What's cooking (2012/02 #07)

Junio C Hamano committed Feb 20, 2012 at 02:20 UTC 3ca00b7ce590370fc4e404c26142976defecf0f1
1 file changed +138 -147
whats-cooking.txt
+138 -147
@@ -1,9 +1,9 @@
1 To: git@vger.kernel.org
2 -Subject: What's cooking in git.git (Feb 2012, #06; Thu, 16)
3 -X-master-at: b0d66b5110faaeb395610ba43b6eb70a18ab5e25
4 -X-next-at: a9004c5cb2204cf950debab328e86de5eefb9da4
2 +Subject: What's cooking in git.git (Feb 2012, #07; Mon, 20)
3 +X-master-at: b3a769dc355b32c95783dc07f59e4dfebdd8bdc7
4 +X-next-at: 80e3ff2c99bd723b0dfb64b7283cf92e398c61b6
5
6 -What's cooking in git.git (Feb 2012, #06; Thu, 16)
6 +What's cooking in git.git (Feb 2012, #07; Mon, 20)
7 --------------------------------------------------
8
9 Here are the topics that have been cooking. Commits prefixed with '-' are
@@ -22,87 +22,6 @@ repositories listed at
22 --------------------------------------------------
23 [Graduated to "master"]
24
25 -* bl/gitweb-project-filter (2012-02-12) 1 commit
26 - (merged to 'next' on 2012-02-13 at 35366b8)
27 - + gitweb: Harden and improve $project_filter page title
28 -
29 -An update to the new feature to "gitweb" to show list of projects for
30 -intermediate levels in directory hierarchies, already slated for 1.7.10.
31 -
32 -* dp/i18n-libcharset (2012-02-13) 1 commit
33 - (merged to 'next' on 2012-02-13 at 528de77)
34 - + Makefile: introduce CHARSET_LIB to link with -lcharset
35 -
36 -Some systems need to explicitly link -lcharset to get locale_charset().
37 -
38 -* jk/git-dir-lookup (2012-02-02) 1 commit
39 - (merged to 'next' on 2012-02-05 at 1856d74)
40 - + standardize and improve lookup rules for external local repos
41 -
42 -When you have both .../foo and .../foo.git, "git clone .../foo" did not
43 -favor the former but the latter.
44 -
45 -* jk/grep-binary-attribute (2012-02-02) 9 commits
46 - (merged to 'next' on 2012-02-05 at 9dffa7e)
47 - + grep: pre-load userdiff drivers when threaded
48 - + grep: load file data after checking binary-ness
49 - + grep: respect diff attributes for binary-ness
50 - + grep: cache userdiff_driver in grep_source
51 - + grep: drop grep_buffer's "name" parameter
52 - + convert git-grep to use grep_source interface
53 - + grep: refactor the concept of "grep source" into an object
54 - + grep: move sha1-reading mutex into low-level code
55 - + grep: make locking flag global
56 -
57 -Fixes a longstanding bug that there was no way to tell "git grep" that a
58 -path may look like text but it is not, which "git diff" can do using the
59 -attributes system. Now "git grep" honors the same "binary" (or "-diff")
60 -attribute.
61 -
62 -* jk/userdiff-config-simplify (2012-02-07) 1 commit
63 - (merged to 'next' on 2012-02-10 at e9854c1)
64 - + drop odd return value semantics from userdiff_config
65 -
66 -Code cleanup.
67 -
68 -* jn/ancient-meld-support (2012-02-10) 1 commit
69 - (merged to 'next' on 2012-02-13 at 28aca31)
70 - + mergetools/meld: Use --help output to detect --output support
71 -
72 -More reliably tell if the given version of "meld" supports --output
73 -option.
74 -
75 -* lt/pull-no-edit (2012-02-12) 1 commit
76 - (merged to 'next' on 2012-02-13 at 352f0cb)
77 - + "git pull" doesn't know "--edit"
78 -
79 -For 1.7.10 where "git merge" by default spawns the editor when it can
80 -automatically commit to record the resulting merge.
81 -
82 -* mh/war-on-extra-refs (2012-02-12) 7 commits
83 - (merged to 'next' on 2012-02-13 at adb7353)
84 - + refs: remove the extra_refs API
85 - + clone: do not add alternate references to extra_refs
86 - + everything_local(): mark alternate refs as complete
87 - + fetch-pack.c: inline insert_alternate_refs()
88 - + fetch-pack.c: rename some parameters from "path" to "refname"
89 - + clone.c: move more code into the "if (refs)" conditional
90 - + t5700: document a failure of alternates to affect fetch
91 -
92 -Internal API clean-up that is very cleanly done.
93 -
94 -* nd/pack-objects-parseopt (2012-02-01) 3 commits
95 - (merged to 'next' on 2012-02-05 at d0dc25d)
96 - + pack-objects: convert to use parse_options()
97 - + pack-objects: remove bogus comment
98 - + pack-objects: do not accept "--index-version=version,"
99 -
100 -"pack-objects" learned use parse-options, losing custom command line
101 -parsing code.
102 -
103 ---------------------------------------------------
104 -[New Topics]
105 -
25 * cb/maint-rev-list-verify-object (2012-02-13) 1 commit
26 (merged to 'next' on 2012-02-16 at a407b9a)
27 + git rev-list: fix invalid typecast
@@ -131,14 +50,19 @@ resulting in false indication of success.
50 The transport programs ignored --no-progress and showed progress when
51 sending their output to a terminal.
52
134 -* jk/diff-highlight (2012-02-13) 5 commits
135 - - diff-highlight: document some non-optimal cases
136 - - diff-highlight: match multi-line hunks
137 - - diff-highlight: refactor to prepare for multi-line hunks
138 - - diff-highlight: don't highlight whole lines
139 - - diff-highlight: make perl strict and warnings fatal
53 +* jc/diff-stat-scaler (2012-02-14) 1 commit
54 + (merged to 'next' on 2012-02-16 at 404d336)
55 + + diff --stat: show bars of same length for paths with same amount of changes
56
141 -Updates diff-highlight (in contrib/).
57 +The output from "git diff --stat" for two paths that have the same amount
58 +of changes showed graph bars of different length due to the way we handled
59 +rounding errors.
60 +
61 +* tg/tag-points-at (2012-02-13) 2 commits
62 + (merged to 'next' on 2012-02-13 at a8f4046)
63 + + builtin/tag.c: Fix a sparse warning
64 + (merged to 'next' on 2012-02-10 at 4bff88f)
65 + + tag: add --points-at list option
66
67 * zj/decimal-width (2012-02-14) 1 commit
68 (merged to 'next' on 2012-02-16 at 72805c4)
@@ -155,12 +79,80 @@ Refactoring.
79 Fixes "git -p cmd" for any subcommand that cares about the true terminal
80 width.
81
82 +--------------------------------------------------
83 +[New Topics]
84 +
85 +* fc/push-prune (2012-02-17) 2 commits
86 + - remote: reorganize check_pattern_match()
87 + - remote: use a local variable in match_push_refs()
88 +
89 +The first two of the three-patch RFC series. The second one and the third
90 +one (not queued) both need better explanation, and the third one also had
91 +a handful of minor yuck factors all of which should be straightforward to
92 +improve.
93 +
94 +Expecting a reroll.
95 +
96 +* jc/add-refresh-unmerged (2012-02-17) 1 commit
97 + - refresh_index: do not show unmerged path that is outside pathspec
98 +
99 +"git add --refresh <pathspec>" warned about unmerged paths outside the
100 +given pathspec.
101 +
102 +* jc/diff-ignore-case (2012-02-19) 6 commits
103 + - diff -i
104 + - diff: --ignore-case
105 + - xdiff: introduce XDF_IGNORE_CASE
106 + - xdiff: introduce XDF_INEXACT_MATCH
107 + - xdiff: PATIENCE/HISTOGRAM are not independent option bits
108 + - xdiff: remove XDL_PATCH_* macros
109 +
110 +"git diff" learned "--ignore-case" option.
111 +
112 +* jn/gitweb-hilite-regions (2012-02-19) 5 commits
113 + - gitweb: Use esc_html_match_hl() in 'grep' search
114 + - gitweb: Highlight matched part of shortened project description
115 + - gitweb: Highlight matched part of project description when searching projects
116 + - gitweb: Highlight matched part of project name when searching projects
117 + - gitweb: Introduce esc_html_hl_regions
118 + (this branch uses jn/gitweb-search-optim.)
119 +
120 +Not reviewed and do not know what this is about yet ;-).
121 +
122 +* jn/gitweb-search-optim (2012-02-19) 3 commits
123 + - gitweb: Faster project search
124 + - gitweb: Option for filling only specified info in fill_project_list_info
125 + - gitweb: Refactor checking if part of project info need filling
126 + (this branch is used by jn/gitweb-hilite-regions.)
127 +
128 +The API introduced in the second step still has yucky design, but at least
129 +it is more clear than the previous rounds what this is trying to do.
130 +
131 +* js/configure-libintl (2012-02-20) 1 commit
132 + - configure: don't use -lintl when there is no gettext support
133 +
134 +Build fix for autoconf, meant for 'maint' track.
135 +
136 +* pj/remote-set-branches-usage-fix (2012-02-19) 1 commit
137 + - remote: fix set-branches usage and documentation
138 +
139 +Documentation fix.
140 +
141 +--------------------------------------------------
142 +[Stalled]
143 +
144 +* jl/maint-submodule-relative (2012-02-09) 2 commits
145 + - submodules: always use a relative path from gitdir to work tree
146 + - submodules: always use a relative path to gitdir
147 +
148 +The second one looked iffy.
149 +
150 * hv/submodule-recurse-push (2012-02-13) 3 commits
151 - push: teach --recurse-submodules the on-demand option
152 - Refactor submodule push check to use string list instead of integer
153 - Teach revision walking machinery to walk multiple times sequencially
154
163 -The bottom one was not clearly explained.
155 +The bottom one was not clearly explained and needs a reroll.
156
157 * zj/diff-stat-dyncol (2012-02-15) 6 commits
158 . diff --stat: use less columns for change counts
@@ -169,38 +161,9 @@ The bottom one was not clearly explained.
161 - (squash to the previous -- replace the log message with this)
162 - diff --stat: tests for long filenames and big change counts
163 - Merge branches zj/decimal-width and zj/term-columns
172 - (this branch uses zj/decimal-width and zj/term-columns.)
164
174 -* jc/diff-stat-scaler (2012-02-14) 1 commit
175 - (merged to 'next' on 2012-02-16 at 404d336)
176 - + diff --stat: show bars of same length for paths with same amount of changes
177 -
178 -The output from "git diff --stat" for two paths that have the same amount
179 -of changes showed graph bars of different length due to the way we handled
180 -rounding errors.
181 -
182 -* jn/gitweb-unborn-head (2012-02-16) 1 commit
183 - - gitweb: Deal with HEAD pointing to unborn branch in "heads" view
184 -
185 -"gitweb" compared non-existent value of HEAD with the names of commit
186 -objects at tips of branches, triggering runtime warnings.
187 -
188 -* tr/perftest (2012-02-16) 3 commits
189 - . Add a performance test for git-grep
190 - . Introduce a performance testing framework
191 - . Move the user-facing test library to test-lib-functions.sh
192 -
193 -When merged to 'pu' this seems to break quite a lot of tests. One way it
194 -does so is by removing the write_script helper, but there may be others.
195 -
196 -* jb/required-filter (2012-02-16) 1 commit
197 - - Add a setting to require a filter to be successful
198 -
199 -A content filter used to be a way to make the recorded contents "more
200 -useful", but this defines a way to optionally mark a filter "required".
201 -
202 ---------------------------------------------------
203 -[Stalled]
165 +I am beginning to think that the last one should wait until the dust from
166 +the earlier part settles.
167
168 * jc/advise-push-default (2011-12-18) 1 commit
169 - push: hint to use push.default=upstream when appropriate
@@ -268,35 +231,61 @@ Expecting a reroll on top of zj/term-columns topic.
231 --------------------------------------------------
232 [Cooking]
233
271 -* jk/config-include (2012-02-16) 9 commits
272 - - config: add include directive
273 - - config: eliminate config_exclusive_filename
274 - - config: stop using config_exclusive_filename
275 - - config: provide a version of git_config with more options
276 - - config: teach git_config_rename_section a file argument
277 - - config: teach git_config_set_multivar_in_file a default path
278 - - config: copy the return value of prefix_filename
279 - - t1300: add missing &&-chaining
280 - + docs: add a basic description of the config API
234 +* jk/diff-highlight (2012-02-13) 5 commits
235 + (merged to 'next' on 2012-02-20 at ba040ae)
236 + + diff-highlight: document some non-optimal cases
237 + + diff-highlight: match multi-line hunks
238 + + diff-highlight: refactor to prepare for multi-line hunks
239 + + diff-highlight: don't highlight whole lines
240 + + diff-highlight: make perl strict and warnings fatal
241
282 -An assignment to the include.path pseudo-variable causes the named file
283 -to be included in-place when Git looks up configuration variables.
242 +Updates diff-highlight (in contrib/).
243
285 -Reverted the earlier round from 'next'.
244 +* jn/gitweb-unborn-head (2012-02-17) 1 commit
245 + (merged to 'next' on 2012-02-20 at 80e3ff2)
246 + + gitweb: Fix "heads" view when there is no current branch
247
287 -* tg/tag-points-at (2012-02-13) 2 commits
288 - (merged to 'next' on 2012-02-13 at a8f4046)
289 - + builtin/tag.c: Fix a sparse warning
290 - (merged to 'next' on 2012-02-10 at 4bff88f)
291 - + tag: add --points-at list option
248 +"gitweb" compared non-existent value of HEAD with the names of commit
249 +objects at tips of branches, triggering runtime warnings.
250
293 -Will merge to 'master'.
251 +* tr/perftest (2012-02-17) 3 commits
252 + (merged to 'next' on 2012-02-20 at 4c75ba9)
253 + + Add a performance test for git-grep
254 + + Introduce a performance testing framework
255 + + Move the user-facing test library to test-lib-functions.sh
256
295 -* jl/maint-submodule-relative (2012-02-09) 2 commits
296 - - submodules: always use a relative path from gitdir to work tree
297 - - submodules: always use a relative path to gitdir
257 +* jb/required-filter (2012-02-17) 1 commit
258 + . Add a setting to require a filter to be successful
259
299 -The second one looked iffy.
260 +A content filter used to be a way to make the recorded contents "more
261 +useful", but this defines a way to optionally mark a filter "required".
262 +
263 +A few test in t0021 use 'false' as the filter, which can exit without
264 +reading any byte from us, before we start writing and causes us to die
265 +with SIGPIPE, leading to intermittent test failure. I think treating this
266 +as a failure of running the filter (the end user's filter should read what
267 +is fed in full, produce its output and write the result back to us) is the
268 +right thing to do, so this patch needs update the surrounding code to
269 +handle such a situation better, probably by using sigchain_push(SIGPIPE)
270 +or something.
271 +
272 +* jk/config-include (2012-02-17) 10 commits
273 + (merged to 'next' on 2012-02-20 at 7b150b7)
274 + + config: add include directive
275 + + config: eliminate config_exclusive_filename
276 + + config: stop using config_exclusive_filename
277 + + config: provide a version of git_config with more options
278 + + config: teach git_config_rename_section a file argument
279 + + config: teach git_config_set_multivar_in_file a default path
280 + + config: copy the return value of prefix_filename
281 + + t1300: add missing &&-chaining
282 + + docs/api-config: minor clarifications
283 + + docs: add a basic description of the config API
284 +
285 +An assignment to the include.path pseudo-variable causes the named file
286 +to be included in-place when Git looks up configuration variables.
287 +
288 +Reverted the earlier round from 'next' and then fixed up further.
289
290 * ld/git-p4-expanded-keywords (2012-02-14) 1 commit
291 (merged to 'next' on 2012-02-16 at a9004c5)
@@ -304,3 +293,5 @@ The second one looked iffy.
293
294 Teach git-p4 to unexpand $RCS$-like keywords that are embedded in
295 tracked contents in order to reduce unnecessary merge conflicts.
296 +
297 +Is this ready for 'master'?