What's cooking (2011/05 #11)

Junio C Hamano committed May 24, 2011 at 14:44 UTC 03aa3f26f04704bc128a4ba1db14db2981f0e498
1 file changed +149 -257
whats-cooking.txt
+149 -257
@@ -1,9 +1,9 @@
1 To: git@vger.kernel.org
2 -Subject: What's cooking in git.git (May 2011, #10; Mon, 23)
2 +Subject: What's cooking in git.git (May 2011, #11; Tue, 24)
3 X-master-at: 5df3e2b3ca5ebe8123927a81d682993ad597a584
4 -X-next-at: fbd4877f301a9ed26bffbebd340b19d5a18c8573
4 +X-next-at: 6dbbf00d8c08050a919971f0006f4045e4fec9f1
5
6 -What's cooking in git.git (May 2011, #10; Mon, 23)
6 +What's cooking in git.git (May 2011, #11; Tue, 24)
7 --------------------------------------------------
8
9 Here are the topics that have been cooking. Commits prefixed with '-' are
@@ -18,6 +18,116 @@ with what is in 'master' at the end of the month.
18 --------------------------------------------------
19 [New Topics]
20
21 +* jc/fix-config-environment-passing (2011-05-24) 1 commit
22 + - Allow built-ins to also use -c var=val via alias
23 +
24 +--------------------------------------------------
25 +[Stalled]
26 +
27 +* jl/read-tree-m-dry-run (2011-05-11) 1 commit
28 + - Teach read-tree the -n|--dry-run option
29 +
30 +Waiting for a reroll.
31 +
32 +* mg/diff-stat-count (2011-05-03) 2 commits
33 + - diff-options.txt: describe --stat-{width,name-width,count}
34 + - diff: introduce --stat-count to limit the stat lines
35 +
36 +There was a miscounting spotted. Need another round.
37 +
38 +* jk/maint-merge-rename-create (2011-03-25) 3 commits
39 + - merge: turn on rewrite detection
40 + - merge: handle renames with replacement content
41 + - t3030: fix accidental success in symlink rename
42 +
43 +Peff wanted to reroll this.
44 +
45 +* jc/index-pack (2011-02-27) 5 commits
46 + - index-pack --verify: read anomalous offsets from v2 idx file
47 + - write_idx_file: need_large_offset() helper function
48 + - index-pack: --verify
49 + - write_idx_file: introduce a struct to hold idx customization options
50 + - index-pack: group the delta-base array entries also by type
51 +
52 +Still a WIP. Need to put histogram output into index-pack --verify to
53 +really kill verify-pack.
54 +
55 +* jk/tag-contains (2010-07-05) 4 commits
56 + - Why is "git tag --contains" so slow?
57 + - default core.clockskew variable to one day
58 + - limit "contains" traversals based on commit timestamp
59 + - tag: speed up --contains calculation
60 +
61 +The idea of the bottom one is probably Ok, except that the use of object
62 +flags needs to be rethought, or at least the helper needs to be moved to
63 +builtin/tag.c to make it clear that it should not be used outside the
64 +current usage context.
65 +
66 +--------------------------------------------------
67 +[Cooking]
68 +
69 +* da/git-prefix-everywhere (2011-05-23) 3 commits
70 + - git-mergetool--lib: Make vimdiff retain the current directory
71 + - git: Remove handling for GIT_PREFIX
72 + - setup: Provide GIT_PREFIX to built-ins
73 +
74 +Is everybody happy with this?
75 +Will merge to "next".
76 +
77 +* jh/receive-count-limit (2011-05-23) 10 commits
78 + - receive-pack: Allow server to refuse pushes with too many objects
79 + - pack-objects: Estimate pack size; abort early if pack size limit is exceeded
80 + - send-pack/receive-pack: Allow server to refuse pushing too large packs
81 + - pack-objects: Allow --max-pack-size to be used together with --stdout
82 + - send-pack/receive-pack: Allow server to refuse pushes with too many commits
83 + - pack-objects: Teach new option --max-commit-count, limiting #commits in pack
84 + - receive-pack: Prepare for addition of the new 'limit-*' family of capabilities
85 + - Tighten rules for matching server capabilities in server_supports()
86 + - send-pack: Attempt to retrieve remote status even if pack-objects fails
87 + - Update technical docs to reflect side-band-64k capability in receive-pack
88 +
89 +Would need another round to separate per-pack and per-session limits.
90 +
91 +* jc/require-work-tree-exists (2011-05-24) 1 commit
92 + (merged to 'next' on 2011-05-24 at 6dbbf00)
93 + + require-work-tree wants more than what its name says
94 +
95 +Rerolled _without_ any in-tree users, just to make sure that we can easily
96 +adapt scripts when necessary in the future.
97 +
98 +Will merge to "master" by the end of week #5.
99 +
100 +* jn/gitweb-js (2011-05-24) 11 commits
101 + (merged to 'next' on 2011-05-24 at c385b9d)
102 + + gitweb: Make JavaScript ability to adjust timezones configurable
103 + + gitweb.js: Add UI for selecting common timezone to display dates
104 + + gitweb: JavaScript ability to adjust time based on timezone
105 + + gitweb: Unify the way long timestamp is displayed
106 + + gitweb: Refactor generating of long dates into format_timestamp_html
107 + + gitweb.js: Provide getElementsByClassName method (if it not exists)
108 + + gitweb.js: Introduce code to handle cookies from JavaScript
109 + + gitweb.js: Extract and improve datetime handling
110 + + gitweb.js: Provide default values for padding in padLeftStr and padLeft
111 + + gitweb.js: Update and improve comments in JavaScript files
112 + + gitweb: Split JavaScript for maintability, combining on build
113 +
114 +Rebased on top of a more recent "master" that already has the two bottom
115 +commits in the previous round of series.
116 +
117 +Will merge to "master" by the end of week #5.
118 +
119 +* jn/ctags-more (2011-04-29) 3 commits
120 + (merged to 'next' on 2011-05-24 at 41b50be)
121 + + gitweb: Optional grouping of projects by category
122 + + gitweb: Modularized git_get_project_description to be more generic
123 + + gitweb: Split git_project_list_body in two functions
124 +
125 +As nobody seems to be helping gitweb maintenance, I'll merge the above two
126 +series to "master" and see if anybody screams. That seems to be the only
127 +way to make any progress.
128 +
129 +Will merge to "master" by the end of week #5.
130 +
131 * jc/notes-batch-removal (2011-05-19) 4 commits
132 (merged to 'next' on 2011-05-23 at 5bd78a6)
133 + show: --ignore-missing
@@ -75,266 +185,27 @@ merge before 1.7.6.
185
186 Will merge to "master" by the end of week #5.
187
78 -* da/git-prefix-everywhere (2011-05-23) 3 commits
79 - - git-mergetool--lib: Make vimdiff retain the current directory
80 - - git: Remove handling for GIT_PREFIX
81 - - setup: Provide GIT_PREFIX to built-ins
82 -
83 -* jc/git-handle-options-miscount (2011-05-23) 1 commit
188 +* jc/git-handle-options-miscount (2011-05-24) 1 commit
189 - handle_options(): do not miscount how many arguments were used
190
86 -* jk/combine-diff-binary-etc (2011-05-23) 5 commits
87 - - combine-diff: respect textconv attributes
88 - - refactor get_textconv to not require diff_filespec
89 - - combine-diff: handle binary files as binary
90 - - combine-diff: calculate mode_differs earlier
91 - - combine-diff: split header printing into its own function
191 +Updated with a trivial test, which turned out to uncover an unrelated bug
192 +(see jc/fix-config-environment-passing topic).
193
93 ---------------------------------------------------
94 -[Graduated to "master"]
95 -
96 -* ab/i18n-scripts-basic (2011-05-14) 4 commits
97 - (merged to 'next' on 2011-05-18 at 3dc69f0)
98 - + Makefile: add xgettext target for *.sh files
99 - + git-sh-i18n.sh: add GIT_GETTEXT_POISON support
100 - + git-sh-i18n.sh: add no-op gettext() and eval_gettext() wrappers
101 - + git-sh-i18n--envsubst: our own envsubst(1) for eval_gettext()
102 - (this branch is used by ab/i18n-scripts.)
103 -
104 -* jc/magic-pathspec (2011-05-17) 14 commits
105 - (merged to 'next' on 2011-05-18 at d7180b9)
106 - + setup.c: Fix some "symbol not declared" sparse warnings
107 - (merged to 'next' on 2011-05-12 at 656c8b5)
108 - + t3703: Skip tests using directory name ":" on Windows
109 - (merged to 'next' on 2011-05-11 at 86d1d6d)
110 - + revision.c: leave a note for "a lone :" enhancement
111 - + Merge branch 'jc/maint-1.6.6-pathspec-stdin-and-cmdline'
112 - (merged to 'next' on 2011-05-11 at aaad95e)
113 - + t3703, t4208: add test cases for magic pathspec
114 - + rev/path disambiguation: further restrict "misspelled index entry" diag
115 - + fix overslow :/no-such-string-ever-existed diagnostics
116 - + fix overstrict :<path> diagnosis
117 - + grep: use get_pathspec() correctly
118 - + pathspec: drop "lone : means no pathspec" from get_pathspec()
119 - + Revert "magic pathspec: add ":(icase)path" to match case insensitively"
120 - (merged to 'next' on 2011-04-25 at 788cd46)
121 - + magic pathspec: add ":(icase)path" to match case insensitively
122 - + magic pathspec: futureproof shorthand form
123 - + magic pathspec: add tentative ":/path/from/top/level" pathspec support
124 -
125 -* jc/maint-1.6.6-pathspec-stdin-and-cmdline (2011-05-11) 1 commit
126 - + setup_revisions(): take pathspec from command line and --stdin correctly
127 - (this branch is used by jc/magic-pathspec, jc/maint-1.7.4-pathspec-stdin-and-cmdline and jc/maint-pathspec-stdin-and-cmdline.)
128 -
129 -* jc/maint-1.7.4-pathspec-stdin-and-cmdline (2011-05-11) 1 commit
130 - + Merge branch 'jc/maint-1.6.6-pathspec-stdin-and-cmdline'
131 -
132 -* jc/maint-pathspec-stdin-and-cmdline (2011-05-11) 2 commits
133 - (merged to 'next' on 2011-05-15 at b4e3a75)
134 - + Merge branch 'jc/maint-1.7.4-pathspec-stdin-and-cmdline'
135 - + Merge branch 'jc/maint-1.6.6-pathspec-stdin-and-cmdline'
136 - (this branch uses jc/maint-1.6.6-pathspec-stdin-and-cmdline; is tangled with jc/magic-pathspec and jc/maint-1.7.4-pathspec-stdin-and-cmdline.)
137 -
138 -These are to fix an ancient bug where feeding "git log --stdin path" with
139 -an input that also has additional pathspec corrupted memory.
140 -
141 -* jc/replacing (2011-05-15) 5 commits
142 - (merged to 'next' on 2011-05-16 at 5d0b584)
143 - + read_sha1_file(): allow selective bypassing of replacement mechanism
144 - + inline lookup_replace_object() calls
145 - + read_sha1_file(): get rid of read_sha1_file_repl() madness
146 - + t6050: make sure we test not just commit replacement
147 - + Declare lookup_replace_object() in cache.h, not in commit.h
148 - (this branch is used by jc/streaming.)
149 -
150 -Code restructuring without much new feature.
151 -
152 -* jk/blame-line-porcelain (2011-05-09) 3 commits
153 - (merged to 'next' on 2011-05-15 at 93232b3)
154 - + blame: add --line-porcelain output format
155 - + blame: refactor porcelain output
156 - + add tests for various blame formats
157 -
158 -* jk/cherry-pick-root-with-resolve (2011-05-16) 3 commits
159 - (merged to 'next' on 2011-05-16 at 2a63845)
160 - + t3503: test cherry picking and reverting root commits
161 - + revert: allow reverting a root commit
162 - + cherry-pick: handle root commits with external strategies
163 -
164 -* jk/git-connection-deadlock-fix (2011-05-18) 4 commits
165 - (merged to 'next' on 2011-05-18 at d4567da)
166 - + test core.gitproxy configuration
167 - (merged to 'next' on 2011-05-16 at 33e6c11)
168 - + send-pack: avoid deadlock on git:// push with failed pack-objects
169 - + connect: let callers know if connection is a socket
170 - + connect: treat generic proxy processes like ssh processes
171 - (this branch uses js/maint-1.6.6-send-pack-stateless-rpc-deadlock-fix; is tangled with js/maint-send-pack-stateless-rpc-deadlock-fix.)
172 -
173 -* jl/submodule-conflicted-gitmodules (2011-05-14) 2 commits
174 - (merged to 'next' on 2011-05-16 at cc6c7e8)
175 - + Submodules: Don't parse .gitmodules when it contains, merge conflicts
176 - + test that git status works with merge conflict in, .gitmodules
177 -
178 -* jm/maint-diff-words-with-sbe (2011-05-20) 1 commit
179 - (merged to 'next' on 2011-05-20 at c0db85c)
180 - + do not read beyond end of malloc'd buffer
181 -
182 -* js/maint-1.6.6-send-pack-stateless-rpc-deadlock-fix (2011-05-16) 1 commit
183 - + sideband_demux(): fix decl-after-stmt
184 - (this branch is used by jk/git-connection-deadlock-fix and js/maint-send-pack-stateless-rpc-deadlock-fix.)
185 -
186 -* js/maint-send-pack-stateless-rpc-deadlock-fix (2011-05-16) 1 commit
187 - (merged to 'next' on 2011-05-16 at ad83470)
188 - + Merge branch 'js/maint-1.6.6-send-pack-stateless-rpc-deadlock-fix' into js/maint-send-pack-stateless-rpc-deadlock-fix
189 - (this branch uses js/maint-1.6.6-send-pack-stateless-rpc-deadlock-fix; is tangled with jk/git-connection-deadlock-fix.)
190 -
191 -* js/mingw-shutdown (2011-05-18) 1 commit
192 - (merged to 'next' on 2011-05-19 at c085b22)
193 - + Windows: add a wrapper for the shutdown() system call
194 -
195 -* ld/p4-preserve-user-names (2011-05-13) 3 commits
196 - (merged to 'next' on 2011-05-15 at bdedb60)
197 - + git-p4: warn if git authorship won't be retained
198 - (merged to 'next' on 2011-05-11 at bc14314)
199 - + git-p4: small improvements to user-preservation
200 - (merged to 'next' on 2011-04-29 at 25116c8)
201 - + git-p4: add option to preserve user names
202 -
203 -* mg/config-symbolic-constants (2011-05-17) 2 commits
204 - (merged to 'next' on 2011-05-19 at ac538c0)
205 - + config: Give error message when not changing a multivar
206 - + config: define and document exit codes
207 -
208 -* ms/ls-remote-exit-with-status (2011-05-18) 1 commit
209 - (merged to 'next' on 2011-05-19 at c4238ce)
210 - + ls-remote: the --exit-code option reports "no matching refs"
211 -
212 -* ms/tagname-does-not-begin-with-dash (2011-05-10) 1 commit
213 - (merged to 'next' on 2011-05-15 at 7df0461)
214 - + tag: disallow '-' as tag name
215 -
216 -* nd/sparse-co-fix (2011-05-10) 2 commits
217 - (merged to 'next' on 2011-05-15 at 1a85791)
218 - + sparse checkout: do not eagerly decide the fate for whole directory
219 - + t1011: fix sparse-checkout initialization and add new file
220 -
221 -* rg/copy-gecos-username (2011-05-17) 1 commit
222 - (merged to 'next' on 2011-05-17 at 6080944)
223 - + copy_gecos: fix not adding nlen to len when processing "&"
224 -
225 -* tr/add-i-no-escape (2011-05-17) 1 commit
226 - (merged to 'next' on 2011-05-19 at d1c7a10)
227 - + add -i: ignore terminal escape sequences
194 +* jk/combine-diff-binary-etc (2011-05-24) 5 commits
195 + (merged to 'next' on 2011-05-24 at 07cf180)
196 + + combine-diff: respect textconv attributes
197 + + refactor get_textconv to not require diff_filespec
198 + + combine-diff: handle binary files as binary
199 + + combine-diff: calculate mode_differs earlier
200 + + combine-diff: split header printing into its own function
201
229 ---------------------------------------------------
230 -[Stalled]
202 +Will merge to "master" by the end of month.
203
232 -* hg/add-i-qq (2011-05-17) 1 commit
233 - - add-interactive: add 'Quit' shortcut to add hunk and quit
204 +* jc/advice-about-to-lose-commit (2011-05-24) 1 commit
205 + - checkout: make advice when reattaching the HEAD less loud
206
235 -The question "why is y+q any more deserving for a short-cut than d+q" was
236 -raised late in the discussion, and I think it a very valid concern.
237 -
238 -May drop.
239 -
240 -* jl/read-tree-m-dry-run (2011-05-11) 1 commit
241 - - Teach read-tree the -n|--dry-run option
242 -
243 -Waiting for a reroll.
244 -
245 -* mg/diff-stat-count (2011-05-03) 2 commits
246 - - diff-options.txt: describe --stat-{width,name-width,count}
247 - - diff: introduce --stat-count to limit the stat lines
248 -
249 -There was a miscounting spotted. Need another round.
250 -
251 -* jn/gitweb-js (2011-04-28) 13 commits
252 - - gitweb: Make JavaScript ability to adjust timezones configurable
253 - - gitweb.js: Add UI for selecting common timezone to display dates
254 - - gitweb: JavaScript ability to adjust time based on timezone
255 - - gitweb: Unify the way long timestamp is displayed
256 - - gitweb: Refactor generating of long dates into format_timestamp_html
257 - - gitweb.js: Provide getElementsByClassName method (if it not exists)
258 - - gitweb.js: Introduce code to handle cookies from JavaScript
259 - - gitweb.js: Extract and improve datetime handling
260 - - gitweb.js: Provide default values for padding in padLeftStr and padLeft
261 - - gitweb.js: Update and improve comments in JavaScript files
262 - - gitweb: Split JavaScript for maintability, combining on build
263 - - Remove gitweb/gitweb.cgi and other legacy targets from main Makefile
264 - - git-instaweb: Simplify build dependency on gitweb
265 -
266 -Rerolled. Waiting for comments.
267 -
268 -* jn/ctags-more (2011-04-29) 3 commits
269 - - gitweb: Optional grouping of projects by category
270 - - gitweb: Modularized git_get_project_description to be more generic
271 - - gitweb: Split git_project_list_body in two functions
272 -
273 -Waiting for comments.
274 -
275 -* jc/require-work-tree-exists (2011-05-08) 1 commit
276 - - require-work-tree wants more than what its name says
277 -
278 -Make "git pull" run from a random place work as long as GIT_DIR and
279 -GIT_WORK_TREE are set up correctly. I am not absolutely sure if that
280 -is a sane use case, though. May drop.
281 -
282 -* jc/add-delete-default (2011-04-27) 1 commit
283 - - git add: notice removal of tracked paths by default
284 -
285 -* jk/maint-merge-rename-create (2011-03-25) 3 commits
286 - - merge: turn on rewrite detection
287 - - merge: handle renames with replacement content
288 - - t3030: fix accidental success in symlink rename
289 -
290 -Peff wanted to reroll this.
291 -
292 -* jc/index-pack (2011-02-27) 5 commits
293 - - index-pack --verify: read anomalous offsets from v2 idx file
294 - - write_idx_file: need_large_offset() helper function
295 - - index-pack: --verify
296 - - write_idx_file: introduce a struct to hold idx customization options
297 - - index-pack: group the delta-base array entries also by type
298 -
299 -Still a WIP. Need to put histogram output into index-pack --verify to
300 -really kill verify-pack.
301 -
302 -* jc/dotdot-is-parent-directory (2011-05-04) 1 commit
303 - - specifying ranges: we did not mean to make ".." an empty set
304 -
305 -Updated documentation, but I do not feel a strong enough annoyance to push
306 -this forward. May drop.
307 -
308 -* jk/tag-contains (2010-07-05) 4 commits
309 - - Why is "git tag --contains" so slow?
310 - - default core.clockskew variable to one day
311 - - limit "contains" traversals based on commit timestamp
312 - - tag: speed up --contains calculation
313 -
314 -The idea of the bottom one is probably Ok, except that the use of object
315 -flags needs to be rethought, or at least the helper needs to be moved to
316 -builtin/tag.c to make it clear that it should not be used outside the
317 -current usage context.
318 -
319 -* jc/advice-about-to-lose-commit (2011-05-06) 1 commit
320 - - checkout: honor advice.detachedHead when reattaching to a branch
321 -
322 -Needs to tighten the advice to only govern the generic How-to part.
323 -
324 ---------------------------------------------------
325 -[Cooking]
326 -
327 -* jh/receive-count-limit (2011-05-23) 10 commits
328 - - receive-pack: Allow server to refuse pushes with too many objects
329 - - pack-objects: Estimate pack size; abort early if pack size limit is exceeded
330 - - send-pack/receive-pack: Allow server to refuse pushing too large packs
331 - - pack-objects: Allow --max-pack-size to be used together with --stdout
332 - - send-pack/receive-pack: Allow server to refuse pushes with too many commits
333 - - pack-objects: Teach new option --max-commit-count, limiting #commits in pack
334 - - receive-pack: Prepare for addition of the new 'limit-*' family of capabilities
335 - - Tighten rules for matching server capabilities in server_supports()
336 - - send-pack: Attempt to retrieve remote status even if pack-objects fails
337 - - Update technical docs to reflect side-band-64k capability in receive-pack
207 +Re-rolled and resurrected from "Stalled" status.
208 +Will merge to "next".
209
210 * js/log-abbrev-commit-config (2011-05-18) 2 commits
211 (merged to 'next' on 2011-05-23 at 4e7e932)
@@ -451,3 +322,24 @@ Rerolled.
322 + Documentation: Add --line-number to git-grep synopsis
323
324 Will merge to "master" by the middle of week #5.
325 +
326 +--------------------------------------------------
327 +[Discarded]
328 +
329 +* hg/add-i-qq (2011-05-17) 1 commit
330 + . add-interactive: add 'Quit' shortcut to add hunk and quit
331 +
332 +The question "why is y+q any more deserving for a short-cut than d+q" was
333 +raised late in the discussion, and I think it a very valid concern.
334 +
335 +* jc/add-delete-default (2011-04-27) 1 commit
336 + . git add: notice removal of tracked paths by default
337 +
338 +This is ill-thought out. When the user does want to say "take all changes
339 +into account, not just addition", they can easily say "add -u" (update).
340 +
341 +* jc/dotdot-is-parent-directory (2011-05-04) 1 commit
342 + . specifying ranges: we did not mean to make ".." an empty set
343 +
344 +Updated documentation, but I do not feel a strong enough annoyance to push
345 +this forward.