What's cooking (2012/07 #03)
Junio C Hamano committed
Jul 11, 2012 at 15:12 UTC
dbf298f6ff94c33cb0958fe27f40ba2056eb93f0
1 file changed
+181
-315
whats-cooking.txt
+181
-315
@@ -1,15 +1,17 @@
1
To: git@vger.kernel.org
2
-Subject: What's cooking in git.git (Jul 2012, #02; Mon, 9)
3
-X-master-at: 299666cc5c792dbc3b67622fcc03d41ff95f4346
4
-X-next-at: 61b12c254ff58ac7c57a6e1ab2aa50b1ada84e72
2
+Subject: What's cooking in git.git (Jul 2012, #03; Wed, 11)
3
+X-master-at: 2b533592900ecebeb69e2dd1b70744433700ff2d
4
+X-next-at: 03fbcf6d885017d6d49b32c4b9a7f578ebd6d0de
5
6
-What's cooking in git.git (Jul 2012, #02; Mon, 9)
6
+What's cooking in git.git (Jul 2012, #03; Wed, 11)
7
--------------------------------------------------
8
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 'next'.
11
12
-The fourth and fifth batches of topics have graduated to 'master'.
12
+Many topics in flight are now in 'next' and all of them look more or
13
+less done (knock wood). Many fixes that are already in 'master' are
14
+now merged down to the 'maint' branch and 1.7.11.2 has been tagged.
15
16
You can find the changes described here in the integration branches of the
17
repositories listed at
@@ -19,6 +21,115 @@ repositories listed at
21
--------------------------------------------------
22
[New Topics]
23
24
+* jc/maint-filter-branch-epoch-date (2012-07-09) 1 commit
25
+ (merged to 'next' on 2012-07-10 at f9774ca)
26
+ + filter-branch: do not forget the '@' prefix to force git-timestamp
27
+
28
+In 1.7.9 era, we taught "git rebase" about the raw timestamp format
29
+but we did not teach the same trick to "filter-branch", which rolled
30
+a similar logic on its own.
31
+
32
+* jk/maint-commit-amend-only-no-paths (2012-07-10) 1 commit
33
+ (merged to 'next' on 2012-07-10 at 1c650c2)
34
+ + commit: fix "--amend --only" with no pathspec
35
+
36
+"git commit --amend --only --" was meant to allow "Clever" people to
37
+rewrite the commit message without making any change even when they
38
+have already changes for the next commit added to their index, but
39
+it never worked as advertised since it was introduced in 1.3.0 era.
40
+
41
+* jk/index-pack-streaming-fix (2012-07-10) 1 commit
42
+ - index-pack: loop while inflating objects in unpack_data
43
+
44
+This needs to go to 'master' soonish.
45
+
46
+* mb/remote-default-nn-origin (2012-07-11) 6 commits
47
+ - Teach get_default_remote to respect remote.default.
48
+ - Test that plain "git fetch" uses remote.default when on a detached HEAD.
49
+ - Teach clone to set remote.default.
50
+ - Teach "git remote" about remote.default.
51
+ - Teach remote.c about the remote.default configuration setting.
52
+ - Rename remote.c's default_remote_name static variables.
53
+
54
+* tg/ce-namelen-field (2012-07-11) 2 commits
55
+ - Strip namelen out of ce_flags into a ce_namelen field
56
+ - Merge branch 'tg/maint-cache-name-compare' into tg/ce-namelen-field
57
+ (this branch uses tg/maint-cache-name-compare.)
58
+
59
+* tg/maint-cache-name-compare (2012-07-11) 1 commit
60
+ - cache_name_compare(): do not truncate while comparing paths
61
+ (this branch is used by tg/ce-namelen-field.)
62
+
63
+--------------------------------------------------
64
+[Stalled]
65
+
66
+* jc/maint-push-refs-all (2012-05-04) 2 commits
67
+ - get_fetch_map(): tighten checks on dest refs
68
+ - fetch/push: allow refs/*:refs/*
69
+
70
+Allows pushing and fetching refs/stash.
71
+There still seem to be other bugs hiding (e.g. try pushing twice).
72
+
73
+Not ready.
74
+
75
+* jc/split-blob (2012-04-03) 6 commits
76
+ - chunked-object: streaming checkout
77
+ - chunked-object: fallback checkout codepaths
78
+ - bulk-checkin: support chunked-object encoding
79
+ - bulk-checkin: allow the same data to be multiply hashed
80
+ - new representation types in the packstream
81
+ - packfile: use varint functions
82
+
83
+Not ready.
84
+
85
+I finished the streaming checkout codepath, but as explained in
86
+127b177 (bulk-checkin: support chunked-object encoding, 2011-11-30),
87
+these are still early steps of a long and painful journey. At least
88
+pack-objects and fsck need to learn the new encoding for the series
89
+to be usable locally, and then index-pack/unpack-objects needs to
90
+learn it to be used remotely.
91
+
92
+Given that I heard a lot of noise that people want large files, and
93
+that I was asked by somebody at GitTogether'11 privately for an
94
+advice on how to pay developers (not me) to help adding necessary
95
+support, I am somewhat dissapointed that the original patch series
96
+that was sent almost two months ago still remains here without much
97
+comments and updates from the developer community. I even made the
98
+interface to the logic that decides where to split chunks easily
99
+replaceable, and I deliberately made the logic in the original patch
100
+extremely stupid to entice others, especially the "bup" fanboys, to
101
+come up with a better logic, thinking that giving people an easy
102
+target to shoot for, they may be encouraged to help out. The plan is
103
+not working :-(.
104
+
105
+--------------------------------------------------
106
+[Cooking]
107
+
108
+* jc/apply-3way (2012-07-09) 19 commits
109
+ (merged to 'next' on 2012-07-10 at fb5c872)
110
+ + apply: tests for the --3way option
111
+ + apply: document --3way option
112
+ + apply: allow rerere() to work on --3way results
113
+ + apply: register conflicted stages to the index
114
+ + apply: --3way with add/add conflict
115
+ + apply: move verify_index_match() higher
116
+ + apply: plug the three-way merge logic in
117
+ + apply: fall back on three-way merge
118
+ + apply: accept -3/--3way command line option
119
+ + apply: move "already exists" logic to check_to_create()
120
+ + apply: move check_to_create_blob() closer to its sole caller
121
+ + apply: further split load_preimage()
122
+ + apply: refactor "previous patch" logic
123
+ + apply: split load_preimage() helper function out
124
+ + apply: factor out checkout_target() helper function
125
+ + apply: refactor read_file_or_gitlink()
126
+ + apply: clear_image() clears things a bit more
127
+ + apply: a bit more comments on PATH_TO_BE_DELETED
128
+ + apply: fix an incomplete comment in check_patch()
129
+
130
+"git apply" learns to wiggle the base version and perform three-way merge
131
+when a patch does not exactly apply to the version you have.
132
+
133
* jl/submodule-rm (2012-07-05) 2 commits
134
- rm: remove submodules from the index and the .gitmodules file
135
- rm: don't fail when removing populated submodules
@@ -27,7 +138,7 @@ repositories listed at
138
- Document rev^! and rev^@ as revision specifiers
139
- Make <refname> documentation more consistent.
140
30
-The tip one is a bit suspicious.
141
+The tip one will be rerolled.
142
143
* nk/maint-gitweb-log-by-lines (2012-07-05) 3 commits
144
- gitweb: Add support to Link: tag
@@ -37,11 +148,13 @@ The tip one is a bit suspicious.
148
Is the use of esc_html() in the tip one correct?
149
150
* pw/git-p4-jobs (2012-07-05) 3 commits
40
- - git p4: notice Jobs lines in git commit messages
41
- - git p4 test: refactor marshal_dump
42
- - git p4: remove unused P4Submit interactive setting
151
+ (merged to 'next' on 2012-07-10 at c6ef8f9)
152
+ + git p4: notice Jobs lines in git commit messages
153
+ + git p4 test: refactor marshal_dump
154
+ + git p4: remove unused P4Submit interactive setting
155
44
-Will merge to 'next'.
156
+Teach "git p4" to notice "Jobs:" in the log message and relay it to
157
+Perforce to trigger its "jobs" support.
158
159
* pw/git-p4-move (2012-07-05) 2 commits
160
- git p4: add support for 'p4 move' in P4Submit
@@ -49,12 +162,15 @@ Will merge to 'next'.
162
163
Will be rerolled.
164
52
-* ar/clone-honor-umask-at-top (2012-07-08) 1 commit
53
- - Restore umasks influence on the permissions of work tree created by clone
54
-
55
-May want to rephrase the log message a bit.
165
+* ar/clone-honor-umask-at-top (2012-07-09) 3 commits
166
+ (merged to 'next' on 2012-07-10 at 680bc22)
167
+ + add: create ADD_EDIT.patch with mode 0666
168
+ + rerere: make rr-cache fanout directory honor umask
169
+ + Restore umasks influence on the permissions of work tree created by clone
170
57
-Will merge to 'next'.
171
+A handful of files and directories we create had tighter than
172
+necessary permission bits when the user wanted to have group
173
+writability (e.g. by setting "umask 002").
174
175
* jn/vcs-svn (2012-07-05) 12 commits
176
(merged to 'next' on 2012-07-09 at 1d97a8f)
@@ -100,241 +216,39 @@ Will merge to 'master'.
216
Will be rerolled but is going in the right direction.
217
218
* tg/ce-namelen (2012-07-08) 1 commit
103
- - Replace strlen() with ce_namelen()
219
+ (merged to 'next' on 2012-07-10 at 87755d8)
220
+ + Replace strlen() with ce_namelen()
221
222
Trivially correct clean-up and micro optimization.
223
224
+Will merge to 'master'.
225
+
226
* jn/makefile-cleanup (2012-07-09) 14 commits
108
- - Makefile: document ground rules for target-specific dependencies
109
- - Makefile: move GIT-VERSION-FILE dependencies closer to use
110
- - Makefile: build instaweb similar to other scripts
111
- - Makefile: update scripts when build-time parameters change
112
- - Makefile: do not replace @@GIT_VERSION@@ in shell scripts
113
- - Makefile: split prefix flags from GIT-CFLAGS
114
- - Makefile: be silent when only GIT_USER_AGENT changes
115
- - Makefile: split GIT_USER_AGENT from GIT-CFLAGS
116
- - Makefile: do not replace @@GIT_USER_AGENT@@ in scripts
117
- - Makefile: apply dependencies consistently to sparse/asm targets
118
- - Makefile: do not have git.o depend on common-cmds.h
119
- - Makefile: fold XDIFF_H and VCSSVN_H into LIB_H
120
- - Makefile: fold MISC_H into LIB_H
121
- - Makefile: sort LIB_H list
122
- (this branch is tangled with jk/version-string-dependency.)
123
-
124
-Tweaks and replaces jk/version-string-dependency.
125
-
126
-Will merge to 'next'.
227
+ (merged to 'next' on 2012-07-10 at f55e6a1)
228
+ + Makefile: document ground rules for target-specific dependencies
229
+ + Makefile: move GIT-VERSION-FILE dependencies closer to use
230
+ + Makefile: build instaweb similar to other scripts
231
+ + Makefile: update scripts when build-time parameters change
232
+ + Makefile: do not replace @@GIT_VERSION@@ in shell scripts
233
+ + Makefile: split prefix flags from GIT-CFLAGS
234
+ + Makefile: be silent when only GIT_USER_AGENT changes
235
+ + Makefile: split GIT_USER_AGENT from GIT-CFLAGS
236
+ + Makefile: do not replace @@GIT_USER_AGENT@@ in scripts
237
+ + Makefile: apply dependencies consistently to sparse/asm targets
238
+ + Makefile: do not have git.o depend on common-cmds.h
239
+ + Makefile: fold XDIFF_H and VCSSVN_H into LIB_H
240
+ + Makefile: fold MISC_H into LIB_H
241
+ + Makefile: sort LIB_H list
242
+
243
+Tightens dependency rules to avoid unnecessary recompilation.
244
245
* cw/amend-commit-without-message (2012-07-09) 1 commit
129
- - Allow edit of empty message with commit --amend
246
+ (merged to 'next' on 2012-07-10 at e7c75d5)
247
+ + Allow edit of empty message with commit --amend
248
249
"commit --amend" used to refuse amending a commit with an empty log
250
message, with or without "--allow-empty-message".
251
134
-Will merge to 'next'.
135
-
136
---------------------------------------------------
137
-[Graduated to "master"]
138
-
139
-* cw/help-over-network (2012-06-28) 2 commits
140
- (merged to 'next' on 2012-07-03 at 99e8d28)
141
- + Allow help.htmlpath to be a URL prefix
142
- + Add config variable to set HTML path for git-help --web
143
-
144
-"git help -w $cmd" can show HTML version of documentation for
145
-"git-$cmd" by setting help.htmlpath to somewhere other than the
146
-default location where the build procedure installs them locally;
147
-the variable can even point at a http:// URL.
148
-
149
-* cw/no-detaching-an-unborn (2012-06-26) 1 commit
150
- (merged to 'next' on 2012-07-03 at 552c5ca)
151
- + git-checkout: disallow --detach on unborn branch
152
-
153
-"git checkout --detach", when you are still on an unborn branch,
154
-should be forbidden, but it wasn't.
155
-
156
-* hv/remote-end-hung-up (2012-06-19) 1 commit
157
- (merged to 'next' on 2012-06-25 at d3f29e2)
158
- + remove the impression of unexpectedness when access is denied
159
-
160
-When we get disconnected while expecting a response from the remote
161
-side because authentication failed, we issued an error message "The
162
-remote side hung up unexpectedly." Give hint that it may be a
163
-permission problem in the message when we can reasonably suspect it.
164
-
165
-* jc/maint-blame-unique-abbrev (2012-07-02) 1 commit
166
- (merged to 'next' on 2012-07-03 at c7ed6b9)
167
- + blame: compute abbreviation width that ensures uniqueness
168
-
169
-"git blame" did not try to make sure the abbreviated commit object
170
-names in its output are unique.
171
-
172
-* jn/perl-makemaker-leading-paths (2012-06-29) 2 commits
173
- (merged to 'next' on 2012-06-29 at 0f17059)
174
- + perl/Makefile: Fix a missing double-quote
175
- (merged to 'next' on 2012-06-28 at 286d854)
176
- + perl/Makefile: move "mkdir -p" to module installation loop for maintainability
177
-
178
-Minor build tweak. People who do not use ExtUtils::MakeMaker may
179
-want to test this before it hits the 'master' branch.
180
-
181
-* js/fast-export-paths-with-spaces (2012-06-27) 1 commit
182
- (merged to 'next' on 2012-07-03 at f9c23e9)
183
- + fast-export: quote paths with spaces
184
-
185
-"git fast-export" produced an input stream for fast-import without
186
-properly quoting pathnames when they contain SPs in them.
187
-
188
-* mm/config-xdg (2012-06-25) 4 commits
189
- (merged to 'next' on 2012-06-28 at 57049d1)
190
- + config: write to $XDG_CONFIG_HOME/git/config file when appropriate
191
- + Let core.attributesfile default to $XDG_CONFIG_HOME/git/attributes
192
- + Let core.excludesfile default to $XDG_CONFIG_HOME/git/ignore
193
- + config: read (but not write) from $XDG_CONFIG_HOME/git/config file
194
-
195
-Teach git to read various information from $XDG_CONFIG_HOME/git/ to allow
196
-the user to avoid cluttering $HOME.
197
-
198
-* mm/credential-plumbing (2012-07-08) 4 commits
199
- (merged to 'next' on 2012-07-08 at fce8ea2)
200
- + git-remote-mediawiki: update comments to reflect credential support
201
- (merged to 'next' on 2012-06-28 at 6e6fe9f)
202
- + git-remote-mediawiki: add credential support
203
- + git credential fill: output the whole 'struct credential'
204
- + add 'git credential' plumbing command
205
- (this branch is tangled with mm/mediawiki-file-attachments and mm/mediawiki-tests.)
206
-
207
-Expose the credential API to scripted Porcelain writers.
208
-
209
-* mz/empty-rebase-test (2012-06-27) 1 commit
210
- (merged to 'next' on 2012-07-03 at 1e7de3f)
211
- + add test case for rebase of empty commit
212
-
213
-We did not have test to make sure "git rebase" without extra options
214
-filters out an empty commit in the original history.
215
-
216
-* nd/clone-single-fix (2012-06-22) 1 commit
217
- (merged to 'next' on 2012-06-26 at c9fff5e)
218
- + clone: fix ref selection in --single-branch --branch=xxx
219
-
220
-"git clone --single-branch" to clone a single branch did not limit
221
-the cloning to the specified branch.
222
-
223
-* pw/git-p4-tests (2012-06-27) 10 commits
224
- (merged to 'next' on 2012-07-03 at 9bd9cda)
225
- + git p4 test: fix badp4dir test
226
- + git p4 test: split up big t9800 test
227
- + git p4 test: cleanup_git should make a new $git
228
- + git p4 test: copy source indeterminate
229
- + git p4 test: check for error message in failed test
230
- + git p4 test: rename some "git-p4 command" strings
231
- + git p4 test: never create default test repo
232
- + git p4 test: simplify quoting involving TRASH_DIRECTORY
233
- + git p4 test: use real_path to resolve p4 client symlinks
234
- + git p4 test: wait longer for p4d to start and test its pid
235
- (this branch is used by pw/git-p4-jobs and pw/git-p4-move.)
236
-
237
-More "git p4" tests.
238
-
239
-* rj/platform-pread-may-be-thread-unsafe (2012-06-26) 1 commit
240
- (merged to 'next' on 2012-06-28 at ce5f79f)
241
- + index-pack: Disable threading on cygwin
242
-
243
-On Cygwin, the platform pread(3) is not thread safe, just like our
244
-own compat/ emulation, and cannot be used in the index-pack program.
245
-
246
-* th/diff-no-index-fixes (2012-06-22) 2 commits
247
- (merged to 'next' on 2012-06-26 at 450e781)
248
- + diff-no-index: exit(1) if 'diff --quiet <repo file> <external file>' finds changes
249
- + diff: handle relative paths in no-index
250
-
251
-"git diff --no-index" did not correctly handle relative paths and
252
-did not correctly give exit codes when run under "--quiet" option.
253
-
254
-* vr/use-our-perl-in-tests (2012-06-24) 4 commits
255
- (merged to 'next' on 2012-07-03 at 90dfcf0)
256
- + t/README: add a bit more Don'ts
257
- + tests: enclose $PERL_PATH in double quotes
258
- + t/test-lib.sh: export PERL_PATH for use in scripts
259
- + t: Replace 'perl' by $PERL_PATH
260
- (this branch is used by jc/test-lib-source-build-options-early.)
261
-
262
-Some implementations of Perl terminates "lines" with CRLF even when
263
-the script is operating on just a sequence of bytes. Make sure to
264
-use "$PERL_PATH", the version of Perl the user told Git to use, in
265
-our tests to avoid unnecessary breakages in tests.
266
-
267
---------------------------------------------------
268
-[Stalled]
269
-
270
-* jc/apply-3way (2012-06-13) 19 commits
271
- - apply --3way: tests
272
- - apply: document --3way option
273
- - apply: allow rerere() upon --3way results
274
- - apply: register conflicted stages to the index
275
- - apply: --3way with add/add conflict
276
- - apply: move verify_index_match() higher
277
- - apply: plug the three-way merge logic in
278
- - apply: fall back on three-way merge
279
- - apply: accept -3/--3way command line option
280
- - apply: move "already exists" logic to check_to_create()
281
- - apply: move check_to_create_blob() closer to its sole caller
282
- - apply: further split load_preimage()
283
- - apply: refactor "previous patch" logic
284
- - apply: split load_preimage() helper function out
285
- - apply: factor out checkout_target() helper function
286
- - apply: refactor read_file_or_gitlink()
287
- - apply: clear_image() clears things a bit more
288
- - apply: a bit more comments on PATH_TO_BE_DELETED
289
- - apply: fix an incomplete comment in check_patch()
290
-
291
-"git apply" learns to wiggle the base version and perform three-way merge
292
-when a patch does not exactly apply to the version you have.
293
-
294
-Will merge to 'next' after reading it over once again.
295
-
296
-* jc/maint-push-refs-all (2012-05-04) 2 commits
297
- - get_fetch_map(): tighten checks on dest refs
298
- - fetch/push: allow refs/*:refs/*
299
-
300
-Allows pushing and fetching refs/stash.
301
-There still seem to be other bugs hiding (e.g. try pushing twice).
302
-
303
-Not ready.
304
-
305
-* jc/split-blob (2012-04-03) 6 commits
306
- - chunked-object: streaming checkout
307
- - chunked-object: fallback checkout codepaths
308
- - bulk-checkin: support chunked-object encoding
309
- - bulk-checkin: allow the same data to be multiply hashed
310
- - new representation types in the packstream
311
- - packfile: use varint functions
312
-
313
-Not ready.
314
-
315
-I finished the streaming checkout codepath, but as explained in
316
-127b177 (bulk-checkin: support chunked-object encoding, 2011-11-30),
317
-these are still early steps of a long and painful journey. At least
318
-pack-objects and fsck need to learn the new encoding for the series
319
-to be usable locally, and then index-pack/unpack-objects needs to
320
-learn it to be used remotely.
321
-
322
-Given that I heard a lot of noise that people want large files, and
323
-that I was asked by somebody at GitTogether'11 privately for an
324
-advice on how to pay developers (not me) to help adding necessary
325
-support, I am somewhat dissapointed that the original patch series
326
-that was sent almost two months ago still remains here without much
327
-comments and updates from the developer community. I even made the
328
-interface to the logic that decides where to split chunks easily
329
-replaceable, and I deliberately made the logic in the original patch
330
-extremely stupid to entice others, especially the "bup" fanboys, to
331
-come up with a better logic, thinking that giving people an easy
332
-target to shoot for, they may be encouraged to help out. The plan is
333
-not working :-(.
334
-
335
---------------------------------------------------
336
-[Cooking]
337
-
252
* jk/push-delete-ref-error-message (2012-07-03) 1 commit
253
(merged to 'next' on 2012-07-09 at 82c6bd2)
254
+ push: don't guess at qualifying remote refs on deletion
@@ -436,40 +350,39 @@ one of its early operations.
350
Needs to be eyeballed by people who run tests with exotic options
351
like valgrind, --root=/dev/shm/somewhere, etc.
352
439
-* jc/sha1-name-more (2012-07-04) 25 commits
440
- - rev-parse --disambiguate=<prefix>
441
- - rev-parse: A and B in "rev-parse A..B" refer to committish
442
- - reset: the command takes committish
443
- - commit-tree: the command wants a tree and commits
444
- - apply: --build-fake-ancestor expects blobs
445
- - sha1_name.c: add support for disambiguating other types
446
- - revision.c: the "log" family, except for "show", takes committish
447
- - revision.c: allow handle_revision_arg() to take other flags
448
- - sha1_name.c: introduce get_sha1_committish()
449
- - sha1_name.c: teach lookup context to get_sha1_with_context()
450
- - sha1_name.c: many short names can only be committish
451
- - sha1_name.c: get_sha1_1() takes lookup flags
452
- - sha1_name.c: get_describe_name() by definition groks only commits
453
- - sha1_name.c: teach get_short_sha1() a commit-only option
454
- - sha1_name.c: allow get_short_sha1() to take other flags
455
- - get_sha1(): fix error status regression
456
- - sha1_name.c: restructure disambiguation of short names
457
- - sha1_name.c: correct misnamed "canonical" and "res"
458
- - sha1_name.c: refactor find_short_packed_object()
459
- - sha1_name.c: rename "now" to "current"
460
- - sha1_name.c: clarify what "fake" is for in find_short_object_filename()
461
- - sha1_name.c: get rid of get_sha1_with_mode()
462
- - sha1_name.c: get rid of get_sha1_with_mode_1()
463
- - sha1_name.c: hide get_sha1_with_context_1() ugliness
464
- - sha1_name.c: indentation fix
353
+* jc/sha1-name-more (2012-07-09) 25 commits
354
+ (merged to 'next' on 2012-07-10 at 669dd92)
355
+ + rev-parse --disambiguate=<prefix>
356
+ + rev-parse: A and B in "rev-parse A..B" refer to committish
357
+ + reset: the command takes committish
358
+ + commit-tree: the command wants a tree and commits
359
+ + apply: --build-fake-ancestor expects blobs
360
+ + sha1_name.c: add support for disambiguating other types
361
+ + revision.c: the "log" family, except for "show", takes committish
362
+ + revision.c: allow handle_revision_arg() to take other flags
363
+ + sha1_name.c: introduce get_sha1_committish()
364
+ + sha1_name.c: teach lookup context to get_sha1_with_context()
365
+ + sha1_name.c: many short names can only be committish
366
+ + sha1_name.c: get_sha1_1() takes lookup flags
367
+ + sha1_name.c: get_describe_name() by definition groks only commits
368
+ + sha1_name.c: teach get_short_sha1() a commit-only option
369
+ + sha1_name.c: allow get_short_sha1() to take other flags
370
+ + get_sha1(): fix error status regression
371
+ + sha1_name.c: restructure disambiguation of short names
372
+ + sha1_name.c: correct misnamed "canonical" and "res"
373
+ + sha1_name.c: refactor find_short_packed_object()
374
+ + sha1_name.c: rename "now" to "current"
375
+ + sha1_name.c: clarify what "fake" is for in find_short_object_filename()
376
+ + sha1_name.c: get rid of get_sha1_with_mode()
377
+ + sha1_name.c: get rid of get_sha1_with_mode_1()
378
+ + sha1_name.c: hide get_sha1_with_context_1() ugliness
379
+ + sha1_name.c: indentation fix
380
381
Teaches the object name parser things like a "git describe" output
382
is always a commit object, "A" in "git log A" must be a committish,
383
and "A" and "B" in "git log A...B" both must be committish, etc., to
384
prolong the lifetime of abbreviated object names.
385
471
-Will merge to 'next'.
472
-
386
* tr/maint-show-walk (2012-06-19) 2 commits
387
(merged to 'next' on 2012-07-09 at c8e0e06)
388
+ show: fix "range implies walking"
@@ -492,50 +405,3 @@ Will defer for 6 months until ancient "less" goes extinct.
405
+ push: start warning upcoming default change for push.default
406
407
Will defer for another cycle.
495
-
496
---------------------------------------------------
497
-[Discarded]
498
-
499
-* nl/http-proxy-more (2012-05-11) 2 commits
500
- . http: rename HTTP_REAUTH to HTTP_AUTH_RETRY
501
- . http: Avoid limit of retrying request only twice
502
-
503
-I queued only the later two patches from this series, even though they do
504
-not make much sense without the first one that seems to need a bit more
505
-work, so that we won't forget.
506
-
507
-* jk/no-op-push-message (2012-05-30) 1 commit
508
- . improve no-op push output
509
-
510
-Rewords the status message of "git push" that pushed only one ref
511
-differently from "Everything up-to-date", to give a bit more help to
512
-people who get the message when their current branch is not pushed.
513
-
514
-I had an impression after the discussion thread that a redesign is
515
-coming, but it hasn't happened yet.
516
-
517
-* db/vcs-svn (2012-06-01) 6 commits
518
- . vcs-svn: drop no-op reset methods
519
- . vcs-svn: fix signedness warnings
520
- . vcs-svn: prefer strstr over memmem
521
- . vcs-svn: prefer constcmp to prefixcmp
522
- . vcs-svn: simplify cleanup in apply_one_window()
523
- . vcs-svn: fix clang-analyzer error
524
-
525
-Replaced by jn/vcs-svn.
526
-
527
-* jk/version-string-dependency (2012-06-20) 11 commits
528
- . Makefile: move GIT-VERSION-FILE dependencies closer to use
529
- . Makefile: build instaweb similar to other scripts
530
- . Makefile: update scripts when build-time parameters change
531
- . Makefile: do not replace @@GIT_VERSION@@ in shell scripts
532
- . Makefile: split prefix flags from GIT-CFLAGS
533
- . Makefile: split GIT_USER_AGENT from GIT-CFLAGS
534
- . Makefile: do not replace @@GIT_USER_AGENT@@ in scripts
535
- . Makefile: apply dependencies consistently to sparse/asm targets
536
- . Makefile: do not have git.o depend on common-cmds.h
537
- - Makefile: fold MISC_H into LIB_H
538
- - Makefile: sort LIB_H list
539
- (this branch is tangled with jn/makefile-cleanup.)
540
-
541
-Replaced by jn/makefile-cleanup.