What's cooking (2009/10 #03)

Junio C Hamano committed Oct 19, 2009 at 01:03 UTC e1854702c6450ba4023c183aef887e0877aa4774
1 file changed +184 -201
whats-cooking.txt
+184 -201
@@ -1,9 +1,9 @@
1 To: git@vger.kernel.org
2 -Subject: What's cooking in git.git (Oct 2009, #02; Sun, 11)
3 -X-master-at: 78d553b7d7b269bb22ebd8b1198657c37484a3a0
4 -X-next-at: 0ff2e96835a8890f86b5b69d8967d350fe89e9e8
2 +Subject: What's cooking in git.git (Oct 2009, #03; Mon, 19)
3 +X-master-at: e79999b1a285d4dcf1d84bc893c864516a390cfa
4 +X-next-at: e01f7ce260cd5923cc9a088b74793a58379b5cc0
5
6 -What's cooking in git.git (Oct 2009, #02; Sun, 11)
6 +What's cooking in git.git (Oct 2009, #03; Mon, 19)
7 --------------------------------------------------
8
9 Here are the topics that have been cooking. Commits prefixed with '-' are
@@ -51,58 +51,134 @@ backward compatible. Currently planeed changes are:
51
52 http://thread.gmane.org/gmane.comp.version-control.git/119731/focus=119751
53
54 -As a post-release ritual, I've rebuilt 'next' on top of 'master'.
55 -
54 --------------------------------------------------
55 [Graduated to "master"]
56
59 -* ch/am-header (2009-09-25) 2 commits
60 - (merged to 'next' on 2009-09-25 at f86e197)
61 - + git-am: force egrep to use correct characters set
62 - + git-am: fixed patch_format detection according to RFC2822
57 +* jp/maint-send-email-fold (2009-10-08) 1 commit.
58 + + git-send-email.perl: fold multiple entry "Cc:" and multiple single line "RCPT TO:"s
59 +
60 +* pv/maint-add-p-no-exclude (2009-10-10) 1 commit.
61 + (merged to 'next' on 2009-10-11 at 0ff2e96)
62 + + git-add--interactive: never skip files included in index
63 +
64 +* dk/blame-el (2009-09-29) 1 commit.
65 + + git-blame.el: Change how blame information is shown.
66 +
67 +* jn/gitweb-patch (2009-10-09) 1 commit.
68 + + gitweb: Do not show 'patch' link for merge commits
69 +
70 +* mr/instaweb-cgid (2009-09-26) 1 commit.
71 + (merged to 'next' on 2009-10-11 at 99bcfaf)
72 + + instaweb: support mod_cgid for apache2
73
64 -* ef/msvc-noreturn (2009-09-30) 2 commits
65 - (merged to 'next' on 2009-10-07 at 66137a0)
66 - + add NORETURN_PTR for function pointers
67 - + increase portability of NORETURN declarations
74 +* tf/doc-pt-br (2009-09-23) 1 commit.
75 + + Documentation: update pt-BR
76
69 -jk: This is the latest round and I think should be ready for at least
70 -'next' (maybe even 'master' as it is really about the build and not about
71 -functionality).
77 +The current AsciiDoc may barf on NOME and SINOPSE, as pt_BR language
78 +definition is not widely distributed yet (it just hit the development
79 +tree).
80
73 -* fc/mutt-alias (2009-09-30) 1 commit
74 - (merged to 'next' on 2009-10-07 at df7ac20)
75 - + send-email: fix mutt regex for grouped aliases
81 +* bg/rebase-reword (2009-10-07) 2 commits.
82 + + rebase-i: fix reword when using a terminal editor
83 + + Teach 'rebase -i' the command "reword"
84
77 -jk: Latest round that addressed comments. Ready for 'next' if not
78 -'master'.
85 +* jp/fetch-tag-match (2009-09-17) 1 commit.
86 + + fetch: Speed up fetch by rewriting find_non_local_tags
87 +
88 +* jn/gitweb-show-size (2009-09-07) 1 commit.
89 + + gitweb: Add 'show-sizes' feature to show blob sizes in tree view
90 +
91 +* jc/maint-1.6.0-blank-at-eof (2009-09-14) 15 commits.
92 + (merged to 'next' on 2009-10-11 at 7f34fb3)
93 + + diff -B: colour whitespace errors
94 + + diff.c: emit_add_line() takes only the rest of the line
95 + + diff.c: split emit_line() from the first char and the rest of the line
96 + + diff.c: shuffling code around
97 + + diff --whitespace: fix blank lines at end
98 + + core.whitespace: split trailing-space into blank-at-{eol,eof}
99 + + diff --color: color blank-at-eof
100 + + diff --whitespace=warn/error: fix blank-at-eof check
101 + + diff --whitespace=warn/error: obey blank-at-eof
102 + + diff.c: the builtin_diff() deals with only two-file comparison
103 + + apply --whitespace: warn blank but not necessarily empty lines at EOF
104 + + apply --whitespace=warn/error: diagnose blank at EOF
105 + + apply.c: split check_whitespace() into two
106 + + apply --whitespace=fix: detect new blank lines at eof correctly
107 + + apply --whitespace=fix: fix handling of blank lines at the eof
108 + (this branch is used by jc/maint-blank-at-eof.)
109
80 -* jk/reflog-date (2009-09-24) 1 commit
81 - (merged to 'next' on 2009-09-29 at 43d444a)
82 - + improve reflog date/number heuristic
110 +This is a fix for an ancient bug (or inconsistent set of features); the
111 +topic is based on an ancient codebase and is designed to be merged
112 +upwards. jc/maint-blank-at-eof serves that purpose.
113
114 --------------------------------------------------
115 [New Topics]
116
87 -* jp/maint-send-email-fold (2009-10-08) 1 commit.
88 - - git-send-email.perl: fold multiple entry "Cc:" and multiple single line "RCPT TO:"s
117 +* ak/bisect-reset-to-switch (2009-10-13) 1 commit.
118 + - bisect reset: Allow resetting to any commit, not just a branch
119
90 -* ks/precompute-completion (2009-10-05) 1 commit.
91 - - Speedup bash completion loading
120 +* fc/doc-fast-forward (2009-10-11) 1 commit.
121 + - user-manual: use 'fast-forward'
122
93 -* pv/maint-add-p-no-exclude (2009-10-10) 1 commit.
94 - (merged to 'next' on 2009-10-11 at 0ff2e96)
95 - + git-add--interactive: never skip files included in index
123 +* gb/maint-gitweb-esc-param (2009-10-13) 1 commit.
124 + (merged to 'next' on 2009-10-14 at 105f997)
125 + + gitweb: fix esc_param
126 + (this branch is used by sb/gitweb-link-author.)
127
97 -* sp/smart-http (2009-10-08) 4 commits.
98 - - Smart fetch and push over HTTP: server side
99 - - Add smart-http options to upload-pack, receive-pack
100 - - Git-aware CGI to provide dumb HTTP transport
101 - - Document the HTTP transport protocol
128 +* jc/maint-1.6.3-graft-trailing-space (2009-10-14) 1 commit.
129 + - info/grafts: allow trailing whitespaces at the end of line
130 +
131 +* jc/strbuf-nested-expand (2009-10-18) 5 commits
132 + - Teach --wrap to only indent without wrapping
133 + - Add %[wrap(width,in1,in2)<<any-string>>%] implementation
134 + - strbuf_nested_expand(): allow expansion to interrupt in the middle
135 + - Add strbuf_add_wrapped_text() to utf8.[ch]
136 + - print_wrapped_text(): allow hard newlines
137 + (this branch uses js/log-rewrap.)
138 +
139 +* jk/maint-cvsimport-pathname (2009-10-19) 1 commit.
140 + (merged to 'next' on 2009-10-19 at 77824f2)
141 + + cvsimport: fix relative argument filenames
142 +
143 +* jn/show-normalized-refs (2009-10-12) 3 commits.
144 + - check-ref-format: simplify --print implementation
145 + - git check-ref-format --print
146 + - Add tests for git check-ref-format
147 +
148 +* sb/gitweb-link-author (2009-10-14) 1 commit.
149 + - gitweb: linkify author/committer names with search
150 + (this branch uses gb/maint-gitweb-esc-param.)
151 +
152 +* jc/checkout-auto-track (2009-10-18) 3 commits
153 + - git checkout --nodwim
154 + - DWIM "git checkout frotz" to "git checkout -b frotz origin/frotz"
155 + - check_filename(): make verify_filename() callable without dying
156
157 --------------------------------------------------
158 [Stalled]
159
160 +* db/vcs-helper-rest (2009-09-03) 6 commits.
161 + - Allow helpers to report in "list" command that the ref is unchanged
162 + - Add support for "import" helper command
163 + - Add a config option for remotes to specify a foreign vcs
164 + - Allow programs to not depend on remotes having urls
165 + - Allow fetch to modify refs
166 + - Use a function to determine whether a remote is valid
167 + (this branch is used by jh/cvs-helper.)
168 +
169 +This holds the remainder of the db/vcs-helper topic that has already
170 +merged in 1.6.5. If people want to replace this with improvements it
171 +would be a good time to do so.
172 +
173 +* jl/submodule-add-noname (2009-09-22) 1 commit.
174 + - git submodule add: make the <path> parameter optional
175 +
176 +Dscho started an interesting discussion regarding the larger workflow in
177 +which the "submodule add" is used. I think the patch itself makes sense
178 +but at the same time it probably makes sense to also take the <path> and
179 +infer the <repository> as Dscho suggested, probably in "git submodule
180 +add", not in "git add" proper, at least initially.
181 +
182 * sr/gfi-options (2009-09-06) 6 commits.
183 - fast-import: test the new option command
184 - fast-import: add option command
@@ -110,6 +186,7 @@ jk: Latest round that addressed comments. Ready for 'next' if not
186 - fast-import: add feature command
187 - fast-import: put marks reading in it's own function
188 - fast-import: put option parsing code in separate functions
189 + (this branch is used by jh/notes.)
190
191 Ejected from 'next' as fast-import folks still seem to be discussing with
192 how to proceed.
@@ -135,28 +212,44 @@ says "no"?)
212 - Allow helpers to request marks for fast-import
213 (this branch uses db/vcs-helper-rest.)
214
138 -Builds on db/vcs-helper. There is a re-roll planned.
139 -
140 -* ne/rev-cache (2009-09-07) 7 commits.
141 - . support for commit grafts, slight change to general mechanism
142 - . support for path name caching in rev-cache
143 - . full integration of rev-cache into git, completed test suite
144 - . administrative functions for rev-cache, start of integration into git
145 - . support for non-commit object caching in rev-cache
146 - . basic revision cache system, no integration or features
147 - . man page and technical discussion for rev-cache
148 -
149 -This seems to break at least t6001 when merged to 'pu'.
215 +Builds on db/vcs-helper.
216
217 --------------------------------------------------
218 [Cooking]
219
154 -* dk/blame-el (2009-09-29) 1 commit.
155 - - git-blame.el: Change how blame information is shown.
220 +* tr/stash-format (2009-10-16) 5 commits
221 + . stash list: drop the default limit of 10 stashes
222 + . stash list: use new %g formats instead of sed
223 + . Introduce new pretty formats %g[sdD] for reflog information
224 + . reflog-walk: refactor the branch@{num} formatting
225 + . Refactor pretty_print_commit arguments into a struct
226
157 -Soon in 'next'.
227 +Expecting a re-roll.
228
159 -* ef/msys-imap (2009-10-09) 8 commits.
229 +* ks/precompute-completion (2009-10-05) 1 commit.
230 + (merged to 'next' on 2009-10-14 at adf722a)
231 + + Speedup bash completion loading
232 +
233 +* sp/smart-http (2009-10-14) 17 commits
234 + - Smart HTTP fetch: gzip requests
235 + - Smart fetch over HTTP: client side
236 + - Smart push over HTTP: client side
237 + - Discover refs via smart HTTP server when available
238 + - Smart fetch and push over HTTP: server side
239 + - Add stateless RPC options to upload-pack, receive-pack
240 + - Git-aware CGI to provide dumb HTTP transport
241 + - Move WebDAV HTTP push under remote-curl
242 + - remote-helpers: Support custom transport options
243 + - remote-helpers: Fetch more than one ref in a batch
244 + - fetch: Allow transport -v -v -v to set verbosity to 3
245 + - remote-curl: Refactor walker initialization
246 + - Add multi_ack_detailed capability to fetch-pack/upload-pack
247 + - Move "get_ack()" back to fetch-pack
248 + - fetch-pack: Use a strbuf to compose the want list
249 + - pkt-line: Make packet_read_line easier to debug
250 + - pkt-line: Add strbuf based functions
251 +
252 +* ef/msys-imap (2009-10-13) 8 commits.
253 - MSVC: Enable OpenSSL, and translate -lcrypto
254 - mingw: enable OpenSSL
255 - mingw: wrap SSL_set_(w|r)fd to call _get_osfhandle
@@ -166,81 +259,43 @@ Soon in 'next'.
259 - imap-send: use separate read and write fds
260 - imap-send: remove useless uid code
261
169 -* jn/gitweb-patch (2009-10-09) 1 commit.
170 - - gitweb: Do not show 'patch' link for merge commits
171 -
172 -Soon in 'next'
262 +Is this good to go yet?
263
264 * mr/gitweb-snapshot (2009-09-26) 2 commits.
265 - gitweb: append short hash ids to snapshot files
266 (merged to 'next' on 2009-10-11 at 22ba047)
267 + gitweb: check given hash before trying to create snapshot
268
179 -* mr/instaweb-cgid (2009-09-26) 1 commit.
180 - (merged to 'next' on 2009-10-11 at 99bcfaf)
181 - + instaweb: support mod_cgid for apache2
182 -
183 -* tf/doc-pt-br (2009-09-23) 1 commit.
184 - - Documentation: update pt-BR
185 -
186 -The current AsciiDoc may barf on NOME and SINOPSE, as pt_BR language
187 -definition is not widely distributed yet (it just hit the development
188 -tree). Need to revert these headings (or change the length of the section
189 -underlines to match the length of translated names).
190 -
269 * jc/pretty-lf (2009-10-04) 1 commit.
270 - Pretty-format: %[+-]x to tweak inter-item newlines
271
194 -I am not happy with this one yet. I am contemplating to introduce a new
195 -syntax "%[magic(param)<anything>%]" to generalize expressions of this and
196 -line wrapping features in an extensible way.
197 -
198 -* js/log-rewrap (2008-11-10) 3 commits.
199 - . Add "%w" to pretty formats, which rewraps the commit message
272 +* js/log-rewrap (2008-11-10) 2 commits
273 - Add strbuf_add_wrapped_text() to utf8.[ch]
274 - print_wrapped_text(): allow hard newlines
202 -... and the first two from this series will be useful to implement an
203 -example magic "wrap", e.g. "%[wrap(i,j,w)%s%+b%]".
275 + (this branch is used by jc/strbuf-nested-expand.)
276
205 -* jg/log-format-body-indent (2009-09-19) 1 commit.
206 - - git-log --format: Add %B tag with %B(x) option
207 -
208 -I think we should redo this on top of the first two patches from
209 -js/log-rewrap series; %B(x) is just a special case %B(x,x,0), no? If a
210 -magic value 0 (or negative) given to wrap-width does not disable wrapping,
211 -we probably should make it so.
212 -
213 -* bg/rebase-reword (2009-10-07) 1 commit.
214 - - Teach 'rebase -i' the command "reword"
215 -
216 -Soon in 'next'
217 -
218 -* js/diff-verbose-submodule (2009-10-04) 1 commit.
277 +* js/diff-verbose-submodule (2009-10-14) 4 commits.
278 + - add tests for git diff --submodule-summary
279 + - fix output for deleted submodules in git diff --submodule-summary
280 + - fix indentation depth for git diff --submodule-summary
281 - Add the --submodule-summary option to the diff option family
282
221 -Dscho sounded like he has some corrections after list comments, but I did
222 -not pick up his interdiff in the middle.
223 -
224 -* jl/submodule-add-noname (2009-09-22) 1 commit.
225 - - git submodule add: make the <path> parameter optional
226 -
227 -Dscho started an interesting discussion regarding the larger workflow in
228 -which the "submodule add" is used. I think the patch itself makes sense
229 -but at the same time it probably makes sense to also take the <path> and
230 -infer the <repository> as Dscho suggested, probably in "git submodule
231 -add", not in "git add" proper, at least initially.
283 +There was an attempted re-roll and it looked very promising with a minor
284 +fix-up, but I ran out of time to pick it up and unilaterally tweaking it.
285 +Waiting for a response.
286
287 * jc/fix-tree-walk (2009-09-14) 10 commits.
288 - read-tree --debug-unpack
235 - - unpack-trees.c: look ahead in the index
236 - - unpack-trees.c: prepare for looking ahead in the index
237 - - Aggressive three-way merge: fix D/F case
238 - - traverse_trees(): handle D/F conflict case sanely
239 - - more D/F conflict tests
240 - - tests: move convenience regexp to match object names to test-lib.sh
241 - - unpack_callback(): use unpack_failed() consistently
242 - - unpack-trees: typofix
243 - - diff-lib.c: fix misleading comments on oneway_diff()
289 + (merged to 'next' on 2009-10-11 at 0b058e2)
290 + + unpack-trees.c: look ahead in the index
291 + + unpack-trees.c: prepare for looking ahead in the index
292 + + Aggressive three-way merge: fix D/F case
293 + + traverse_trees(): handle D/F conflict case sanely
294 + + more D/F conflict tests
295 + + tests: move convenience regexp to match object names to test-lib.sh
296 + + unpack_callback(): use unpack_failed() consistently
297 + + unpack-trees: typofix
298 + + diff-lib.c: fix misleading comments on oneway_diff()
299
300 This is my replacement for Linus's lt/maint-traverse-trees-fix patch. It
301 is not so much as a counter-proposal; I originally thought it might make
@@ -250,37 +305,7 @@ what Linus outlined --- walk the trees, and have the index walker follow
305 it. It turned out that the index side also needed some hairy look-ahead,
306
307 This includes the fix to aggressive mode of three-way merge used by the
253 -resolve strategy. I think the series is Ok for 'next'.
254 -
255 -* jp/fetch-tag-match (2009-09-17) 1 commit.
256 - - fetch: Speed up fetch by rewriting find_non_local_tags
257 -
258 -I just queued this without reading nor thinking about it very much. I
259 -personally liked my version that had far smaller number of lines changed
260 -(which means I can be fairly certain that it did not introduce any
261 -regression), but perhaps the majorly rewritten logic this patch gives us
262 -may be easier to follow and maintain.
263 -
264 -* jc/maint-blank-at-eof (2009-09-15) 0 commits.
265 - (this branch uses jc/maint-1.6.0-blank-at-eof.)
266 -
267 -The series does not have a commit of its own but is a preparation for
268 -merging the original jc/1.6.0-maint-blank-at-eof topic to 'maint' and then
269 -'master'
270 -
271 -Soon in 'next'
272 -
273 -* db/vcs-helper-rest (2009-09-03) 6 commits.
274 - - Allow helpers to report in "list" command that the ref is unchanged
275 - - Add support for "import" helper command
276 - - Add a config option for remotes to specify a foreign vcs
277 - - Allow programs to not depend on remotes having urls
278 - - Allow fetch to modify refs
279 - - Use a function to determine whether a remote is valid
280 - (this branch is used by jh/cvs-helper.)
281 -
282 -This holds the remainder of the db/vcs-helper topic that has already
283 -merged in 1.6.5.
308 +resolve strategy.
309
310 * jh/notes (2009-10-09) 22 commits.
311 - fast-import: Proper notes tree manipulation using the notes API
@@ -305,36 +330,9 @@ merged in 1.6.5.
330 - Speed up git notes lookup
331 - Add a script to edit/inspect notes
332 - Introduce commit notes
333 + (this branch uses sr/gfi-options.)
334
309 -Rerolled and queued.
310 -
311 -* jn/gitweb-show-size (2009-09-07) 1 commit.
312 - - gitweb: Add 'show-sizes' feature to show blob sizes in tree view
313 -
314 -Soon in 'next'
315 -
316 -* jc/maint-1.6.0-blank-at-eof (2009-09-14) 15 commits.
317 - (merged to 'next' on 2009-10-11 at 7f34fb3)
318 - + diff -B: colour whitespace errors
319 - + diff.c: emit_add_line() takes only the rest of the line
320 - + diff.c: split emit_line() from the first char and the rest of the line
321 - + diff.c: shuffling code around
322 - + diff --whitespace: fix blank lines at end
323 - + core.whitespace: split trailing-space into blank-at-{eol,eof}
324 - + diff --color: color blank-at-eof
325 - + diff --whitespace=warn/error: fix blank-at-eof check
326 - + diff --whitespace=warn/error: obey blank-at-eof
327 - + diff.c: the builtin_diff() deals with only two-file comparison
328 - + apply --whitespace: warn blank but not necessarily empty lines at EOF
329 - + apply --whitespace=warn/error: diagnose blank at EOF
330 - + apply.c: split check_whitespace() into two
331 - + apply --whitespace=fix: detect new blank lines at eof correctly
332 - + apply --whitespace=fix: fix handling of blank lines at the eof
333 - (this branch is used by jc/maint-blank-at-eof.)
334 -
335 -This is a fix for an ancient bug (or inconsistent set of features); the
336 -topic is based on an ancient codebase and is designed to be merged
337 -upwards. jc/maint-blank-at-eof serves that purpose.
335 +Is this good for 'next' now?
336
337 * jn/gitweb-blame (2009-09-01) 5 commits.
338 - gitweb: Minify gitweb.js if JSMIN is defined
@@ -425,33 +423,18 @@ Maybe some people care about this. I dunno.
423
424 Maybe some people care about this. I dunno.
425
428 -* lt/read-directory (2009-05-15) 3 commits.
429 - . Add initial support for pathname conversion to UTF-8
430 - . read_directory(): infrastructure for pathname character set conversion
431 - . Add 'fill_directory()' helper function for directory traversal
432 -
433 -* cc/reset-merge (2009-09-16) 4 commits.
434 - . reset: add test cases for "--merge-safe" option
435 - . reset: add option "--merge-safe" to "git reset"
436 - . reset: use "unpack_trees()" directly instead of "git read-tree"
437 - . reset: add a few tests for "git reset --merge"
438 -
439 -* cc/sequencer-rebase-i (2009-08-28) 15 commits.
440 - . rebase -i: use "git sequencer--helper --cherry-pick"
441 - . sequencer: add "--cherry-pick" option to "git sequencer--helper"
442 - . sequencer: add "do_commit()" and related functions working on "next_commit"
443 - . pick: libify "pick_help_msg()"
444 - . revert: libify cherry-pick and revert functionnality
445 - . rebase -i: use "git sequencer--helper --fast-forward"
446 - . sequencer: let "git sequencer--helper" callers set "allow_dirty"
447 - . sequencer: add "--fast-forward" option to "git sequencer--helper"
448 - . sequencer: add "do_fast_forward()" to perform a fast forward
449 - . rebase -i: use "git sequencer--helper --reset-hard"
450 - . sequencer: add "--reset-hard" option to "git sequencer--helper"
451 - . sequencer: add "reset_almost_hard()" and related functions
452 - . rebase -i: use "git sequencer--helper --make-patch"
453 - . sequencer: add "make_patch" function to save a patch
454 - . sequencer: add "builtin-sequencer--helper.c"
455 -
456 -* lt/maint-traverse-trees-fix (2009-09-06) 1 commit.
457 - . Prepare 'traverse_trees()' for D/F conflict lookahead
426 +* ne/rev-cache (2009-09-07) 7 commits.
427 + . support for commit grafts, slight change to general mechanism
428 + . support for path name caching in rev-cache
429 + . full integration of rev-cache into git, completed test suite
430 + . administrative functions for rev-cache, start of integration into git
431 + . support for non-commit object caching in rev-cache
432 + . basic revision cache system, no integration or features
433 + . man page and technical discussion for rev-cache
434 +
435 +This seems to break at least t6001 when merged to 'pu'.
436 +
437 +* jg/log-format-body-indent (2009-09-19) 1 commit.
438 + . git-log --format: Add %B tag with %B(x) option
439 +
440 +Ejected from 'pu' as this is merely a short-hand for %[wrap(0,x,x)%b%].