7bbd09786d · matheusbsbs/matheus-bagliano

@@ -1,9 +1,9 @@

1 To: git@vger.kernel.org

2 -Subject: What's cooking in git.git (Feb 2010, #05; Sun, 21)

3 -X-master-at: 241b9254e1c8ff071d8054f8b6fbe1883b389d69

4 -X-next-at: d52c051f421da80f7dc5fed68a7046d70d06d008

2 +Subject: What's cooking in git.git (Feb 2010, #06; Tue, 23)

3 +X-master-at: 0901d5a2ef80996cf64c8afeaec765e1bc09f795

4 +X-next-at: ac4ec8fb00b9defe6763ade89af32f420a97b5ed

5

6 -What's cooking in git.git (Feb 2010, #05; Sun, 21)

6 +What's cooking in git.git (Feb 2010, #06; Tue, 23)

7 --------------------------------------------------

8

9 Here are the topics that have been cooking. Commits prefixed with '-' are

@@ -12,89 +12,167 @@ marked with '.' do not appear in any of the integration branches, but I am

12 still holding onto them.

13

14 --------------------------------------------------

15 -[Graduated to "master"]

16 -

17 -* np/fast-import-idx-v2 (2010-02-17) 6 commits

18 - + fast-import: use the diff_delta() max_delta_size argument

19 - + fast-import: honor pack.indexversion and pack.packsizelimit config vars

20 - + fast-import: make default pack size unlimited

21 - + fast-import: use write_idx_file() instead of custom code

22 - + fast-import: use sha1write() for pack data

23 - + fast-import: start using struct pack_idx_entry

24 -

25 -* ml/maint-grep-doc (2010-02-15) 1 commit

26 - (merged to 'next' on 2010-02-16 at 4059a38)

27 - + grep documentation: clarify what files match

28 -

29 -* tc/maint-transport-ls-remote-with-void (2010-02-16) 1 commit

30 - (merged to 'next' on 2010-02-16 at e6ef1a8)

31 - + transport: add got_remote_refs flag

32 -

33 -* hm/maint-imap-send-crlf (2010-02-12) 1 commit

34 - (merged to 'next' on 2010-02-17 at c6162cb)

35 - + git-imap-send: Convert LF to CRLF before storing patch to draft box

36 -

37 -* sp/maint-push-sideband (2010-02-10) 8 commits

38 - (merged to 'next' on 2010-02-16 at 6f19e5b)

39 - + receive-pack: Send internal errors over side-band #2

40 - + t5401: Use a bare repository for the remote peer

41 - + receive-pack: Send hook output over side band #2

42 - + receive-pack: Wrap status reports inside side-band-64k

43 - + receive-pack: Refactor how capabilities are shown to the client

44 - + send-pack: demultiplex a sideband stream with status data

45 - + run-command: support custom fd-set in async

46 - + run-command: Allow stderr to be a caller supplied pipe

47 - (this branch is used by sp/push-sideband.)

48 -

49 -Based on 1.6.5 maintenance track, later to be merged to 1.6.X maintenance

50 -series if needed.

51 -

52 -* sp/push-sideband (2010-02-10) 0 commits

53 - (this branch uses sp/maint-push-sideband.)

54 -

55 -* jc/checkout-detached (2010-01-29) 1 commit

56 - (merged to 'next' on 2010-02-17 at 7e03edc)

57 - + Reword "detached HEAD" notification

58 -

59 -* jc/maint-fix-test-perm (2010-01-30) 2 commits

60 - (merged to 'next' on 2010-02-16 at 9d2e037)

61 - + lib-patch-mode.sh: Fix permission

62 - + t6000lib: Fix permission

63 -

64 -* jn/makefile-script-lib (2010-01-31) 1 commit

65 - (merged to 'next' on 2010-02-16 at f5334f5)

66 - + Do not install shell libraries executable

67 -

68 -* mv/request-pull-modernize (2010-01-29) 1 commit

69 - (merged to 'next' on 2010-02-16 at be03aad)

70 - + request-pull: avoid mentioning that the start point is a single commit

15 +[New Topics]

16 +

17 +* gb/submodule-env (2010-02-23) 2 commits

18 + - submodules: ensure clean environment when operating in a submodule

19 + - shell setup: clear_local_git_env() function

20 +

21 +This is a round before the round with hardcoded list on the C side. I

22 +think the idea of the new one is basically sound, but I do not think it

23 +should create a new source file nor call it cache.c which obviously this

24 +thing is not (environment.c might be a better home for it).

25 +

26 +* jl/maint-submodule-clear-env (2010-02-23) 1 commit

27 + - is_submodule_modified(): clear environment properly

28 +

29 +I'd prefer to rebuild this on top of rerolled gb/submodule-env before

30 +moving it forward.

31 +

32 +* il/loosen-remote-helper-names (2010-02-23) 1 commit

33 + - Allow '+', '-' and '.' in remote helper names

34 +

35 +Judging from the discussion, the conclusion seemed to be that Dscho was

36 +needlessly worried about using '+' in pathnames on DOS, and the patch is

37 +Ok. There also was a discussion on how to sanely deal with the need to

38 +have duplicated binaries with the current set-up, only to handle

39 +essentially the same svn+ssh:// and ssh+svn:// protocols; I however think

40 +that is pretty much orthogonal to the issue of what letters can be used in

41 +names.

42 +

43 +* jk/maint-add--interactive-delete (2010-02-22) 1 commit

44 + - add-interactive: fix bogus diff header line ordering

45 +

46 +Looked Ok for 'next'.

47 +

48 +* js/runtime-prefix-trace-not-warn (2010-02-23) 1 commit

49 + - Print RUNTIME_PREFIX warning only when GIT_TRACE is set

50 +

51 +Looked Ok for 'next'.

52 +

53 +* lt/deepen-builtin-source (2010-02-22) 1 commit

54 + - Move 'builtin-*' into a 'builtin/' subdirectory

55 +

56 +This is a painful one to keep out of 'master' for a long time, as any

57 +topic with new builtin commands will need evil merges to adjust to it.

58 +

59 +* tc/http-cleanup (2010-02-21) 7 commits

60 + - remote-curl: init walker only when needed

61 + - remote-curl: use http_fetch_ref() instead of walker wrapper

62 + - http: init and cleanup separately from http-walker

63 + - http-walker: cleanup more thoroughly

64 + - http-push: remove useless condition

65 + - t554[01]-http-push: refactor, add non-ff tests

66 + - t5541-http-push: check that ref is unchanged for non-ff test

67 +

68 +Didn't look; there was some discussion that may result in a reroll.

69 +

70 +* tr/notes-display (2010-02-23) 11 commits

71 + - notes: add shorthand --ref to override GIT_NOTES_REF

72 + - commit --amend: copy notes to the new commit

73 + - rebase: support automatic notes copying

74 + - notes: implement helpers needed for note copying during rewrite

75 + - notes: implement 'git notes copy --stdin'

76 + - rebase -i: invoke post-rewrite hook

77 + - rebase: invoke post-rewrite hook

78 + - commit --amend: invoke post-rewrite hook

79 + - Documentation: document post-rewrite hook

80 + - Support showing notes from more than one notes tree

81 + - test-lib: unset GIT_NOTES_REF to stop it from influencing tests

82 + (this branch uses jh/notes.)

83 +

84 +Didn't look too carefully except for the second one.

85 +

86 +* ts/svn-mergeinfo (2010-02-22) 3 commits

87 + - git-svn: Fix discarding of extra parents from svn:mergeinfo

88 + - t9151: Add two new svn:mergeinfo test cases

89 + - t9151: Fix a few commits in the SVN dump

90 +

91 +I won't pick up git-svn patches myself without Acks from Eric and/or Sam;

92 +there are here only so that they won't be forgotten.

93 +

94 +* cw/test-lib-relicense (2010-02-22) 1 commit

95 + . test-lib.sh: Add explicit license detail, with change from GPLv2 to GPLv2+.

96 +

97 +I think a reasonable way forward would be to collect Acks from everybody

98 +who appears on "git blame" or "git shortlog" output. How do people feel

99 +about relicensing the whole (except for the obviously borrowed bits, such

100 +as xdiff/) under GPLv2 _or later_? I am not a big fan of GPLv3 and do not

101 +see a compelling reason to switch to v3 for the sake of v3 license per-se,

102 +but helping others that want to use our code might tip the balance.

103

104 --------------------------------------------------

73 -[New Topics]

105 +[Stalled]

106

75 -* dp/read-not-mmap-small-loose-object (2010-02-21) 1 commit

76 - - hash-object: don't use mmap() for small files

107 +* ld/push-porcelain (2010-02-09) 4 commits

108 + - git-push: fix an error message so it goes to stderr

109 + - git-push: make git push --dry-run --porcelain exit with status 0 even if updates will be rejected

110 + - git-push: send "To <remoteurl>" messages to the standard output in --porcelain mode

111 + - git-push: squelch advice message if in --porcelain mode

112 +

113 +This needs further simplification, judging from the previous discussion?

114 +

115 +* sd/format-patch-to (2010-02-17) 1 commit

116 + - Add 'git format-patch --to=' option and 'format.to' configuration variable.

117

78 -I treaked the cut-off based on my reading of Dmitry's numbers.

118 +Shouldn't be too hard to add tests to t4014; other than that looked ready

119 +for 'next'.

120

80 -* np/compress-loose-object-memsave (2010-02-20) 1 commit

81 - . sha1_file: don't malloc the whole compressed result when writing out objects

121 +* sd/init-template (2010-02-17) 2 commits

122 + - Add a "TEMPLATE DIRECTORY" section to git-init[1].

123 + - Add `init.templatedir` configuration variable.

124

83 -* jc/maint-add-paranoid (2010-02-19) 2 commits

84 - - paranoid: avoid unnecessary re-hashing

85 - - Teach "git add" and friends to be paranoid

125 +Shouldn't be too hard to add tests to t0001; other than that looked ready

126 +for 'next'.

127

87 -This conflicts rather badly with more logical change by Nicolas in the

88 -other thread; perhaps we should drop this "paranoia" for now.

128 +* sd/log-decorate (2010-02-17) 3 commits

129 + - log.decorate: usability fixes

130 + - Add `log.decorate' configuration variable.

131 + - git_config_maybe_bool()

132 +

133 +Probably ready for 'next', except that people need to be warned about

134 +having to update their scripts to explicitly pass --no-decorate to keep

135 +them working. A good idea to disable this when --pretty was given, just

136 +like notes are disabled by default, was floated.

137 +

138 +* pb/log-first-parent-p-m (2010-02-10) 1 commit

139 + (merged to 'next' on 2010-02-17 at 2f8e5ae)

140 + + git log -p -m: document -m and honor --first-parent

141 +

142 +Needs tests but otherwise looked fine. We might want to teach "-m trumps

143 +implicit --cc" to "git show", but that is a totally separate topic.

144 +

145 +I actually care about this "log -p --first-parent" very much, but if Pasky

146 +is counting on that and procrastinating until I write the tests myself, he

147 +is in for a dissapointment. I don't have that much free time these days.

148 +Help is appreciated.

149 +

150 +* js/rebase-origin-x (2010-02-05) 1 commit

151 + - [RFC w/o test and incomplete] rebase: add -x option to record original commit name

152 +

153 +I retract my objection against the idea of -x; needs polishing before

154 +moving forward.

155 +

156 +--------------------------------------------------

157 +[Cooking]

158 +

159 +* dp/read-not-mmap-small-loose-object (2010-02-21) 1 commit

160 + (merged to 'next' on 2010-02-21 at fa39a9a)

161 + + hash-object: don't use mmap() for small files

162 +

163 +Will merge to 'master' after a few more days of cooking.

164 +

165 +* np/compress-loose-object-memsave (2010-02-21) 2 commits

166 + (merged to 'next' on 2010-02-21 at 1e558d6)

167 + + sha1_file: be paranoid when creating loose objects

168 + + sha1_file: don't malloc the whole compressed result when writing out objects

169 +

170 +Will merge to 'master' after a few more days of cooking.

171

172 * jc/maint-fix-mailinfo-strip (2010-02-19) 1 commit

173 - mailinfo: do not strip leading spaces even for a header line

174

93 -Linus noticed that an indented first line in the log message body loses

94 -its indentation. Lukas sent a lot more intrusive patch to keep the

95 -removal of leading spaces for header lines while avoiding this issue, but

96 -I think removing the leading blanks is wrong in either the message or

97 -in-body header.

175 +Ready for 'next', I think.

176

177 * ne/pack-local-doc (2010-02-18) 1 commit

178 - Documentation: pack-objects: Clarify --local's semantics.

@@ -102,26 +180,34 @@ in-body header.

180 Comments from pack experts?

181

182 * ml/connect-refactor (2010-02-17) 1 commit

105 - - connect.c: move duplicated code to a new function 'get_host_and_port'

183 + (merged to 'next' on 2010-02-21 at 7361651)

184 + + connect.c: move duplicated code to a new function 'get_host_and_port'

185

107 -* ml/encode-header-refactor (2010-02-16) 1 commit

108 - - refactor duplicated encode_header in pack-objects and fast-import

186 +* ml/encode-header-refactor (2010-02-23) 2 commits

187 + (merged to 'next' on 2010-02-23 at ac4ec8f)

188 + + move encode_in_pack_object_header() to a better place

189 + (merged to 'next' on 2010-02-21 at efe648b)

190 + + refactor duplicated encode_header in pack-objects and fast-import

191

192 * ml/fill-mm-refactor (2010-02-16) 1 commit

111 - - refactor duplicated fill_mm() in checkout and merge-recursive

193 + (merged to 'next' on 2010-02-21 at 2fc5570)

194 + + refactor duplicated fill_mm() in checkout and merge-recursive

195 +

196 +* ml/send-pack-transport-refactor (2010-02-16) 1 commit

197 + (merged to 'next' on 2010-02-21 at db276f4)

198 + + refactor duplicated code in builtin-send-pack.c and transport.c

199

113 -These three should be safe for 'next' but I postponed them because they

114 -were distractions compared to other topics I wanted to proccess first.

200 +Will merge to 'master' soonish.

201

116 -* mm/mkstemps-mode-for-packfiles (2010-02-20) 6 commits

202 +* mm/mkstemps-mode-for-packfiles (2010-02-22) 6 commits

203 - Use git_mkstemp_mode instead of plain mkstemp to create object files

118 - - git_mkstemps_mode: do not overwrite errno

204 + - git_mkstemps_mode: don't set errno to EINVAL on exit.

205 - Use git_mkstemp_mode and xmkstemp_mode in odb_mkstemp, not chmod later.

206 - git_mkstemp_mode, xmkstemp_mode: variants of gitmkstemps with mode argument.

207 - Move gitmkstemps to path.c

208 - Add a testcase for ACL with restrictive umask.

209

124 -The test does not seem to pass for me.

210 +Looked Ok for 'next'.

211

212 * rs/optim-text-wrap (2010-02-19) 4 commits

213 (merged to 'next' on 2010-02-21 at 70ef189)

@@ -130,10 +216,14 @@ The test does not seem to pass for me.

216 + utf8.c: remove print_spaces()

217 + utf8.c: remove print_wrapped_text()

218

219 +Will merge to 'master' soonish.

220 +

221 * tr/maint-cherry-pick-list (2010-02-20) 1 commit

222 (merged to 'next' on 2010-02-21 at 65fded0)

223 + cherry_pick_list: quit early if one side is empty

224

225 +Will merge to 'master' soonish, and backmerge to 'maint'.

226 +

227 * tc/transport-verbosity (2010-02-18) 9 commits

228 - transport: update flags to be in running order

229 - pull: learn --progress

@@ -145,74 +235,21 @@ The test does not seem to pass for me.

235 - fetch: refactor verbosity option handling into transport.[ch]

236 - Documentation/git-push.txt: put --quiet before --verbose

237

148 ---------------------------------------------------

149 -[Stalled]

150 -

151 -* ld/push-porcelain (2010-02-09) 4 commits

152 - - git-push: fix an error message so it goes to stderr

153 - - git-push: make git push --dry-run --porcelain exit with status 0 even if updates will be rejected

154 - - git-push: send "To <remoteurl>" messages to the standard output in --porcelain mode

155 - - git-push: squelch advice message if in --porcelain mode

156 -

157 -This needs further simplification, judging from the previous discussion?

158 -

159 -* ld/maint-diff-quiet-w (2010-02-16) 1 commit

160 - - git diff --quiet -w: check and report the status

161 -

162 -Needs tests but otherwise looked Ok.

163 -

164 -* sd/format-patch-to (2010-02-17) 1 commit

165 - - Add 'git format-patch --to=' option and 'format.to' configuration variable.

166 -

167 -Shouldn't be too hard to add tests to t4014; other than that looked ready

168 -for 'next'.

169 -

170 -* sd/init-template (2010-02-17) 2 commits

171 - - Add a "TEMPLATE DIRECTORY" section to git-init[1].

172 - - Add `init.templatedir` configuration variable.

238 +Didn't look very carefully. Comments from transport people are very much

239 +appreciated before moving this forward.

240

174 -Shouldn't be too hard to add tests to t0001; other than that looked ready

175 -for 'next'.

241 +* ld/maint-diff-quiet-w (2010-02-21) 2 commits

242 + (merged to 'next' on 2010-02-21 at 4701142)

243 + + git-diff: add a test for git diff --quiet -w

244 + + git diff --quiet -w: check and report the status

245

177 -* sd/log-decorate (2010-02-17) 3 commits

178 - - log.decorate: usability fixes

179 - - Add `log.decorate' configuration variable.

180 - - git_config_maybe_bool()

181 -

182 -Probably ready for 'next', except that people need to be warned about

183 -having to update their scripts to explicitly pass --no-decorate to keep

184 -them working.

185 -

186 -* pb/log-first-parent-p-m (2010-02-10) 1 commit

187 - (merged to 'next' on 2010-02-17 at 2f8e5ae)

188 - + git log -p -m: document -m and honor --first-parent

189 -

190 -Needs tests but otherwise looked fine. We might want to teach "-m trumps

191 -implicit --cc" to "git show", but that is a totally separate topic.

192 -

193 -* js/rebase-origin-x (2010-02-05) 1 commit

194 - - [RFC w/o test and incomplete] rebase: add -x option to record original commit name

195 -

196 -I retract my objection against the idea of -x; needs polishing before

197 -moving forward.

198 -

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

200 -[Cooking]

246 +Will merge to 'master' after a few more days of cooking.

247

248 * cp/add-u-pathspec (2010-02-09) 2 commits

249 - test for add with non-existent pathspec

250 - git add -u: die on unmatched pathspec

251

206 -I am a bit torn on this one. Traditionally we never complained on

207 -unmatched pathspec when talking about tracked files. If we were to go

208 -this route, I think we should probably enhance the "run_diff_files" and

209 -friends in such a way that they mark matched pathspecs, in a way similar

210 -to match_pathspec() in dir.c does, and report unmatched ones based on

211 -that result, instead of adding an extra pass to scan the index. The same

212 -goes for pathspec_matches() in builtin-grep.c

213 -

214 -Incidentally, I've proposed "pathspec unification" as possible GSoC'10

215 -project---with luck, we might finally see a progress on this front ;-)

252 +Will merge to 'next'.

253

254 * jc/for-each-ref (2010-02-13) 4 commits

255 (merged to 'next' on 2010-02-21 at c9a6c2f)

@@ -221,11 +258,17 @@ project---with luck, we might finally see a progress on this front ;-)

258 + builtin-for-each-ref.c: check if we need to peel onion while parsing the format

259 + builtin-for-each-ref.c: comment fixes

260

261 +Will merge to 'master' soonish.

262 +

263 * jn/gitweb-config-error-die (2010-02-14) 1 commit

264 (merged to 'next' on 2010-02-21 at e3ecd65)

265 + gitweb: Die if there are parsing errors in config file

266

228 -* jn/maint-fix-pager (2010-02-20) 7 commits

267 +Will merge to 'master' soonish.

268 +

269 +* jn/maint-fix-pager (2010-02-22) 8 commits

270 + (merged to 'next' on 2010-02-23 at 575e0e4)

271 + + tests: Fix race condition in t7006-pager

272 (merged to 'next' on 2010-02-21 at 640e10c)

273 + t7006-pager: if stdout is not a terminal, make a new one

274 + tests: Add tests for automatic use of pager

@@ -235,19 +278,26 @@ project---with luck, we might finally see a progress on this front ;-)

278 + Make 'git var GIT_PAGER' always print the configured pager

279 + Fix 'git var' usage synopsis

280

281 +Will merge to 'master' after a few more days of cooking.

282 +

283 * ml/color-when (2010-02-16) 1 commit

284 (merged to 'next' on 2010-02-21 at d52c051)

285 + Add an optional argument for --color options

286

287 +Will merge to 'master' soonish.

288 +

289 * hm/imap-send-cram-md5 (2010-02-15) 1 commit

243 - - imap-send: support CRAM-MD5 authentication

290 + (merged to 'next' on 2010-02-21 at de8f650)

291 + + imap-send: support CRAM-MD5 authentication

292

245 -A potential clean-up sent as a counter-proposal; waiting for response.

293 +Will merge to 'master' after a few more days of cooking.

294

295 * jh/maint-submodule-status-in-void (2010-02-16) 1 commit

296 (merged to 'next' on 2010-02-21 at 2e605c3)

297 + submodule summary: Don't barf when invoked in an empty repo

298

299 +Will merge to 'master' after a few more days of cooking.

300 +

301 * bg/apply-blank-at-eof (2010-02-17) 3 commits

302 - t3417: Add test cases for "rebase --whitespace=fix"

303 - t4124: Add additional tests of --whitespace=fix

@@ -259,12 +309,13 @@ RFC.

309 (merged to 'next' on 2010-02-21 at aca55e6)

310 + require_work_tree broken with NONGIT_OK

311

262 -* ml/send-pack-transport-refactor (2010-02-16) 1 commit

263 - (merged to 'next' on 2010-02-21 at db276f4)

264 - + refactor duplicated code in builtin-send-pack.c and transport.c

312 +Will merge to 'master' soonish, and backmerge to 'maint'.

313

314 * jc/maint-status-preload (2010-02-17) 1 commit

267 - - status: preload index to optimize lstat(2) calls

315 + (merged to 'next' on 2010-02-21 at d79e163)

316 + + status: preload index to optimize lstat(2) calls

317 +

318 +Will merge to 'master' soonish, and backmerge to 'maint'.

319

320 * nd/root-git (2010-02-14) 5 commits

321 - Add test for using Git at root of file system

@@ -273,6 +324,8 @@ RFC.

324 - init-db, rev-parse --git-dir: do not append redundant slash

325 - make_absolute_path(): Do not append redundant slash

326

327 +Will merge to 'next' unless I hear objections.

328 +

329 * ac/cvsimport-revision-mapping (2010-02-06) 1 commit

330 (merged to 'next' on 2010-02-17 at 6756446)

331 + cvsimport: new -R option: generate .git/cvs-revisions mapping

@@ -307,6 +360,8 @@ This is to help merging the clean-up to "master".

360 + Makefile: transport.o depends on branch.h now

361 (this branch uses jn/maint-makedepend and jn/master-makedepend.)

362

363 +Will merge to 'master' after a few more days of cooking.

364 +

365 * jh/notes (2010-02-13) 30 commits

366 (merged to 'next' on 2010-02-21 at 75fc451)

367 + builtin-notes: Add "copy" subcommand for copying notes between objects

@@ -339,11 +394,16 @@ This is to help merging the clean-up to "master".

394 + Add tests for checking correct handling of $GIT_NOTES_REF and core.notesRef

395 + Notes API: get_commit_notes() -> format_note() + remove the commit restriction

396 + Minor cosmetic fixes to notes.c

397 + (this branch is used by tr/notes-display.)

398 +

399 +Will merge to 'master' after a few more days of cooking.

400

401 * jc/grep-author-all-match-implicit (2010-01-17) 1 commit

402 (merged to 'next' on 2010-02-17 at 3b7be80)

403 + "log --author=me --grep=it" should find intersection, not union

404

405 +Will merge to 'master' after a few more days of cooking.

406 +

407 * jh/gitweb-caching (2010-01-30) 1 commit

408 - gitweb: Add an option to force version match

409

What's cooking (2010/02 #6)