original empty tar archive test

309 + - t5004: avoid using tar for checking emptiness of archive

310 +

311 + Fixes tests added in 1.8.2 era that are broken on BSDs,

312 + but it is not all that urgent.

313 +

314 +

315 +* vv/help-unknown-ref (2013-05-08) 2 commits

316 + - merge: use help_unknown_ref()

317 + - help: add help_unknown_ref()

318 +

319 + Detect "git merge foo" that might have meant "git merge origin/foo"

320 + and give an error message that is more specific than "foo is not

321 + something we can merge".

322 +

323 +

324 * fc/fast-export-persistent-marks (2013-05-07) 3 commits

325 - fast-{import,export}: use get_sha1_hex() to read from marks file

326 - fast-export: don't parse commits while reading marks file

@@ -278,12 +434,24 @@ of the repositories listed at

434 Will cook in 'next'.

435

436

281 -* js/transport-helper-error-reporting-fix (2013-04-28) 3 commits

437 +* js/transport-helper-error-reporting-fix (2013-04-28) 13 commits

438 (merged to 'next' on 2013-04-29 at 8cc4bb8)

439 + git-remote-testgit: build it to run under $SHELL_PATH

440 + git-remote-testgit: further remove some bashisms

441 + git-remote-testgit: avoid process substitution

286 - (this branch uses fc/transport-helper-error-reporting.)

442 + (merged to 'next' on 2013-04-25 at 3358f1a)

443 + + t5801: "VAR=VAL shell_func args" is forbidden

444 + (merged to 'next' on 2013-04-22 at 5ba6467)

445 + + transport-helper: update remote helper namespace

446 + + transport-helper: trivial code shuffle

447 + + transport-helper: warn when refspec is not used

448 + + transport-helper: clarify pushing without refspecs

449 + + transport-helper: update refspec documentation

450 + + transport-helper: clarify *:* refspec

451 + + transport-helper: improve push messages

452 + + transport-helper: mention helper name when it dies

453 + + transport-helper: report errors properly

454 + (this branch is tangled with fc/transport-helper-error-reporting.)

455

456 Finishing touches to fc/transport-helper-error-reporting topic.

457 Will cook in 'next'.

@@ -298,25 +466,6 @@ of the repositories listed at

466 Will cook in 'next'.

467

468

301 -* kb/full-history-compute-treesame-carefully (2013-05-06) 11 commits

302 - - revision.c: treat A...B merge bases as if manually specified

303 - - revision.c: discount side branches when computing TREESAME

304 - - simplify-merges: drop merge from irrelevant side branch

305 - - simplify-merges: never remove all TREESAME parents

306 - - t6012: update test for tweaked full-history traversal

307 - - revision.c: Make --full-history consider more merges

308 - - rev-list-options.txt: correct TREESAME for P

309 - - t6111: allow checking the parents as well

310 - - t6111: new TREESAME test set

311 - - t6019: test file dropped in -s ours merge

312 - - decorate.c: compact table when growing

313 -

314 - Major update to a very core part of the system to improve culling

315 - of irrelevant parents while traversing a mergy history.

316 -

317 - Will not be a 1.8.3 material, but is an important topic.

318 -

319 -

469 * jh/checkout-auto-tracking (2013-04-21) 8 commits

470 (merged to 'next' on 2013-04-22 at 2356700)

471 + glossary: Update and rephrase the definition of a remote-tracking branch

@@ -414,7 +563,10 @@ of the repositories listed at

563 Will cook in 'next'.

564

565

417 -* fc/transport-helper-error-reporting (2013-04-25) 10 commits

566 +* fc/transport-helper-error-reporting (2013-05-10) 12 commits

567 + (merged to 'next' on 2013-05-10 at 2a9af4b)

568 + + transport-helper: fix remote helper namespace regression

569 + + test: remote-helper: add missing and

570 (merged to 'next' on 2013-04-25 at 3358f1a)

571 + t5801: "VAR=VAL shell_func args" is forbidden

572 (merged to 'next' on 2013-04-22 at 5ba6467)

@@ -427,12 +579,12 @@ of the repositories listed at

579 + transport-helper: improve push messages

580 + transport-helper: mention helper name when it dies

581 + transport-helper: report errors properly

430 - (this branch is used by js/transport-helper-error-reporting-fix.)

582 + (this branch is tangled with js/transport-helper-error-reporting-fix.)

583

584 Update transport helper to report errors and maintain ref hierarchy

585 used to keep track of remote helper state better.

586

435 - Will cook in 'next', but may be 1.8.3 material depending on how things go.

587 + Will cook in 'next'.

588

589

590 * jk/submodule-subdirectory-ok (2013-04-24) 3 commits

@@ -564,3 +716,21 @@ of the repositories listed at

716 aka "--no-ignore-removal".

717

718 Will cook in 'next' until Git 2.0.

719 +

720 +--------------------------------------------------

721 +[Discarded]

722 +

723 +* kb/full-history-compute-treesame-carefully (2013-05-06) 11 commits

724 + . revision.c: treat A...B merge bases as if manually specified

725 + . revision.c: discount side branches when computing TREESAME

726 + . simplify-merges: drop merge from irrelevant side branch

727 + . simplify-merges: never remove all TREESAME parents

728 + . t6012: update test for tweaked full-history traversal

729 + . revision.c: Make --full-history consider more merges

730 + . rev-list-options.txt: correct TREESAME for P

731 + . t6111: allow checking the parents as well

732 + . t6111: new TREESAME test set

733 + . t6019: test file dropped in -s ours merge

734 + . decorate.c: compact table when growing

735 +

736 + Superseded by kb/full-history-compute-treesame-carefully-2

What's cooking (2013/05 #04)