1 To: git@vger.kernel.org
2 Subject: What's cooking in git.git (Jan 2012, #03; Wed, 11)
3 X-master-at: 0e1cfc52de002e2d9b0e6562e8672fee3bf45a67
4 X-next-at: 5c0bfa96c2f67a7ac76afe5096f622dced39734f
6 What's cooking in git.git (Jan 2012, #03; Wed, 11)
7 --------------------------------------------------
9 Here are the topics that have been cooking. Commits prefixed with '-' are
10 only in 'pu' (proposed updates) while commits prefixed with '+' are in
13 Here are the repositories that have my integration branches:
15 With maint, master, next, pu, todo:
17 git://git.kernel.org/pub/scm/git/git.git
18 git://repo.or.cz/alt-git.git
19 https://code.google.com/p/git-core/
20 https://github.com/git/git
22 With only maint and master:
24 git://git.sourceforge.jp/gitroot/git-core/git.git
25 git://git-core.git.sourceforge.net/gitroot/git-core/git-core
27 With all the topics and integration branches:
29 https://github.com/gitster/git
31 The preformatted documentation in HTML and man format are found in:
33 git://git.kernel.org/pub/scm/git/git-{htmldocs,manpages}.git/
34 git://repo.or.cz/git-{htmldocs,manpages}.git/
35 https://code.google.com/p/git-{htmldocs,manpages}.git/
36 https://github.com/gitster/git-{htmldocs,manpages}.git/
38 --------------------------------------------------
41 * jc/pull-signed-tag (2012-01-11) 1 commit
42 - merge: use editor by default in interactive sessions
44 Per Linus's strong suggestion, sugarcoated (aka "taking blame for the
45 original UI screw-ups") so that it is easier for me to swallow and accept
46 a potentially huge backward incompatibility issue, "git merge" is made to
47 launch an editor to explain the merge in the merge commit by default in
50 * jc/request-pull-show-head-4 (2012-01-10) 1 commit
51 (merged to 'next' on 2012-01-11 at 8d98a6b)
52 + request-pull: use the real fork point when preparing the message
54 Hopefully the final finishing touch to the request-pull script that was
55 updated during this cycle.
57 Will merge by 1.7.9-rc1.
59 * jk/maint-upload-archive (2012-01-11) 1 commit
60 (merged to 'next' on 2012-01-11 at 5c0bfa9)
61 + archive: re-allow HEAD:Documentation on a remote invocation
63 Running "git archive" remotely and asking for a partial tree of a ref,
64 e.g. HEAD:$path was forbidden by a recent change to tighten security, but
65 was found to be overly restrictive.
67 Will merge by 1.7.9-rc1.
69 * nd/commit-ignore-i-t-a (2012-01-11) 2 commits
70 - commit: add --skip-intent-to-add to allow commit with i-t-a entries in index
71 - cache-tree: update API to take abitrary flags
73 * nd/maint-refname-in-hierarchy-check (2012-01-11) 1 commit
74 - Fix incorrect ref namespace check
76 Avoid getting confused by "ref/headxxx" and mistaking it as if it is under
77 the "refs/heads/" hierarchy.
80 * rr/sequencer (2012-01-11) 2 commits
81 - sequencer: factor code out of revert builtin
82 - revert: prepare to move replay_action to header
84 Moving large chunk of code out of cherry-pick/revert for later reuse,
85 primarily to prepare for the next cycle.
88 * ss/maint-msys-cvsexportcommit (2012-01-11) 2 commits
89 (merged to 'next' on 2012-01-11 at 007aab1)
90 + git-cvsexportcommit: Fix calling Perl's rel2abs() on MSYS
91 + t9200: On MSYS, do not pass Windows-style paths to CVS
93 Will merge by 1.7.9-rc1.
95 * tr/maint-mailinfo (2012-01-11) 1 commit
96 (merged to 'next' on 2012-01-11 at d9e344a)
97 + mailinfo documentation: accurately describe non -k case
99 Will merge by 1.7.9-rc1.
101 * jk/credentials (2012-01-11) 3 commits
102 - unix-socket: do not let close() or chdir() clobber errno during cleanup
103 - credential-cache: report more daemon connection errors
104 - unix-socket: handle long socket pathnames
106 * pw/p4-view-updates (2012-01-11) 5 commits
107 - git-p4: add tests demonstrating spec overlay ambiguities
108 - git-p4: adjust test to adhere to stricter useClientSpec
109 - git-p4: clarify comment
110 - git-p4: fix verbose comment typo
111 - git-p4: only a single ... wildcard is supported
113 --------------------------------------------------
114 [Graduated to "master"]
116 * bw/maint-t8006-sed-incomplete-line (2012-01-03) 1 commit
117 + Work around sed portability issue in t8006-blame-textconv
119 --------------------------------------------------
122 * jc/advise-push-default (2011-12-18) 1 commit
123 - push: hint to use push.default=upstream when appropriate
125 Peff had a good suggestion outlining an updated code structure so that
126 somebody new can try to dip his or her toes in the development. Any
129 Waiting for a reroll.
131 * jc/split-blob (2011-12-01) 6 commits
132 . WIP (streaming chunked)
133 - chunked-object: fallback checkout codepaths
134 - bulk-checkin: support chunked-object encoding
135 - bulk-checkin: allow the same data to be multiply hashed
136 - new representation types in the packstream
137 - varint-in-pack: refactor varint encoding/decoding
141 At least pack-objects and fsck need to learn the new encoding for the
142 series to be usable locally, and then index-pack/unpack-objects needs to
143 learn it to be used remotely.
145 * jc/advise-i18n (2011-12-22) 1 commit
146 - i18n of multi-line advice messages
148 Allow localization of advice messages that tend to be longer and
149 multi-line formatted. For now this is deliberately limited to advise()
150 interface and not vreportf() in general as touching the latter has
151 interactions with error() that has plumbing callers whose prefix "error: "
152 should never be translated.
154 --------------------------------------------------
157 * rs/diff-postimage-in-context (2012-01-06) 1 commit
158 (merged to 'next' on 2012-01-09 at 9635032)
159 + xdiff: print post-image for common records instead of pre-image
164 * cb/push-quiet (2012-01-08) 3 commits
165 - t5541: avoid TAP test miscounting
166 - fix push --quiet: add 'quiet' capability to receive-pack
167 - server_supports(): parse feature list more carefully
172 * nd/clone-detached (2012-01-11) 11 commits
173 - clone: print advice on checking out detached HEAD
174 - clone: allow --branch to take a tag
175 - clone: refuse to clone if --branch points to bogus ref
176 - clone: --branch=<branch> always means refs/heads/<branch>
178 - clone: delay cloning until after remote HEAD checking
179 - clone: factor out remote ref writing
180 - clone: factor out HEAD update code
181 - clone: factor out checkout code
182 - clone: write detached HEAD in bare repositories
183 - t5601: add missing && cascade
185 Looking good, but I may have screwed up the merge of this into 'pu'.
188 * nd/clone-single-branch (2012-01-08) 1 commit
189 (merged to 'next' on 2012-01-09 at 6c3c759)
190 + clone: add --single-branch to fetch only one branch
195 * jn/gitweb-unspecified-action (2012-01-09) 1 commit
196 - gitweb: Fix actionless dispatch for non-existent objects
201 * nd/index-pack-no-recurse (2012-01-09) 3 commits
202 - index-pack: eliminate unlimited recursion in get_delta_base()
203 - index-pack: eliminate recursion in find_unresolved_deltas
204 - Eliminate recursion in setting/clearing marks in commit list
206 The first one looked sensible; I am not sure if the second and third ones
207 take the right approach.
209 Will defer till the next cycle.
211 * mh/ref-api-rest (2011-12-12) 35 commits
212 - repack_without_ref(): call clear_packed_ref_cache()
213 - read_packed_refs(): keep track of the directory being worked in
214 - is_refname_available(): query only possibly-conflicting references
215 - refs: read loose references lazily
216 - read_loose_refs(): take a (ref_entry *) as argument
217 - struct ref_dir: store a reference to the enclosing ref_cache
218 - sort_ref_dir(): take (ref_entry *) instead of (ref_dir *)
219 - do_for_each_ref_in_dir*(): take (ref_entry *) instead of (ref_dir *)
220 - add_entry(): take (ref_entry *) instead of (ref_dir *)
221 - search_ref_dir(): take (ref_entry *) instead of (ref_dir *)
222 - find_containing_direntry(): use (ref_entry *) instead of (ref_dir *)
223 - add_ref(): take (ref_entry *) instead of (ref_dir *)
224 - read_packed_refs(): take (ref_entry *) instead of (ref_dir *)
225 - find_ref(): take (ref_entry *) instead of (ref_dir *)
226 - is_refname_available(): take (ref_entry *) instead of (ref_dir *)
227 - get_loose_refs(): return (ref_entry *) instead of (ref_dir *)
228 - get_packed_refs(): return (ref_entry *) instead of (ref_dir *)
229 - refs: wrap top-level ref_dirs in ref_entries
230 - get_ref_dir(): keep track of the current ref_dir
231 - do_for_each_ref(): only iterate over the subtree that was requested
232 - refs: sort ref_dirs lazily
233 - sort_ref_dir(): do not sort if already sorted
234 - refs: store references hierarchically
235 - refs.c: rename ref_array -> ref_dir
236 - struct ref_entry: nest the value part in a union
237 - check_refname_component(): return 0 for zero-length components
238 - free_ref_entry(): new function
239 - refs.c: reorder definitions more logically
240 - is_refname_available(): reimplement using do_for_each_ref_in_array()
241 - names_conflict(): simplify implementation
242 - names_conflict(): new function, extracted from is_refname_available()
243 - repack_without_ref(): reimplement using do_for_each_ref_in_array()
244 - do_for_each_ref_in_arrays(): new function
245 - do_for_each_ref_in_array(): new function
246 - do_for_each_ref(): correctly terminate while processesing extra_refs
248 The API for extra anchoring points may require rethought first; that would
249 hopefully make the "ref" part a lot simpler. And that is happening in
250 another topic (which has graduated to 'master').
252 Will defer till the next cycle.
254 * ss/git-svn-prompt-sans-terminal (2012-01-04) 3 commits
256 - git-svn, perl/Git.pm: extend Git::prompt helper for querying users
257 (merged to 'next' on 2012-01-05 at 954f125)
258 + perl/Git.pm: "prompt" helper to honor GIT_ASKPASS and SSH_ASKPASS
260 The bottom one has been replaced with a rewrite based on comments from
261 Ævar. The second one needs more work, both in perl/Git.pm and prompt.c, to
262 give precedence to tty over SSH_ASKPASS when terminal is available.
264 Will defer till the next cycle.
266 * cb/git-daemon-tests (2012-01-08) 5 commits
267 (merged to 'next' on 2012-01-08 at 1db8351)
268 + git-daemon tests: wait until daemon is ready
269 + git-daemon: produce output when ready
270 + git-daemon: add tests
271 + dashed externals: kill children on exit
272 + run-command: optionally kill children on exit
274 Will defer till the next cycle.
276 * jk/parse-object-cached (2012-01-06) 3 commits
277 (merged to 'next' on 2012-01-08 at 8c6fa4a)
278 + upload-pack: avoid parsing tag destinations
279 + upload-pack: avoid parsing objects during ref advertisement
280 + parse_object: try internal cache before reading object db
282 These are a bit scary changes, but I do think they are worth doing.
283 Will defer till the next cycle.
285 * jn/maint-gitweb-grep-fix (2012-01-05) 2 commits
286 - gitweb: Harden "grep" search against filenames with ':'
287 - gitweb: Fix file links in "grep" search
289 Waiting for a confirmation from bug reporter.