What's in/cooking

Junio C Hamano committed Nov 12, 2008 at 18:00 UTC 9f43e271de64695391242f846700cfc32b32e357
2 files changed +369
whats/cooking/2008/11/04.txt new
+197
@@ -0,0 +1,197 @@
1 +Subject: What's cooking in git.git (Nov 2008, #04; Wed, 12)
2 +X-master-at: 171d7661eda111d3e35f6e8097a1a3a07b30026c
3 +X-next-at: c990ab59beabbf7e62799d0a0190cc0a5171aba2
4 +
5 +What's cooking in git.git (Nov 2008, #04; Wed, 12)
6 +--------------------------------------------------
7 +
8 +Here are the topics that have been cooking. Commits prefixed
9 +with '-' are only in 'pu' while commits prefixed with '+' are
10 +in 'next'.
11 +
12 +The topics list the commits in reverse chronological order. The topics
13 +meant to be merged to the maintenance series have "maint-" in their names.
14 +
15 +I think we should declare -rc0 by the end of this week, with the goal of
16 +including everything that are in 'next' as of tonight, plus the "narrow
17 +checkout" and possibly Pierre's send-email that acts as format-patch
18 +frontend, by the time we hit -rc1, perhaps soon after the end of month.
19 +
20 +----------------------------------------------------------------
21 +[New Topics]
22 +
23 +* mk/maint-cg-push (Mon Nov 10 22:47:11 2008 +0100) 1 commit
24 + + git push: Interpret $GIT_DIR/branches in a Cogito compatible way
25 +
26 +* jk/commit-v-strip (Wed Nov 12 03:23:37 2008 -0500) 5 commits
27 + + status: show "-v" diff even for initial commit
28 + + Merge branch 'jk/maint-commit-v-strip' into jk/commit-v-strip
29 + + commit: Fix stripping of patch in verbose mode.
30 + + wt-status: refactor initial commit printing
31 + + define empty tree sha1 as a macro
32 +
33 +* jk/maint-commit-v-strip (Wed Nov 12 03:25:52 2008 -0500) 1 commit
34 + + commit: Fix stripping of patch in verbose mode.
35 +
36 +* ph/send-email (Tue Nov 11 00:54:02 2008 +0100) 4 commits
37 + - git send-email: ask less questions when --compose is used.
38 + - git send-email: add --annotate option
39 + - git send-email: interpret unknown files as revision lists
40 + - git send-email: make the message file name more specific.
41 +
42 +* nd/rename-cache (Sat Nov 8 18:27:33 2008 +0700) 2 commits
43 + - diffcore-rename: add config option to allow to cache renames
44 + - diffcore-rename: support rename cache
45 +
46 +----------------------------------------------------------------
47 +[Will merge to "master" soon]
48 +
49 +* jk/deny-push-to-current (Sat Nov 8 20:49:27 2008 -0500) 2 commits
50 + + receive-pack: detect push to current branch of non-bare repo
51 + + t5516: refactor oddball tests
52 +
53 +* dl/xdiff (Fri Nov 7 21:24:33 2008 -0800) 1 commit
54 + + xdiff: give up scanning similar lines early
55 +
56 +This performance fix may eventually need to be cherry-picked to 'maint'.
57 +
58 +* lt/decorate (Wed Nov 12 11:51:28 2008 +0100) 7 commits
59 + + rev-list documentation: clarify the two parts of history
60 + simplification
61 + + Document "git log --simplify-by-decoration"
62 + + Document "git log --source"
63 + + revision traversal: '--simplify-by-decoration'
64 + + Make '--decorate' set an explicit 'show_decorations' flag
65 + + revision: make tree comparison functions take commits rather than
66 + trees
67 + + Add a 'source' decorator for commits
68 +
69 +* gb/gitweb-snapshot-pathinfo (Sun Nov 2 10:21:39 2008 +0100) 3 commits
70 + + gitweb: embed snapshot format parameter in PATH_INFO
71 + + gitweb: retrieve snapshot format from PATH_INFO
72 + + gitweb: make the supported snapshot formats array global
73 +
74 +Got sick of waiting for re-Ack.
75 +
76 +* jn/gitweb-customlinks (Sun Oct 12 00:02:32 2008 +0200) 1 commit
77 + + gitweb: Better processing format string in custom links in navbar
78 +
79 +Got sick of waiting for some sort of response from Pasky.
80 +
81 +* st/tag (Wed Nov 5 00:20:36 2008 +0100) 2 commits
82 + + tag: Add more tests about mixing incompatible modes and options
83 + + tag: Check that options are only allowed in the appropriate mode
84 +
85 +* ds/uintmax-config (Mon Nov 3 09:14:28 2008 -0900) 5 commits
86 + - autoconf: Enable threaded delta search when pthreads are supported
87 + + Add autoconf tests for pthreads
88 + + Make Pthread link flags configurable
89 + + Add Makefile check for FreeBSD 4.9-SECURITY
90 + + Build: add NO_UINTMAX_T to support ancient systems
91 +
92 +I split the part that unconditionally enable threaded delta search on any
93 +platform on which Pthread library is detected to be available from the
94 +topmost one. Will merge the rest to 'master' soon.
95 +
96 +* jk/diff-convfilter (Sun Oct 26 00:50:02 2008 -0400) 8 commits
97 + + enable textconv for diff in verbose status/commit
98 + + wt-status: load diff ui config
99 + + only textconv regular files
100 + + userdiff: require explicitly allowing textconv
101 + + refactor userdiff textconv code
102 + + add userdiff textconv tests
103 + + document the diff driver textconv feature
104 + + diff: add missing static declaration
105 +
106 +* jk/diff-convfilter-test-fix (Fri Oct 31 01:09:13 2008 -0400) 4 commits
107 + + Avoid using non-portable `echo -n` in tests.
108 + + add userdiff textconv tests
109 + + document the diff driver textconv feature
110 + + diff: add missing static declaration
111 +
112 +An update to the one above.
113 +
114 +----------------------------------------------------------------
115 +[Actively Cooking]
116 +
117 +* bc/maint-keep-pack (Sun Nov 9 23:59:58 2008 -0600) 9 commits
118 + + pack-objects: extend --local to mean ignore non-local loose
119 + objects too
120 + + sha1_file.c: split has_loose_object() into local and non-local
121 + counterparts
122 + + t7700: demonstrate mishandling of loose objects in an alternate
123 + ODB
124 + + builtin-gc.c: use new pack_keep bitfield to detect .keep file
125 + existence
126 + + repack: do not fall back to incremental repacking with [-a|-A]
127 + + repack: don't repack local objects in packs with .keep file
128 + + pack-objects: new option --honor-pack-keep
129 + + packed_git: convert pack_local flag into a bitfield and add
130 + pack_keep
131 + + t7700: demonstrate mishandling of objects in packs with a .keep
132 + file
133 +
134 +A few commits were rewound from 'next' and replaced with new ones.
135 +
136 +* np/pack-safer (Sun Nov 9 13:11:06 2008 -0800) 11 commits
137 + + t5303: fix printf format string for portability
138 + + t5303: work around printf breakage in dash
139 + + pack-objects: don't leak pack window reference when splitting
140 + packs
141 + + extend test coverage for latest pack corruption resilience
142 + improvements
143 + + pack-objects: allow "fixing" a corrupted pack without a full
144 + repack
145 + + make find_pack_revindex() aware of the nasty world
146 + + make check_object() resilient to pack corruptions
147 + + make packed_object_info() resilient to pack corruptions
148 + + make unpack_object_header() non fatal
149 + + better validation on delta base object offsets
150 + + close another possibility for propagating pack corruption
151 +
152 +* mv/remote-rename (Mon Nov 10 21:43:03 2008 +0100) 4 commits
153 + + git-remote: document the migration feature of the rename
154 + subcommand
155 + + git-remote rename: migrate from remotes/ and branches/
156 + + remote: add a new 'origin' variable to the struct
157 + + Implement git remote rename
158 +
159 +* jc/blame (Wed Jun 4 22:58:40 2008 -0700) 2 commits
160 + + blame: show "previous" information in --porcelain/--incremental
161 + format
162 + + git-blame: refactor code to emit "porcelain format" output
163 +
164 +* nd/narrow (Wed Oct 1 11:04:09 2008 +0700) 9 commits
165 + - grep: skip files outside sparse checkout area
166 + - checkout_entry(): CE_NO_CHECKOUT on checked out entries.
167 + - Prevent diff machinery from examining worktree outside sparse
168 + checkout
169 + - ls-files: Add tests for --sparse and friends
170 + - update-index: add --checkout/--no-checkout to update
171 + CE_NO_CHECKOUT bit
172 + - update-index: refactor mark_valid() in preparation for new options
173 + - ls-files: add options to support sparse checkout
174 + - Introduce CE_NO_CHECKOUT bit
175 + - Extend index to save more flags
176 +
177 +Early parts looked Ok; I am trying to find time to review the whole thing
178 +but fixing bugs in topics from other people here and there took almost all
179 +my time this weekend. Not good.
180 +
181 +----------------------------------------------------------------
182 +[On Hold]
183 +
184 +* jc/send-pack-tell-me-more (Thu Mar 20 00:44:11 2008 -0700) 1 commit
185 + - "git push": tellme-more protocol extension
186 +
187 +This seems to have a deadlock during communication between the peers.
188 +Someone needs to pick up this topic and resolve the deadlock before it can
189 +continue.
190 +
191 +* jk/renamelimit (Sat May 3 13:58:42 2008 -0700) 1 commit
192 + - diff: enable "too large a rename" warning when -M/-C is explicitly
193 + asked for
194 +
195 +This would be the right thing to do for command line use,
196 +but gitk will be hit due to tcl/tk's limitation, so I am holding
197 +this back for now.
whats/in/2008/11/02.txt new
+172
@@ -0,0 +1,172 @@
1 +Subject: What's in git.git (Nov 2008, #02; Wed, 12)
2 +X-maint-at: c14639f7b14d3d856c7c1f6c097dd1cb9615a5d0
3 +X-master-at: 171d7661eda111d3e35f6e8097a1a3a07b30026c
4 +X-maint-was: 1a9016aae5aa8c468bb2d8d5dd16303d5c005869
5 +X-master-was: 6a509a6f7f38906996ac791449d5bcc2f32eef23
6 +
7 +What's in git.git (Nov 2008, #02; Wed, 12)
8 +
9 + maint c14639f (Start 1.6.0.5 cycle)
10 + master 171d766 (git.html: Update the links to stale versions)
11 +------------------------------------------------------------------------
12 +
13 +BLURB HERE
14 +
15 +* The 'maint' branch has these fixes since the last announcement.
16 +
17 +Alex Riesen (7):
18 + Add mksnpath which allows you to specify the output buffer
19 + Fix mkpath abuse in dwim_ref and dwim_log of sha1_name.c
20 + Fix potentially dangerous uses of mkpath and git_path
21 + Add git_snpath: a .git path formatting routine with output buffer
22 + Fix potentially dangerous use of git_path in ref.c
23 + git_pathdup: returns xstrdup-ed copy of the formatted path
24 + Use git_pathdup instead of xstrdup(git_path(...))
25 +
26 +Alexandre Julliard (2):
27 + checkout: Don't crash when switching away from an invalid branch.
28 + git-submodule: Avoid printing a spurious message.
29 +
30 +Christian Couder (2):
31 + Documentation: rev-list: change a few instances of "git-cmd" to "git cmd"
32 + Documentation: bisect: change a few instances of "git-cmd" to "git cmd"
33 +
34 +Christian Jaeger (1):
35 + Git.pm: do not break inheritance
36 +
37 +Clemens Buchacher (2):
38 + do not force write of packed refs
39 + push: fix local refs update if already up-to-date
40 +
41 +Daniel Lowe (1):
42 + Fix non-literal format in printf-style calls
43 +
44 +Jeff King (1):
45 + refactor handling of "other" files in ls-files and status
46 +
47 +Johannes Schindelin (1):
48 + Fix fetch/pull when run without --update-head-ok
49 +
50 +Junio C Hamano (11):
51 + pull: allow "git pull origin $something:$current_branch" into an unborn
52 + branch
53 + Fix "checkout --track -b newbranch" on detached HEAD
54 + reset --hard/read-tree --reset -u: remove unmerged new paths
55 + demonstrate breakage of detached checkout with symbolic link HEAD
56 + Enhance hold_lock_file_for_{update,append}() API
57 + Update draft 1.6.0.4 release notes
58 + format-patch documentation: mention the special case of showing a single
59 + commit
60 + GIT 1.6.0.4
61 + Makefile: help people who run 'make check' by mistake
62 + checkout: Fix "initial checkout" detection
63 + Start 1.6.0.5 cycle
64 +
65 +Matt Kraai (1):
66 + Remove the period after the git-check-attr summary
67 +
68 +Mikael Magnusson (1):
69 + Document that git-log takes --all-match.
70 +
71 +Miklos Vajna (5):
72 + Fix git branch -m for symrefs.
73 + rename_ref(): handle the case when the reflog of a ref does not exist
74 + Fix git update-ref --no-deref -d.
75 + git branch -m: forbid renaming of a symref
76 + update-ref --no-deref -d: handle the case when the pointed ref is packed
77 +
78 +Nicolas Pitre (1):
79 + Fix pack.packSizeLimit and --max-pack-size handling
80 +
81 +Quy Tonthat (1):
82 + Update RPM spec for the new location of git-cvsserver.
83 +
84 +Stefan Naewe (1):
85 + git ls-remote: make usage string match manpage
86 +
87 +Yann Dirson (1):
88 + Add reference for status letters in documentation.
89 +
90 +
91 +* The 'master' branch has these since the last announcement
92 + in addition to the above.
93 +
94 +Alex Riesen (1):
95 + Fix potentially dangerous uses of mkpath and git_path
96 +
97 +Alexander Gavrilov (5):
98 + builtin-blame: Reencode commit messages according to git-log rules.
99 + Windows: Make OpenSSH properly detect tty detachment.
100 + gitweb: Use single implementation of export_ok check.
101 + gitweb: Add a per-repository authorization hook.
102 + gitweb: Fix mod_perl support.
103 +
104 +Brian Downing (2):
105 + Always initialize xpparam_t to 0
106 + Allow alternate "low-level" emit function from xdl_diff
107 +
108 +Christian Couder (1):
109 + bisect: fix missing "exit"
110 +
111 +Daniel Lowe (1):
112 + Fix non-literal format in printf-style calls
113 +
114 +David M. Syzdek (1):
115 + autoconf: Add link tests to each AC_CHECK_FUNC() test
116 +
117 +David Symonds (1):
118 + git-diff: Add --staged as a synonym for --cached.
119 +
120 +Giuseppe Bilotta (5):
121 + gitweb: parse project/action/hash_base:filename PATH_INFO
122 + gitweb: generate project/action/hash URLs
123 + gitweb: use_pathinfo filenames start with /
124 + gitweb: parse parent..current syntax from PATH_INFO
125 + gitweb: generate parent..current URLs
126 +
127 +Jeff King (1):
128 + t5400: expect success for denying deletion
129 +
130 +John Chapman (2):
131 + git-p4: Support purged files and optimize memory usage
132 + git-p4: Cache git config for performance
133 +
134 +Junio C Hamano (6):
135 + Fix reading of cloud tags
136 + rebase-i-p: minimum fix to obvious issues
137 + git-rebase--interactive.sh: comparision with == is bashism
138 + Update draft release notes to 1.6.1
139 + Update draft release notes to 1.6.1
140 + git.html: Update the links to stale versions
141 +
142 +Markus Heidelberg (1):
143 + Makefile: add install-man rules (quick and normal)
144 +
145 +Miklos Vajna (1):
146 + parse-opt: migrate builtin-checkout-index.
147 +
148 +Nicolas Pitre (5):
149 + index-pack: rationalize delta resolution code
150 + index-pack: smarter memory usage during delta resolution
151 + fix multiple issues in index-pack
152 + improve index-pack tests
153 + index-pack: don't leak leaf delta result
154 +
155 +Pete Harlan (2):
156 + contrib/hooks/post-receive-email: Put rev display in separate function
157 + contrib/hooks/post-receive-email: Make revision display configurable
158 +
159 +René Scharfe (3):
160 + blame: inline get_patch()
161 + add xdi_diff_hunks() for callers that only need hunk lengths
162 + blame: use xdi_diff_hunks(), get rid of struct patch
163 +
164 +Stephen Haberman (7):
165 + rebase-i-p: test to exclude commits from todo based on its parents
166 + rebase-i-p: use HEAD for updating the ref instead of mapping OLDHEAD
167 + rebase-i-p: delay saving current-commit to REWRITTEN if squashing
168 + rebase-i-p: fix 'no squashing merges' tripping up non-merges
169 + rebase-i-p: only list commits that require rewriting in todo
170 + rebase-i-p: do not include non-first-parent commits touching UPSTREAM
171 + rebase-i-p: if todo was reordered use HEAD as the rewritten parent
172 +