What's cooking (2009/03 #02)

Junio C Hamano committed Mar 5, 2009 at 02:00 UTC e11a22c6b7fd9abde703fbc71002fdb32d3759e9
1 file changed +259 -289
whats-cooking.txt
+259 -289
@@ -1,9 +1,9 @@
1 To: git@vger.kernel.org
2 -Subject: What's cooking in git.git (Mar 2009, #01; Tue, 03)
3 -X-master-at: 1d035f85649cbb0a4fcc3e341f7e854acf9cf3c1
4 -X-next-at: 23381c887989ea63153c562061b1c4703ee75baf
2 +Subject: What's cooking in git.git (Mar 2009, #02; Thu, 05)
3 +X-master-at: 836769e875ecf47e83137a39de5937652e0e861e
4 +X-next-at: c65e767e015b0309eb047b5a4d2b20dcaceeab22
5
6 -What's cooking in git.git (Mar 2009, #01; Tue, 03)
6 +What's cooking in git.git (Mar 2009, #02; Thu, 05)
7 --------------------------------------------------
8
9 Here are the topics that have been cooking. Commits prefixed with '-' are
@@ -14,196 +14,177 @@ holding onto them.
14 The topics list the commits in reverse chronological order. The topics
15 meant to be merged to the maintenance series have "maint-" in their names.
16
17 -The master branch is slightly past 1.6.2-rc2. It's been a week and I
18 -think I can declare the final on this Wednesday, *if* I am not too mired
19 -by the day job, but I do not know what would happen yet.
20 -
21 -As an experiment, 'next' and 'pu' stayed open during this release freeze;
22 -new topics have been accepted. I have to say that the experiment was a
23 -moderate success, and many topics in 'next' seem to be of fairly high
24 -quality already, which would mean that we will have a shorter cycle before
25 -1.6.3.
17 +The master branch is slightly past 1.6.2, but embarrasingly enough we
18 +already have a real commit on 'maint'.
19
20 ----------------------------------------------------------------
21 [New Topics]
22
30 -* jk/clone-post-checkout (Tue Mar 3 00:37:51 2009 -0500) 1 commit
31 - - clone: run post-checkout hook when checking out
32 -
33 -* hv/cvsimport (Mon Mar 2 18:59:36 2009 +0100) 1 commit
23 +* hv/cvsimport-tests (Mon Mar 2 18:59:36 2009 +0100) 1 commit
24 - cvsimport: add test illustrating a bug in cvsps
25
36 -* jc/push-to-create (Mon Mar 2 22:36:16 2009 -0800) 1 commit
37 - - Push to create
26 +Yet more tests without fixing anything...
27
39 -* fc/parseopt-config (Sat Feb 21 02:49:29 2009 +0200) 11 commits
40 - - git config: don't allow --get-color* and variable type
41 - - git config: don't allow extra arguments for -e or -l.
42 - - git config: don't allow multiple variable types
43 - - git config: don't allow multiple config file locations
44 - - git config: reorganize to use parseopt
45 - - git config: reorganize get_color*
46 - - git config: trivial rename in preparation for parseopt
47 - - git_config(): not having a per-repo config file is not an error
48 - + git config: trivial cleanup for editor action
49 - + git config: codestyle cleanups
50 - + config: Add new option to open an editor.
28 +* jc/clone-branch-rebase (Tue Mar 3 22:29:55 2009 -0800) 1 commit
29 + - Make git-clone respect branch.autosetuprebase
30
52 -* tr/format-patch-thread (Thu Feb 19 22:26:33 2009 +0100) 4 commits
53 - - format-patch: support deep threading
54 - - format-patch: thread as reply to cover letter even with in-reply-
55 - to
56 - - format-patch: track several references
57 - - format-patch: threading test reactivation
31 +This is rewrite of a patch from Pat Notz.
32
59 -* ns/pretty-format (Tue Feb 24 15:33:29 2009 +0200) 5 commits
60 - + bash completion: add --format= and --oneline options for "git log"
61 - + Add tests for git log --pretty, --format and --oneline.
62 - + Add --oneline that is a synonym to "--pretty=oneline --abbrev-
63 - commit"
64 - + Give short-hands to --pretty=tformat:%formatstring
65 - + Add --format that is a synonym to --pretty
33 +* mg/maint-submodule-normalize-path (Tue Mar 3 16:08:21 2009 +0100) 2 commits
34 + - git submodule: Fix adding of submodules at paths with ./, .. and //
35 + - git submodule: Add test cases for git submodule add
36
67 -* mh/cvsimport-tests (Mon Feb 23 06:08:14 2009 +0100) 5 commits
68 - - Add a test of "git cvsimport"'s handling of tags and branches
69 - - Add some tests of git-cvsimport's handling of vendor branches
70 - - Test contents of entire cvsimported "master" tree contents
71 - - Use CVS's -f option if available (ignore user's ~/.cvsrc file)
72 - - Start a library for cvsimport-related tests
37 +* kb/tracking-count-no-merges (Wed Mar 4 18:47:39 2009 +0100) 1 commit
38 + - stat_tracking_info(): only count real commits
39
74 -* tr/maint-1.6.0-send-email-irt (Sun Mar 1 23:45:41 2009 +0100) 1 commit
75 - - send-email: respect in-reply-to regardless of threading
40 +This gives the merge commits zero weight when talking about how many
41 +commits you have ahead (or behind) of the branch you are tracking. Even
42 +though I agree that they should carry much less weight than the "real"
43 +commits, because your repeated merge from the other branch does not really
44 +add any real value to the end result, giving them absolute zero weight
45 +somehow feels wrong. At least it shows that your have been _active_ on the
46 +branch. But I do not feel very strongly about it.
47
77 -* jc/maint-1.6.0-keep-pack (Sat Feb 28 00:37:19 2009 -0800) 6 commits
78 - + is_kept_pack(): final clean-up
79 - + Simplify is_kept_pack()
80 - + Consolidate ignore_packed logic more
81 - + has_sha1_kept_pack(): take "struct rev_info"
82 - + has_sha1_pack(): refactor "pretend these packs do not exist"
83 - interface
84 - + git-repack: resist stray environment variable
48 +* js/rebase-i-opt (Tue Mar 3 10:55:31 2009 +0100) 1 commit
49 + - rebase -i: avoid 'git reset' when possible
50
86 -* en/maint-hash-object (Sat Feb 28 12:56:49 2009 -0700) 1 commit
87 - + Ensure proper setup of git_dir for git-hash-object
51 +----------------------------------------------------------------
52 +[Will merge to 'master' soon]
53
89 -* el/blame-date (Fri Feb 20 14:51:11 2009 -0800) 1 commit
90 - - Make git blame's date output format configurable, like git log
54 +* jk/sane-relative-time (Tue Feb 24 00:42:16 2009 -0500) 1 commit
55 + + never fallback relative times to absolute
56
92 -* js/remote-improvements (Wed Feb 25 03:32:28 2009 -0500) 22 commits
93 - - builtin-remote: new show output style for push refspecs
94 - - builtin-remote: new show output style
95 - - remote: make guess_remote_head() use exact HEAD lookup if it is
96 - available
97 - - builtin-remote: add set-head subcommand
98 - - builtin-remote: teach show to display remote HEAD
99 - - builtin-remote: fix two inconsistencies in the output of "show
100 - <remote>"
101 - - builtin-remote: make get_remote_ref_states() always populate
102 - states.tracked
103 - - builtin-remote: rename variables and eliminate redundant function
104 - call
105 - - builtin-remote: remove unused code in get_ref_states
106 - - builtin-remote: refactor duplicated cleanup code
107 - - string-list: new for_each_string_list() function
108 - - remote: make match_refs() not short-circuit
109 - - remote: make match_refs() copy src ref before assigning to
110 - peer_ref
111 - - remote: let guess_remote_head() optionally return all matches
112 - - remote: make copy_ref() perform a deep copy
113 - - remote: simplify guess_remote_head()
114 - - move locate_head() to remote.c
115 - - move duplicated ref_newer() to remote.c
116 - - move duplicated get_local_heads() to remote.c
117 - - refactor find_ref_by_name() to accept const list
118 - - add basic http clone/fetch tests
119 - - test scripts: refactor start_httpd helper
57 +I think I sent out a "here is how" patch for something related to
58 +--date=<format> option that may be related to this topic; I seem to have
59 +lost it.
60
121 -* js/clone-depth-local (Fri Feb 27 00:04:06 2009 -0800) 3 commits
122 - . parse_options(): do not "increment" boolean
123 - . clone: ignore --depth when cloning locally (implicitly --local)
124 - . clone: do not ignore --no-local option
61 +* js/send-email (Mon Mar 2 23:52:18 2009 -0500) 5 commits
62 + + send-email: add --confirm option and configuration setting
63 + + send-email: don't create temporary compose file until it is needed
64 + + send-email: --suppress-cc improvements
65 + + send-email: handle multiple Cc addresses when reading mbox message
66 + + send-email: allow send-email to run outside a repo
67
126 -Jeff had a good suggestion for this series but it was tripped by
127 -a misfeature in parse_options().
68 +* sg/rerere-cleanup (Sat Feb 14 23:21:04 2009 +0100) 1 commit
69 + + rerere: remove duplicated functions
70
129 -* rs/memmem (Tue Mar 3 00:19:30 2009 +0100) 2 commits
130 - - optimize compat/ memmem()
131 - - diffcore-pickaxe: use memmem()
71 +* jc/add-p-unquote (Mon Feb 16 22:43:43 2009 -0800) 1 commit
72 + + git-add -i/-p: learn to unwrap C-quoted paths
73
133 -As always with patches from René, this is already next material.
74 +* jw/imap-preformatted-html (Thu Feb 12 08:58:12 2009 -0600) 1 commit
75 + + imap.preformattedHTML to tell Thunderbird to send non-flowed text
76
135 -* mv/parseopt-ls-files (Tue Feb 17 15:27:11 2009 +0100) 2 commits
136 - - parse-opt: migrate builtin-ls-files.
137 - - Turn the flags in struct dir_struct into a single variable
77 +The patch text should be identical to Jeremy's "Virtual Patch", except
78 +that the configuration variable was renamed per list discussion.
79
139 -* jk/sane-relative-time (Tue Feb 24 00:42:16 2009 -0500) 1 commit
140 - + never fallback relative times to absolute
80 +* jw/format-patch-attach (Thu Feb 12 09:51:55 2009 -0600) 1 commit
81 + + Enable setting attach as the default in .gitconfig for git-format-
82 + patch.
83
142 -* tv/rebase-stat (Sun Mar 1 22:28:28 2009 +0100) 2 commits
143 - - git-pull: Allow --stat and --no-stat to be used with --rebase
144 - - git-rebase: Add --stat and --no-stat for producing diffstat on
145 - rebase
84 +* sr/force-rebase (Fri Feb 13 23:48:01 2009 +0100) 1 commit
85 + + Teach rebase to rebase even if upstream is up to date
86
147 -----------------------------------------------------------------
148 -[Discarded]
87 +* fg/exclude-bq (Tue Feb 10 15:20:17 2009 +0100) 1 commit
88 + + Support "\" in non-wildcard exclusion entries
89
150 -* jc/fsck (Fri Jan 30 02:33:47 2009 -0800) 4 commits
151 - - fsck: three levels of validation
152 - - verify-pack: add --quick
153 - - verify_pack(): allow a quicker verification for a pack with
154 - version 2 idx
155 - - pack-check.c: minor formatting fix to match coding style
90 +* dm/add-i-edit-abort (Thu Feb 12 00:19:41 2009 -0500) 1 commit
91 + + add -i: revisit hunk on editor failure
92
157 -J6t has a good point that if this had any value then medium level should
158 -replace the default. I am tempted to actually dropping this as a failed
159 -experiment.
93 +* tp/completion (Sat Feb 21 15:48:43 2009 +0100) 6 commits
94 + + Fixup: Add bare repository indicator for __git_ps1
95 + + Add bare repository indicator for __git_ps1
96 + + completion: More fixes to prevent unbound variable errors
97 + + completion: Better __git_ps1 support when not in working directory
98 + + completion: Use consistent if [...] convention, not "test"
99 + + completion: For consistency, change "git rev-parse" to __gitdir
100 + calls
101
161 -* js/remote-set-head (Sat Feb 14 05:30:30 2009 -0500) 5 commits
162 - - builtin-remote: better handling of multiple remote HEADs
163 - - builtin-remote: add set-head subcommand
164 - - builtin-remote: teach show to display remote HEAD
165 - - builtin-remote: move duplicated cleanup code its own function
166 - - builtin-clone: move locate_head() to remote.c so it can be re-used
102 +* js/branch-symref (Wed Feb 18 22:34:44 2009 -0500) 4 commits
103 + + add basic branch display tests
104 + + branch: clean up repeated strlen
105 + + Avoid segfault with 'git branch' when the HEAD is detached
106 + + builtin-branch: improve output when displaying remote branches
107
168 -* jk/head-lookup (Sun Feb 15 01:18:18 2009 -0500) 5 commits
169 - - remote: use exact HEAD lookup if it is available
170 - - remote: refactor guess_remote_head
171 - - refactor find_refs_by_name to accept const list
172 - - add basic http clone/fetch tests
173 - - test scripts: refactor start_httpd helper
108 +* al/ansi-color (Fri Feb 13 22:53:41 2009 +0100) 2 commits
109 + + builtin-branch.c: Rename branch category color names
110 + + Clean up use of ANSI color sequences
111 +
112 +* js/valgrind (Thu Feb 5 22:03:00 2009 +0100) 9 commits
113 + + valgrind: do not require valgrind 3.4.0 or newer
114 + + test-lib: avoid assuming that templates/ are in the GIT_EXEC_PATH
115 + + Tests: let --valgrind imply --verbose and --tee
116 + + Add a script to coalesce the valgrind outputs
117 + + t/Makefile: provide a 'valgrind' target
118 + + test-lib.sh: optionally output to test-results/$TEST.out, too
119 + + Valgrind support: check for more than just programming errors
120 + + valgrind: ignore ldso and more libz errors
121 + + Add valgrind support in test scripts
122 +
123 +* fc/config-editor (Sat Feb 21 02:48:54 2009 +0200) 3 commits
124 + + git config: trivial cleanup for editor action
125 + + git config: codestyle cleanups
126 + + config: Add new option to open an editor.
127 +
128 +Rerolled and looked sane.
129 +
130 +* jc/blame (Wed Jun 4 22:58:40 2008 -0700) 2 commits
131 + + blame: show "previous" information in --porcelain/--incremental
132 + format
133 + + git-blame: refactor code to emit "porcelain format" output
134
175 -These two are now consolidated into Jay's remove-improvements series
176 -listed above.
135 +This gives Porcelains (like gitweb) the information on the commit _before_
136 +the one that the final blame is laid on, which should save them one
137 +rev-parse to dig further. Jakub seems to want this for gitweb.
138 +
139 +* ns/pretty-format (Tue Feb 24 15:33:29 2009 +0200) 5 commits
140 + + bash completion: add --format= and --oneline options for "git log"
141 + + Add tests for git log --pretty, --format and --oneline.
142 + + Add --oneline that is a synonym to "--pretty=oneline --abbrev-
143 + commit"
144 + + Give short-hands to --pretty=tformat:%formatstring
145 + + Add --format that is a synonym to --pretty
146
147 ----------------------------------------------------------------
179 -[Stalled and may need help and prodding to go forward]
148 +[Discarded]
149
181 -* gb/gitweb-base (Sun Feb 15 10:18:36 2009 +0100) 1 commit
182 - - gitweb: fix wrong base URL when non-root DirectoryIndex
150 +* js/clone-depth-local (Fri Feb 27 00:04:06 2009 -0800) 3 commits
151 + . parse_options(): do not "increment" boolean
152 + . clone: ignore --depth when cloning locally (implicitly --local)
153 + . clone: do not ignore --no-local option
154 +
155 +Jeff had a good suggestion for this series but it was tripped by
156 +a misfeature in parse_options().
157
158 * ns/stash-keep (Thu Feb 12 06:25:14 2009 +0900) 1 commit
185 - - stash: --keep option just saves
159 + . stash: --keep option just saves
160
187 -Do we want to keep this one?
161 +----------------------------------------------------------------
162 +[Stalled and may need help and prodding to go forward]
163
164 * lh/submodule-tree-traversal (Sun Jan 25 01:52:06 2009 +0100) 1 commit
165 - archive.c: add support for --submodules[=(all|checkedout)]
166
167 Discussion stalled on the submodule selection criteria.
168
194 -* jc/blame (Wed Jun 4 22:58:40 2008 -0700) 2 commits
195 - + blame: show "previous" information in --porcelain/--incremental
196 - format
197 - + git-blame: refactor code to emit "porcelain format" output
169 +* jc/merge-convert (Mon Jan 26 16:45:01 2009 -0800) 1 commit
170 + - git-merge-file: allow converting the results for the work tree
171
199 -This gives Porcelains (like gitweb) the information on the commit _before_
200 -the one that the final blame is laid on, which should save them one
201 -rev-parse to dig further. The line number in the "previous" information
202 -may need refining, and sanity checking code for reference counting may
203 -need to be resurrected before this can move forward.
172 +This is a feature waiting for a user.
173
205 -I thought recent tig discussion may blow new life into it, but is this
206 -unneeded? If so I'd rather revert it (or discard after 1.6.2).
174 +We did not give scripted Porcelains a way to say "this temporary file I am
175 +using for merging is for this path, so use the core.autocrlf and attributes
176 +rules for that final path". Instead, merge-file simply wrote out the
177 +data in the canonical repository representation.
178 +
179 +rerere has the same issue, but it is a lot worse. It reads the three
180 +files (preimage, postimage and thisimage) from the work tree in the work
181 +tree representation, merges them without converting them to the canonical
182 +representation first but inserts the conflict markers with the canonical
183 +representation and writes the resulting mess out. It needs to be fixed to
184 +read with convert_to_git(), merge them while they are still in the
185 +canonical representation and possibly add conflict markers, and then write
186 +the results out after convert_to_working_tree(). It also needs to write
187 +in binary mode as well.
188
189 * db/foreign-scm (Sun Jan 11 15:12:10 2009 -0500) 3 commits
190 - Support fetching from foreign VCSes
@@ -229,7 +210,7 @@ the other end is a native git repository is a very worthy goal.
210 - refs: add a "for_each_replace_ref" function
211
212 I think the code is much cleaner than the first round, but I am not
232 -convinced it is doing the right thing in the connectivity traverser.
213 +convinced it is doing the right thing in the connectivity traverser.
214 Independent review sorely needed.
215
216 * sc/gitweb-category (Fri Dec 12 00:45:12 2008 +0100) 3 commits
@@ -238,127 +219,140 @@ Independent review sorely needed.
219 - gitweb: Modularized git_get_project_description to be more generic
220
221 Design discussion between Jakub and Sebastien seems to have stalled, but
241 -Jakub seems to be taking this over.
222 +Jakub seems to be taking this over, so I'll probably discard these three
223 +shortly.
224
225 ----------------------------------------------------------------
244 -[Reverted]
245 -
246 -* mh/unify-color (Fri Jan 23 01:25:23 2009 -0800) 3 commits
247 - ? Revert previous two commits
248 - ? move the color variables to color.c
249 - ? handle color.ui at a central place
250 -
251 -This broke git-format-patch badly.
252 -
253 -* js/rebase-error-a-bit-more-verbose (Sun Feb 8 21:22:18 2009 -0800) 2 commits
254 - ? Revert "rebase: explain why when the HEAD could not be detached"
255 - ? rebase: explain why when the HEAD could not be detached
256 -
257 -This turned out to be unnecessary.
226 +[Actively cooking]
227
259 -* rs/maint-1.6.0-windows-ceiling (Sat Feb 7 12:40:40 2009 -0800) 2 commits
260 - ? Revert "fix t1504 on Windows"
261 - ? fix t1504 on Windows
228 +* kb/checkout-optim (Wed Mar 4 18:47:40 2009 +0100) 15 commits
229 + - better introduction of GIT with USE_NSEC defined
230 + + write_index(): update index_state->timestamp after flushing to
231 + disk
232 + + verify_uptodate(): add ce_uptodate(ce) test
233 + + make USE_NSEC work as expected
234 + + fix compile error when USE_NSEC is defined
235 + + check_updates(): effective removal of cache entries marked
236 + CE_REMOVE
237 + + lstat_cache(): print a warning if doing ping-pong between cache
238 + types
239 + + show_patch_diff(): remove a call to fstat()
240 + + write_entry(): use fstat() instead of lstat() when file is open
241 + + write_entry(): cleanup of some duplicated code
242 + + create_directories(): remove some memcpy() and strchr() calls
243 + + unlink_entry(): introduce schedule_dir_for_removal()
244 + + lstat_cache(): swap func(length, string) into func(string, length)
245 + + lstat_cache(): generalise longest_match_lstat_cache()
246 + + lstat_cache(): small cleanup and optimisation
247
263 -I'm giving a fresh start to J6t's series which contains this.
248 +* rs/memmem (Tue Mar 3 00:19:30 2009 +0100) 2 commits
249 + - optimize compat/ memmem()
250 + - diffcore-pickaxe: use memmem()
251
265 -* lh/reverted-submodule-tree-traversal (Sun Jan 25 18:39:55 2009 -0800) 4 commits
266 - ? Revert round #1 of the series
267 - ? builtin-ls-tree: enable traversal of submodules
268 - ? archive.c: enable traversal of submodules
269 - ? tree.c: add support for traversal of submodules
252 +As always with patches from René, this is already next material.
253
271 -I'm giving a fresh start to Lars's second iteration.
254 +* jc/push-to-create (Mon Mar 2 22:36:16 2009 -0800) 1 commit
255 + - Push to create
256
273 -----------------------------------------------------------------
274 -[Will merge to 'next' soon]
257 +This was a failed weatherbaloon patch to allow creation of a new
258 +repository from the remote side. Will discard.
259
276 -* js/send-email (Mon Mar 2 23:52:18 2009 -0500) 5 commits
277 - - send-email: add --confirm option and configuration setting
278 - + send-email: don't create temporary compose file until it is needed
279 - + send-email: --suppress-cc improvements
280 - + send-email: handle multiple Cc addresses when reading mbox message
281 - + send-email: allow send-email to run outside a repo
260 +* mv/parseopt-ls-files (Tue Feb 17 15:27:11 2009 +0100) 2 commits
261 + - parse-opt: migrate builtin-ls-files.
262 + - Turn the flags in struct dir_struct into a single variable
263
283 -----------------------------------------------------------------
284 -[Ready for 'master' after 1.6.2]
264 +* tv/rebase-stat (Sun Mar 1 22:28:28 2009 +0100) 2 commits
265 + - git-pull: Allow --stat and --no-stat to be used with --rebase
266 + - git-rebase: Add --stat and --no-stat for producing diffstat on
267 + rebase
268
286 -* sg/rerere-cleanup (Sat Feb 14 23:21:04 2009 +0100) 1 commit
287 - + rerere: remove duplicated functions
269 +* jk/clone-post-checkout (Tue Mar 3 00:37:51 2009 -0500) 1 commit
270 + - clone: run post-checkout hook when checking out
271
289 -* jc/add-p-unquote (Mon Feb 16 22:43:43 2009 -0800) 1 commit
290 - + git-add -i/-p: learn to unwrap C-quoted paths
272 +* js/remote-improvements (Wed Feb 25 03:32:28 2009 -0500) 22 commits
273 + - builtin-remote: new show output style for push refspecs
274 + - builtin-remote: new show output style
275 + - remote: make guess_remote_head() use exact HEAD lookup if it is
276 + available
277 + - builtin-remote: add set-head subcommand
278 + - builtin-remote: teach show to display remote HEAD
279 + - builtin-remote: fix two inconsistencies in the output of "show
280 + <remote>"
281 + - builtin-remote: make get_remote_ref_states() always populate
282 + states.tracked
283 + - builtin-remote: rename variables and eliminate redundant function
284 + call
285 + - builtin-remote: remove unused code in get_ref_states
286 + - builtin-remote: refactor duplicated cleanup code
287 + - string-list: new for_each_string_list() function
288 + - remote: make match_refs() not short-circuit
289 + - remote: make match_refs() copy src ref before assigning to
290 + peer_ref
291 + - remote: let guess_remote_head() optionally return all matches
292 + - remote: make copy_ref() perform a deep copy
293 + - remote: simplify guess_remote_head()
294 + - move locate_head() to remote.c
295 + - move duplicated ref_newer() to remote.c
296 + - move duplicated get_local_heads() to remote.c
297 + - refactor find_ref_by_name() to accept const list
298 + - add basic http clone/fetch tests
299 + - test scripts: refactor start_httpd helper
300
292 -* jw/imap-preformatted-html (Thu Feb 12 08:58:12 2009 -0600) 1 commit
293 - + imap.preformattedHTML to tell Thunderbird to send non-flowed text
301 +* fc/parseopt-config (Sat Feb 21 02:49:29 2009 +0200) 11 commits
302 + - git config: don't allow --get-color* and variable type
303 + - git config: don't allow extra arguments for -e or -l.
304 + - git config: don't allow multiple variable types
305 + - git config: don't allow multiple config file locations
306 + - git config: reorganize to use parseopt
307 + - git config: reorganize get_color*
308 + - git config: trivial rename in preparation for parseopt
309 + - git_config(): not having a per-repo config file is not an error
310 + + git config: trivial cleanup for editor action
311 + + git config: codestyle cleanups
312 + + config: Add new option to open an editor.
313
295 -The patch text should be identical to Jeremy's "Virtual Patch", except
296 -that the configuration variable was renamed per list discussion.
314 +* tr/format-patch-thread (Thu Feb 19 22:26:33 2009 +0100) 4 commits
315 + - format-patch: support deep threading
316 + - format-patch: thread as reply to cover letter even with in-reply-
317 + to
318 + - format-patch: track several references
319 + - format-patch: threading test reactivation
320
298 -* jw/format-patch-attach (Thu Feb 12 09:51:55 2009 -0600) 1 commit
299 - + Enable setting attach as the default in .gitconfig for git-format-
300 - patch.
321 +* el/blame-date (Fri Feb 20 14:51:11 2009 -0800) 1 commit
322 + - Make git blame's date output format configurable, like git log
323
302 -* sr/force-rebase (Fri Feb 13 23:48:01 2009 +0100) 1 commit
303 - + Teach rebase to rebase even if upstream is up to date
324 +I think the above seven series were basically Ok; I'll hopefully have time
325 +to re-read them and merge them to 'next'.
326
305 -* fg/exclude-bq (Tue Feb 10 15:20:17 2009 +0100) 1 commit
306 - + Support "\" in non-wildcard exclusion entries
327 +* mh/cvsimport-tests (Mon Feb 23 06:08:14 2009 +0100) 5 commits
328 + - Add a test of "git cvsimport"'s handling of tags and branches
329 + - Add some tests of git-cvsimport's handling of vendor branches
330 + - Test contents of entire cvsimported "master" tree contents
331 + - Use CVS's -f option if available (ignore user's ~/.cvsrc file)
332 + - Start a library for cvsimport-related tests
333
308 -* dm/add-i-edit-abort (Thu Feb 12 00:19:41 2009 -0500) 1 commit
309 - + add -i: revisit hunk on editor failure
334 +Tests without fixes are of dubious value. Any takers?
335
311 -* tp/completion (Sat Feb 21 15:48:43 2009 +0100) 6 commits
312 - + Fixup: Add bare repository indicator for __git_ps1
313 - + Add bare repository indicator for __git_ps1
314 - + completion: More fixes to prevent unbound variable errors
315 - + completion: Better __git_ps1 support when not in working directory
316 - + completion: Use consistent if [...] convention, not "test"
317 - + completion: For consistency, change "git rev-parse" to __gitdir
318 - calls
336 +* tr/maint-1.6.0-send-email-irt (Sun Mar 1 23:45:41 2009 +0100) 1 commit
337 + - send-email: respect in-reply-to regardless of threading
338
320 -* js/branch-symref (Wed Feb 18 22:34:44 2009 -0500) 4 commits
321 - + add basic branch display tests
322 - + branch: clean up repeated strlen
323 - + Avoid segfault with 'git branch' when the HEAD is detached
324 - + builtin-branch: improve output when displaying remote branches
339 +Tests?
340
326 -* al/ansi-color (Fri Feb 13 22:53:41 2009 +0100) 2 commits
327 - + builtin-branch.c: Rename branch category color names
328 - + Clean up use of ANSI color sequences
341 +* jc/maint-1.6.0-keep-pack (Sat Feb 28 00:37:19 2009 -0800) 6 commits
342 + + is_kept_pack(): final clean-up
343 + + Simplify is_kept_pack()
344 + + Consolidate ignore_packed logic more
345 + + has_sha1_kept_pack(): take "struct rev_info"
346 + + has_sha1_pack(): refactor "pretend these packs do not exist"
347 + interface
348 + + git-repack: resist stray environment variable
349
330 -* kb/checkout-optim (Mon Feb 23 19:02:57 2009 +0100) 14 commits
331 - + write_index(): update index_state->timestamp after flushing to
332 - disk
333 - + verify_uptodate(): add ce_uptodate(ce) test
334 - + make USE_NSEC work as expected
335 - + fix compile error when USE_NSEC is defined
336 - + check_updates(): effective removal of cache entries marked
337 - CE_REMOVE
338 - + lstat_cache(): print a warning if doing ping-pong between cache
339 - types
340 - + show_patch_diff(): remove a call to fstat()
341 - + write_entry(): use fstat() instead of lstat() when file is open
342 - + write_entry(): cleanup of some duplicated code
343 - + create_directories(): remove some memcpy() and strchr() calls
344 - + unlink_entry(): introduce schedule_dir_for_removal()
345 - + lstat_cache(): swap func(length, string) into func(string, length)
346 - + lstat_cache(): generalise longest_match_lstat_cache()
347 - + lstat_cache(): small cleanup and optimisation
350 +This is in response to Linus's "Really slow 'git gc'" ($gmane/110743)
351
349 -* js/valgrind (Thu Feb 5 22:03:00 2009 +0100) 9 commits
350 - + valgrind: do not require valgrind 3.4.0 or newer
351 - + test-lib: avoid assuming that templates/ are in the GIT_EXEC_PATH
352 - + Tests: let --valgrind imply --verbose and --tee
353 - + Add a script to coalesce the valgrind outputs
354 - + t/Makefile: provide a 'valgrind' target
355 - + test-lib.sh: optionally output to test-results/$TEST.out, too
356 - + Valgrind support: check for more than just programming errors
357 - + valgrind: ignore ldso and more libz errors
358 - + Add valgrind support in test scripts
352 +* en/maint-hash-object (Sat Feb 28 12:56:49 2009 -0700) 1 commit
353 + + Ensure proper setup of git_dir for git-hash-object
354
360 -----------------------------------------------------------------
361 -[Actively cooking]
355 +Obvious fix that can also go to 1.6.1.X
356
357 * tr/gcov (Thu Feb 19 12:13:42 2009 +0100) 8 commits
358 - Test git-patch-id
@@ -370,30 +364,24 @@ that the configuration variable was renamed per list discussion.
364 - Test that diff can read from stdin
365 - Support coverage testing with GCC/gcov
366
373 -* js/notes (Sat Feb 14 21:23:25 2009 +0100) 13 commits
367 +* js/notes (Wed Feb 18 11:17:27 2009 -0800) 14 commits
368 + - tests: fix "export var=val"
369 - notes: refuse to edit notes outside refs/notes/
370 - t3301: use test_must_fail instead of !
371 - t3301: fix confusing quoting in test for valid notes ref
372 - notes: use GIT_EDITOR and core.editor over VISUAL/EDITOR
373 - notes: only clean up message file when editing
374 - handle empty notes gracefully
380 - + git notes show: test empty notes
381 - + git-notes: fix printing of multi-line notes
382 - + notes: fix core.notesRef documentation
383 - + Add an expensive test for git-notes
384 - + Speed up git notes lookup
385 - + Add a script to edit/inspect notes
386 - + Introduce commit notes
375 + - git notes show: test empty notes
376 + - git-notes: fix printing of multi-line notes
377 + - notes: fix core.notesRef documentation
378 + - Add an expensive test for git-notes
379 + - Speed up git notes lookup
380 + - Add a script to edit/inspect notes
381 + - Introduce commit notes
382
388 -Earlier part was merged to master and then reverted there. Will be
389 -rebased 1.6.2 to keep my sanity.
390 -
391 -* fc/config-editor (Sat Feb 21 02:48:54 2009 +0200) 3 commits
392 - + git config: trivial cleanup for editor action
393 - + git config: codestyle cleanups
394 - + config: Add new option to open an editor.
395 -
396 -Rerolled and looked sane.
383 +Rebased and then kicked back to 'pu' to give the author a chance to
384 +rearrange if necessary. I might do some trivial squashing myself.
385
386 ----------------------------------------------------------------
387 [On Hold]
@@ -405,7 +393,8 @@ Rerolled and looked sane.
393 - Refuse updating the current branch in a non-bare repository via
394 push
395
408 -This is for 1.7.0.
396 +These are for 1.7.0, but the messages when they trigger together may need
397 +to be rethought.
398
399 * jc/commit-assume-also-during-merge (Thu Jan 22 22:21:49 2009 -0800) 3 commits
400 - git commit: pathspec without -i/-o implies -i semantics during a
@@ -414,23 +403,4 @@ This is for 1.7.0.
403 - Add "partial commit" tests during a conflicted merge
404
405 This was only meant as a weatherballoon to help facilitate discussion.
417 -
418 -* jc/merge-convert (Mon Jan 26 16:45:01 2009 -0800) 1 commit
419 - - git-merge-file: allow converting the results for the work tree
420 -
421 -This is a feature waiting for a user. I do not need it in 1.6.2
422 -
423 -We did not give scripted Porcelains a way to say "this temporary file I am
424 -using for merging is for this path, so use the core.autocrlf and attributes
425 -rules for that final path". Instead, merge-file simply wrote out the
426 -data in the canonical repository representation.
427 -
428 -rerere has the same issue, but it is a lot worse. It reads the three
429 -files (preimage, postimage and thisimage) from the work tree in the work
430 -tree representation, merges them without converting them to the canonical
431 -representation first but inserts the conflict markers with the canonical
432 -representation and writes the resulting mess out. It needs to be fixed to
433 -read with convert_to_git(), merge them while they are still in the
434 -canonical representation and possibly add conflict markers, and then write
435 -the results out after convert_to_working_tree(). It also needs to write
436 -in binary mode as well.
406 +Will be discarded.