What's cooking (2013/05 #09)
Junio C Hamano committed
May 29, 2013 at 16:56 UTC
03ba10812e3960ab9120363903dee94e99ace5a0
1 file changed
+368
-353
whats-cooking.txt
+368
-353
@@ -1,36 +1,371 @@
1
To: git@vger.kernel.org
2
Bcc: lwn@lwn.net
3
-Subject: What's cooking in git.git (May 2013, #08; Tue, 28)
4
-X-master-at: edca4152560522a431a51fc0a06147fc680b5b18
5
-X-next-at: c6abf3f6e02deba43158d8c73669e079c83d9722
3
+Subject: What's cooking in git.git (May 2013, #09; Wed, 29)
4
+X-master-at: 727a46b2f9a1ce69eaf09bc46cb129f1c40833d8
5
+X-next-at: f3f2a46de0016cbb91a672fc1179759b6ad9a32c
6
7
-What's cooking in git.git (May 2013, #08; Tue, 28)
7
+What's cooking in git.git (May 2013, #09; Wed, 29)
8
--------------------------------------------------
9
10
Here are the topics that have been cooking. Commits prefixed with
11
'-' are only in 'pu' (proposed updates) while commits prefixed with
12
'+' are in 'next'.
13
14
-There are topics that are still in 'next', not because they needed
15
-more testing, but only because we already were in the -rc period.
16
-Now that 1.8.3 is out, we will start merging them to 'master' (and
17
-some to 'maint' for 1.8.3.1) probably tomorrow (after waiting for a
18
-few days to make sure we can have a clean start of the post 1.8.3
19
-maint branch with brown paper bag fixes and nothing else).
14
+Usually I avoid issuing "What's cooking" back-to-back, but because I
15
+will be offline for a few days, here is a snapshot of tonight's
16
+status.
17
21
-The post 1.8.3 cycle will start early next month, at which time the
22
-'next' branch will be rewound and rebuilt. Until then, I expect
23
-that my patch queue may stay slow and leaky while I take a bit of
24
-break.
18
+Some topics that have been cooking in 'next' from the previous cycle
19
+have now graduated to 'master', so the RelNotes have been updated.
20
+
21
+Nothing new in 'next', and it has not been rewound yet, which will
22
+hopefully happen this weekend, and after that post 1.8.3 cycle
23
+really starts.
24
+
25
+Also a rather serious regression on path-exclusion logic (most
26
+notably seen in .gitignore) has been found and quickly patched (it
27
+hasn't been merged to 'master' yet, though).
28
29
You can find the changes described here in the integration branches
30
of the repositories listed at
31
32
http://git-blame.blogspot.com/p/git-public-repositories.html
33
34
+--------------------------------------------------
35
+[Graduated to "master"]
36
+
37
+* as/check-ignore (2013-04-29) 6 commits
38
+ (merged to 'next' on 2013-04-30 at 646931f)
39
+ + t0008: use named pipe (FIFO) to test check-ignore streaming
40
+ (merged to 'next' on 2013-04-21 at 7515aa8)
41
+ + Documentation: add caveats about I/O buffering for check-{attr,ignore}
42
+ + check-ignore: allow incremental streaming of queries via --stdin
43
+ + check-ignore: move setup into cmd_check_ignore()
44
+ + check-ignore: add -n / --non-matching option
45
+ + t0008: remove duplicated test fixture data
46
+
47
+ Enhance "check-ignore" (1.8.2 update) to work more like "check-attr"
48
+ over bidi-pipes.
49
+
50
+
51
+* fc/transport-helper-error-reporting (2013-05-10) 12 commits
52
+ (merged to 'next' on 2013-05-10 at 2a9af4b)
53
+ + transport-helper: fix remote helper namespace regression
54
+ + test: remote-helper: add missing and
55
+ (merged to 'next' on 2013-04-25 at 3358f1a)
56
+ + t5801: "VAR=VAL shell_func args" is forbidden
57
+ (merged to 'next' on 2013-04-22 at 5ba6467)
58
+ + transport-helper: update remote helper namespace
59
+ + transport-helper: trivial code shuffle
60
+ + transport-helper: warn when refspec is not used
61
+ + transport-helper: clarify pushing without refspecs
62
+ + transport-helper: update refspec documentation
63
+ + transport-helper: clarify *:* refspec
64
+ + transport-helper: improve push messages
65
+ + transport-helper: mention helper name when it dies
66
+ + transport-helper: report errors properly
67
+ (this branch is tangled with js/transport-helper-error-reporting-fix.)
68
+
69
+ Update transport helper to report errors and maintain ref hierarchy
70
+ used to keep track of remote helper state better.
71
+
72
+
73
+* jc/prune-all (2013-04-25) 4 commits
74
+ (merged to 'next' on 2013-04-26 at 97a7387)
75
+ + prune: introduce OPT_EXPIRY_DATE() and use it
76
+ (merged to 'next' on 2013-04-22 at b00ccf6)
77
+ + api-parse-options.txt: document "no-" for non-boolean options
78
+ + git-gc.txt, git-reflog.txt: document new expiry options
79
+ + date.c: add parse_expiry_date()
80
+ (this branch is used by mh/packed-refs-various.)
81
+
82
+ We used the approxidate() parser for "--expire=<timestamp>" options
83
+ of various commands, but it is better to treat --expire=all and
84
+ --expire=now a bit more specially than using the current timestamp.
85
+ Update "git gc" and "git reflog" with a new parsing function for
86
+ expiry dates.
87
+
88
+
89
+* jh/checkout-auto-tracking (2013-04-21) 8 commits
90
+ (merged to 'next' on 2013-04-22 at 2356700)
91
+ + glossary: Update and rephrase the definition of a remote-tracking branch
92
+ + branch.c: Validate tracking branches with refspecs instead of refs/remotes/*
93
+ + t9114.2: Don't use --track option against "svn-remote"-tracking branches
94
+ + t7201.24: Add refspec to keep --track working
95
+ + t3200.39: tracking setup should fail if there is no matching refspec.
96
+ + checkout: Use remote refspecs when DWIMming tracking branches
97
+ + t2024: Show failure to use refspec when DWIMming remote branch names
98
+ + t2024: Add tests verifying current DWIM behavior of 'git checkout <branch>'
99
+
100
+ Updates "git checkout foo" that DWIMs the intended "upstream" and
101
+ turns it into "git checkout -t -b foo remotes/origin/foo" to
102
+ correctly take existing remote definitions into account. The
103
+ remote "origin" may be what uniquely map its own branch to
104
+ remotes/some/where/foo but that some/where may not be "origin".
105
+
106
+
107
+* jk/lookup-object-prefer-latest (2013-05-02) 1 commit
108
+ (merged to 'next' on 2013-05-06 at cc59dcc)
109
+ + lookup_object: prioritize recently found objects
110
+
111
+ Optimizes object lookup when the object hashtable starts to become
112
+ crowded.
113
+
114
+
115
+* jk/subtree-do-not-push-if-split-fails (2013-05-01) 1 commit
116
+ (merged to 'next' on 2013-05-06 at 81bdf37)
117
+ + contrib/subtree: don't delete remote branches if split fails
118
+
119
+ "git subtree" (in contrib/) had one codepath with loose error
120
+ checks to lose data at the remote side.
121
+
122
+
123
+* jk/test-output (2013-05-06) 3 commits
124
+ (merged to 'next' on 2013-05-06 at 7c03af3)
125
+ + t/Makefile: don't define TEST_RESULTS_DIRECTORY recursively
126
+ (merged to 'next' on 2013-05-01 at 63827c9)
127
+ + test output: respect $TEST_OUTPUT_DIRECTORY
128
+ + t/Makefile: fix result handling with TEST_OUTPUT_DIRECTORY
129
+
130
+ When TEST_OUTPUT_DIRECTORY setting is used, it was handled somewhat
131
+ inconsistently between the test framework and t/Makefile, and logic
132
+ to summarize the results looked at a wrong place.
133
+
134
+
135
+* jn/config-ignore-inaccessible (2013-04-15) 1 commit
136
+ (merged to 'next' on 2013-05-07 at 4f09e24)
137
+ + config: allow inaccessible configuration under $HOME
138
+
139
+ When $HOME is misconfigured to point at an unreadable directory, we
140
+ used to complain and die. This loosens the check.
141
+
142
+
143
+* js/transport-helper-error-reporting-fix (2013-04-28) 13 commits
144
+ (merged to 'next' on 2013-04-29 at 8cc4bb8)
145
+ + git-remote-testgit: build it to run under $SHELL_PATH
146
+ + git-remote-testgit: further remove some bashisms
147
+ + git-remote-testgit: avoid process substitution
148
+ (merged to 'next' on 2013-04-25 at 3358f1a)
149
+ + t5801: "VAR=VAL shell_func args" is forbidden
150
+ (merged to 'next' on 2013-04-22 at 5ba6467)
151
+ + transport-helper: update remote helper namespace
152
+ + transport-helper: trivial code shuffle
153
+ + transport-helper: warn when refspec is not used
154
+ + transport-helper: clarify pushing without refspecs
155
+ + transport-helper: update refspec documentation
156
+ + transport-helper: clarify *:* refspec
157
+ + transport-helper: improve push messages
158
+ + transport-helper: mention helper name when it dies
159
+ + transport-helper: report errors properly
160
+ (this branch is tangled with fc/transport-helper-error-reporting.)
161
+
162
+ Finishing touches to fc/transport-helper-error-reporting topic.
163
+
164
+
165
+* mh/fetch-into-shallow (2013-05-02) 2 commits
166
+ (merged to 'next' on 2013-05-03 at 3fadc61)
167
+ + t5500: add test for fetching with an unknown 'shallow'
168
+ (merged to 'next' on 2013-04-29 at a167d3e)
169
+ + upload-pack: ignore 'shallow' lines with unknown obj-ids
170
+
171
+
172
+* mh/packed-refs-various (2013-05-01) 33 commits
173
+ (merged to 'next' on 2013-05-01 at e527153)
174
+ + refs: handle the main ref_cache specially
175
+ + refs: change do_for_each_*() functions to take ref_cache arguments
176
+ + pack_one_ref(): do some cheap tests before a more expensive one
177
+ + pack_one_ref(): use write_packed_entry() to do the writing
178
+ + pack_one_ref(): use function peel_entry()
179
+ + refs: inline function do_not_prune()
180
+ + pack_refs(): change to use do_for_each_entry()
181
+ + refs: use same lock_file object for both ref-packing functions
182
+ + pack_one_ref(): rename "path" parameter to "refname"
183
+ + pack-refs: merge code from pack-refs.{c,h} into refs.{c,h}
184
+ + pack-refs: rename handle_one_ref() to pack_one_ref()
185
+ + refs: extract a function write_packed_entry()
186
+ + repack_without_ref(): write peeled refs in the rewritten file
187
+ + t3211: demonstrate loss of peeled refs if a packed ref is deleted
188
+ + refs: change how packed refs are deleted
189
+ + search_ref_dir(): return an index rather than a pointer
190
+ + repack_without_ref(): silence errors for dangling packed refs
191
+ + t3210: test for spurious error messages for dangling packed refs
192
+ + refs: change the internal reference-iteration API
193
+ + refs: extract a function peel_entry()
194
+ + peel_ref(): fix return value for non-peelable, not-current reference
195
+ + peel_object(): give more specific information in return value
196
+ + refs: extract function peel_object()
197
+ + refs: extract a function ref_resolves_to_object()
198
+ + repack_without_ref(): use function get_packed_ref()
199
+ + peel_ref(): use function get_packed_ref()
200
+ + get_packed_ref(): return a ref_entry
201
+ + do_for_each_ref_in_dirs(): remove dead code
202
+ + refs: define constant PEELED_LINE_LENGTH
203
+ + refs: document how current_ref is used
204
+ + refs: document do_for_each_ref() and do_one_ref()
205
+ + refs: document the fields of struct ref_value
206
+ + refs: document flags constants REF_*
207
+ (this branch uses jc/prune-all.)
208
+
209
+ Updates reading and updating packed-refs file, correcting corner
210
+ case bugs.
211
+
212
+
213
+* rj/sparse (2013-04-28) 10 commits
214
+ (merged to 'next' on 2013-05-01 at 649e16c)
215
+ + sparse: Fix mingw_main() argument number/type errors
216
+ + compat/mingw.c: Fix some sparse warnings
217
+ + compat/win32mmap.c: Fix some sparse warnings
218
+ + compat/poll/poll.c: Fix a sparse warning
219
+ + compat/win32/pthread.c: Fix a sparse warning
220
+ + compat/unsetenv.c: Fix a sparse warning
221
+ + compat/nedmalloc: Fix compiler warnings on linux
222
+ + compat/nedmalloc: Fix some sparse warnings
223
+ + compat/fnmatch/fnmatch.c: Fix a sparse error
224
+ + compat/regex/regexec.c: Fix some sparse warnings
225
+
226
--------------------------------------------------
227
[New Topics]
228
229
+* ap/diff-ignore-blank-lines (2013-05-29) 1 commit
230
+ - diff: add --ignore-blank-lines option
231
+
232
+ "git diff" learned a mode that ignores hunks whose change consists
233
+ only of additions and removals of blank lines, which is the same as
234
+ "diff -B" (ignore blank lines) of GNU diff.
235
+
236
+
237
+* fc/show-branch-in-rebase-am (2013-05-29) 1 commit
238
+ - prompt: fix for simple rebase
239
+
240
+ The bash prompt code (in contrib/) displayed the name of the branch
241
+ being rebased when "rebase -i/-m/-p" modes are in use, but not the
242
+ plain vanilla "rebase".
243
+
244
+ Will merge to 'next'.
245
+
246
+
247
+* kb/status-ignored-optim-2 (2013-05-29) 1 commit
248
+ - dir.c: fix ignore processing within not-ignored directories
249
+
250
+ Fix 1.8.3 regressions in the .gitignore path exclusion logic.
251
+
252
+
253
+* ks/difftool-dir-diff-copy-fix (2013-05-29) 1 commit
254
+ - difftool --dir-diff: allow changing any clean working tree file
255
+
256
+ "difftool --dir-diff" did not copy back changes made by the
257
+ end-user in the diff tool backend to the working tree in some
258
+ cases.
259
+
260
+
261
+* rr/push-head (2013-05-29) 3 commits
262
+ - push: make push.default = current use resolved HEAD
263
+ - push: fail early with detached HEAD and current
264
+ - push: factor out the detached HEAD error message
265
+
266
+ "git push $there HEAD:branch" did not resolve HEAD early enough, so
267
+ it was easy to flip it around while push is still going on and push
268
+ out a branch that the user did not originally intended when the
269
+ command was started.
270
+
271
+ Will merge to 'next'.
272
+
273
+
274
+* sb/archive-zip-double-assignment-fix (2013-05-29) 1 commit
275
+ - archive-zip:write_zip_entry: Remove second reset of size variable to zero.
276
+
277
+ Will merge to 'next'.
278
+
279
+--------------------------------------------------
280
+[Stalled]
281
+
282
+* mg/more-textconv (2013-05-10) 7 commits
283
+ (merged to 'next' on 2013-05-21 at 701cdb7)
284
+ + grep: honor --textconv for the case rev:path
285
+ + grep: allow to use textconv filters
286
+ + t7008: demonstrate behavior of grep with textconv
287
+ + cat-file: do not die on --textconv without textconv filters
288
+ + show: honor --textconv for blobs
289
+ + diff_opt: track whether flags have been set explicitly
290
+ + t4030: demonstrate behavior of show with textconv
291
+
292
+ Make "git grep" and "git show" pay attention to --textconv when
293
+ dealing with blob objects.
294
+
295
+ I thought this was pretty well designed and executed, but it seems
296
+ there are some doubts on the list.
297
+
298
+
299
+* mh/multimail (2013-04-21) 1 commit
300
+ - git-multimail: a replacement for post-receive-email
301
+
302
+ Waiting for the initial history to pull from.
303
+ $gmane/223564
304
+
305
+
306
+* jc/format-patch (2013-04-22) 2 commits
307
+ - format-patch: --inline-single
308
+ - format-patch: rename "no_inline" field
309
+
310
+ A new option to send a single patch to the standard output to be
311
+ appended at the bottom of a message. I personally have no need for
312
+ this, but it was easy enough to cobble together. Tests, docs and
313
+ stripping out more MIMEy stuff are left as exercises to interested
314
+ parties.
315
+
316
+ Not ready for inclusion.
317
+
318
+ Will discard unless we hear from anybody who is interested in
319
+ tying its loose ends.
320
+
321
+
322
+* jk/gitweb-utf8 (2013-04-08) 4 commits
323
+ - gitweb: Fix broken blob action parameters on blob/commitdiff pages
324
+ - gitweb: Don't append ';js=(0|1)' to external links
325
+ - gitweb: Make feed title valid utf8
326
+ - gitweb: Fix utf8 encoding for blob_plain, blobdiff_plain, commitdiff_plain, and patch
327
+
328
+ Various fixes to gitweb.
329
+
330
+ Waiting for a reroll after a review.
331
+ Will discard unless we hear from anybody who is interested in
332
+ tying its loose ends.
333
+
334
+* jk/commit-info-slab (2013-04-19) 3 commits
335
+ - commit-slab: introduce a macro to define a slab for new type
336
+ - commit-slab: avoid large realloc
337
+ - commit: allow associating auxiliary info on-demand
338
+ (this branch is used by jc/show-branch.)
339
+
340
+ Technology demonstration to show a way we could use unbound number
341
+ of flag bits on commit objects.
342
+
343
+--------------------------------------------------
344
+[Cooking]
345
+
346
+* rj/mingw-cygwin (2013-05-08) 2 commits
347
+ - cygwin: Remove the CYGWIN_V15_WIN32API build variable
348
+ - mingw: rename WIN32 cpp macro to GIT_WINDOWS_NATIVE
349
+
350
+ Update build for Cygwin 1.[57]. Torsten Bögershausen reports that
351
+ this is fine with Cygwin 1.7 ($gmane/225824) so let's try moving it
352
+ ahead.
353
+
354
+ Will merge to 'next'.
355
+
356
+
357
+* rr/rebase-autostash (2013-05-29) 7 commits
358
+ - rebase: implement --[no-]autostash and rebase.autostash
359
+ - rebase --merge: return control to caller, for housekeeping
360
+ - rebase -i: return control to caller, for housekeeping
361
+ - am: return control to caller, for housekeeping
362
+ - rebase: prepare to do generic housekeeping
363
+ - rebase -i: don't error out if $state_dir already exists
364
+ - am: tighten a conditional that checks for $dotest
365
+
366
+ Will merge to 'next'.
367
+
368
+
369
* nd/urls-doc-no-file-hyperlink-fix (2013-05-24) 1 commit
370
- urls.txt: avoid auto converting to hyperlink
371
@@ -91,8 +426,7 @@ of the repositories listed at
426
weren't expecting the user to modify them and did not take the
427
change back to the working tree.
428
94
- The log message has room for clarification, but the change looked
95
- sane.
429
+ Will merge to 'next'.
430
431
432
* nd/prune-packed-dryrun-verbose (2013-05-28) 1 commit
@@ -116,82 +450,15 @@ of the repositories listed at
450
* xq/credential-osxkeychain (2013-05-28) 1 commit
451
- credential-osxkeychain: support more protocols
452
119
- Will merge to 'next'.
120
-
121
-
122
-* jc/core-checkstat (2013-05-06) 1 commit
123
- - deprecate core.statinfo at Git 2.0 boundary
124
- (this branch is used by jc/core-checkstat-2.0.)
125
-
126
- Will merge to 'next'.
127
-
128
---------------------------------------------------
129
-[Stalled]
130
-
131
-* rr/rebase-autostash (2013-05-12) 7 commits
132
- - rebase: implement --[no-]autostash and rebase.autostash
133
- - rebase --merge: return control to caller, for housekeeping
134
- - rebase -i: return control to caller, for housekeeping
135
- - am: return control to caller, for housekeeping
136
- - rebase: prepare to do generic housekeeping
137
- - rebase -i: don't error out if $state_dir already exists
138
- - am: tighten a conditional that checks for $dotest
139
-
140
- This is from v3, but after a "Fixed" message in $gmane/224111 we
141
- haven't seen a reroll yet. Also there was an attempt for a
142
- follow-up, but it was never completed.
143
-
144
-
145
-* rj/mingw-cygwin (2013-05-08) 2 commits
146
- - cygwin: Remove the CYGWIN_V15_WIN32API build variable
147
- - mingw: rename WIN32 cpp macro to GIT_WINDOWS_NATIVE
148
-
149
- Update build for Cygwin 1.[57]. There seems to have been some
150
- discussion but did anything concrete come out of it???
151
-
152
-
153
-* mh/multimail (2013-04-21) 1 commit
154
- - git-multimail: a replacement for post-receive-email
155
-
156
- Waiting for the initial history to pull from.
157
- $gmane/223564
158
-
159
-
160
-* jc/format-patch (2013-04-22) 2 commits
161
- - format-patch: --inline-single
162
- - format-patch: rename "no_inline" field
163
-
164
- A new option to send a single patch to the standard output to be
165
- appended at the bottom of a message. I personally have no need for
166
- this, but it was easy enough to cobble together. Tests, docs and
167
- stripping out more MIMEy stuff are left as exercises to interested
168
- parties.
169
-
170
- Not ready for inclusion.
171
-
172
-
173
-* jk/gitweb-utf8 (2013-04-08) 4 commits
174
- - gitweb: Fix broken blob action parameters on blob/commitdiff pages
175
- - gitweb: Don't append ';js=(0|1)' to external links
176
- - gitweb: Make feed title valid utf8
177
- - gitweb: Fix utf8 encoding for blob_plain, blobdiff_plain, commitdiff_plain, and patch
178
-
179
- Various fixes to gitweb.
180
-
181
- Waiting for a reroll after a review.
453
+ Will merge to 'next'.
454
455
184
-* jk/commit-info-slab (2013-04-19) 3 commits
185
- - commit-slab: introduce a macro to define a slab for new type
186
- - commit-slab: avoid large realloc
187
- - commit: allow associating auxiliary info on-demand
188
- (this branch is used by jc/show-branch.)
456
+* jc/core-checkstat (2013-05-06) 1 commit
457
+ - deprecate core.statinfo at Git 2.0 boundary
458
+ (this branch is used by jc/core-checkstat-2.0.)
459
190
- Technology demonstration to show a way we could use unbound number
191
- of flag bits on commit objects.
460
+ Will merge to 'next'.
461
193
---------------------------------------------------
194
-[Cooking]
462
463
* tr/line-log (2013-04-22) 13 commits
464
(merged to 'next' on 2013-04-22 at 8f2c1de)
@@ -216,16 +483,6 @@ of the repositories listed at
483
Will merge to 'master'.
484
485
219
-* jn/config-ignore-inaccessible (2013-04-15) 1 commit
220
- (merged to 'next' on 2013-05-07 at 4f09e24)
221
- + config: allow inaccessible configuration under $HOME
222
-
223
- When $HOME is misconfigured to point at an unreadable directory, we
224
- used to complain and die. This loosens the check.
225
-
226
- Will merge to 'master'.
227
-
228
-
486
* mh/reflife (2013-05-28) 25 commits
487
- refs: document the lifetime of the args passed to each_ref_fn
488
- register_ref(): make a copy of the bad reference SHA-1
@@ -257,6 +514,8 @@ of the repositories listed at
514
feeds to its callbacks (in short, "you do not own it, so make a
515
copy if you want to keep it").
516
517
+ Will merge to 'next'.
518
+
519
520
* th/bisect-skip-report-range-fix (2013-05-22) 1 commit
521
- bisect: Fix log output for multi-parent skip ranges
@@ -266,15 +525,14 @@ of the repositories listed at
525
Will merge to 'next'.
526
527
269
-* mm/mediawiki-https-fail-message (2013-05-23) 1 commit
528
+* mm/mediawiki-https-fail-message (2013-05-29) 1 commit
529
- git-remote-mediawiki: better error message when HTTP(S) access fails
530
531
Hint users when https:// connection failed to check the
532
certificate; it is a good hint if we assumie that it is common
533
error for the end users to make.
534
276
- Still under discussion.
277
- $gmane/225678
535
+ Will merge to 'next'.
536
537
538
* tg/maint-zsh-svn-remote-prompt (2013-05-22) 1 commit
@@ -324,7 +582,7 @@ of the repositories listed at
582
When a reflog notation is used for implicit "current branch", we
583
did not say which branch and worse said "branch ''".
584
327
- Waiting for series of rerolls to settle.
585
+ Will merge to 'next'.
586
587
588
* jc/show-branch (2013-05-21) 5 commits
@@ -348,10 +606,7 @@ of the repositories listed at
606
- remote-bzr: fix for files with spaces
607
- remote-bzr: recover from failed clones
608
351
- The ones near the tip conflicted with the hotfix for 1.8.3 so I
352
- discarded them for now.
353
-
354
- Will merge to 'next'?
609
+ Will merge to 'next'.
610
611
612
* jx/clean-interactive (2013-05-22) 15 commits
@@ -386,7 +641,9 @@ of the repositories listed at
641
of the test suite under Valgrind, to speed things up.
642
643
The tip one may be useful in practice but is a tad ugly ;-)
389
- Will merge to 'next'.
644
+
645
+ There seem to be some miscounting by toggling the verbose/valgrind
646
+ mode at wrong places? Cf. $gmane/225735
647
648
649
* rh/merge-options-doc-fix (2013-05-16) 1 commit
@@ -650,23 +907,6 @@ of the repositories listed at
907
Will merge to 'master'.
908
909
653
-* mg/more-textconv (2013-05-10) 7 commits
654
- (merged to 'next' on 2013-05-21 at 701cdb7)
655
- + grep: honor --textconv for the case rev:path
656
- + grep: allow to use textconv filters
657
- + t7008: demonstrate behavior of grep with textconv
658
- + cat-file: do not die on --textconv without textconv filters
659
- + show: honor --textconv for blobs
660
- + diff_opt: track whether flags have been set explicitly
661
- + t4030: demonstrate behavior of show with textconv
662
-
663
- Make "git grep" and "git show" pay attention to --textconv when
664
- dealing with blob objects.
665
-
666
- I thought this was pretty well designed and executed, but it seems
667
- there are some doubts on the list.
668
-
669
-
910
* jh/shorten-refname (2013-05-07) 4 commits
911
- t1514: refname shortening is done after dereferencing symbolic refs
912
- shorten_unambiguous_ref(): Fix shortening refs/remotes/origin/HEAD to origin
@@ -693,8 +933,8 @@ of the repositories listed at
933
Will merge to 'master'.
934
935
696
-* nd/warn-ambiguous-object-name (2013-05-07) 1 commit
697
- - get_sha1: improve ambiguity warning regarding SHA-1 and ref names
936
+* nd/warn-ambiguous-object-name (2013-05-29) 1 commit
937
+ - get_sha1: warn about full or short object names that look like refs
938
939
"git cmd <name>", when <name> happens to be a 40-hex string,
940
directly uses the 40-hex string as an object name, even if a ref
@@ -703,12 +943,9 @@ of the repositories listed at
943
like we warn when more than one refs/ hierachies share the same
944
name.
945
706
- Will be rerolled.
707
- $gmane/225585
708
-
946
710
-* fc/zsh-leftover-bits (2013-05-28) 3 commits
711
- (merged to 'next' on 2013-05-28 at d4a7aee)
947
+* fc/zsh-leftover-bits (2013-05-29) 3 commits
948
+ (merged to 'next' on 2013-05-29 at b367b4e)
949
+ completion: zsh: improve bash script loading
950
(merged to 'next' on 2013-05-21 at 1b84ec1)
951
+ completion: synchronize zsh wrapper
@@ -782,26 +1019,6 @@ of the repositories listed at
1019
Will merge to 'next'.
1020
1021
785
-* jk/lookup-object-prefer-latest (2013-05-02) 1 commit
786
- (merged to 'next' on 2013-05-06 at cc59dcc)
787
- + lookup_object: prioritize recently found objects
788
-
789
- Optimizes object lookup when the object hashtable starts to become
790
- crowded.
791
-
792
- Will merge to 'master'.
793
-
794
-
795
-* jk/subtree-do-not-push-if-split-fails (2013-05-01) 1 commit
796
- (merged to 'next' on 2013-05-06 at 81bdf37)
797
- + contrib/subtree: don't delete remote branches if split fails
798
-
799
- "git subtree" (in contrib/) had one codepath with loose error
800
- checks to lose data at the remote side.
801
-
802
- Will merge to 'master'.
803
-
804
-
1022
* fc/completion (2013-04-27) 9 commits
1023
(merged to 'next' on 2013-05-21 at aba6ef1)
1024
+ completion: remove __git_index_file_list_filter()
@@ -817,190 +1034,6 @@ of the repositories listed at
1034
Will merge to 'master'.
1035
1036
820
-* jk/test-output (2013-05-06) 3 commits
821
- (merged to 'next' on 2013-05-06 at 7c03af3)
822
- + t/Makefile: don't define TEST_RESULTS_DIRECTORY recursively
823
- (merged to 'next' on 2013-05-01 at 63827c9)
824
- + test output: respect $TEST_OUTPUT_DIRECTORY
825
- + t/Makefile: fix result handling with TEST_OUTPUT_DIRECTORY
826
-
827
- When TEST_OUTPUT_DIRECTORY setting is used, it was handled somewhat
828
- inconsistently between the test framework and t/Makefile, and logic
829
- to summarize the results looked at a wrong place.
830
-
831
- Will merge to 'master'.
832
-
833
-
834
-* rj/sparse (2013-04-28) 10 commits
835
- (merged to 'next' on 2013-05-01 at 649e16c)
836
- + sparse: Fix mingw_main() argument number/type errors
837
- + compat/mingw.c: Fix some sparse warnings
838
- + compat/win32mmap.c: Fix some sparse warnings
839
- + compat/poll/poll.c: Fix a sparse warning
840
- + compat/win32/pthread.c: Fix a sparse warning
841
- + compat/unsetenv.c: Fix a sparse warning
842
- + compat/nedmalloc: Fix compiler warnings on linux
843
- + compat/nedmalloc: Fix some sparse warnings
844
- + compat/fnmatch/fnmatch.c: Fix a sparse error
845
- + compat/regex/regexec.c: Fix some sparse warnings
846
-
847
- Will merge to 'master'.
848
-
849
-
850
-* mh/fetch-into-shallow (2013-05-02) 2 commits
851
- (merged to 'next' on 2013-05-03 at 3fadc61)
852
- + t5500: add test for fetching with an unknown 'shallow'
853
- (merged to 'next' on 2013-04-29 at a167d3e)
854
- + upload-pack: ignore 'shallow' lines with unknown obj-ids
855
-
856
- Will merge to 'master'.
857
-
858
-
859
-* jh/checkout-auto-tracking (2013-04-21) 8 commits
860
- (merged to 'next' on 2013-04-22 at 2356700)
861
- + glossary: Update and rephrase the definition of a remote-tracking branch
862
- + branch.c: Validate tracking branches with refspecs instead of refs/remotes/*
863
- + t9114.2: Don't use --track option against "svn-remote"-tracking branches
864
- + t7201.24: Add refspec to keep --track working
865
- + t3200.39: tracking setup should fail if there is no matching refspec.
866
- + checkout: Use remote refspecs when DWIMming tracking branches
867
- + t2024: Show failure to use refspec when DWIMming remote branch names
868
- + t2024: Add tests verifying current DWIM behavior of 'git checkout <branch>'
869
-
870
- Updates "git checkout foo" that DWIMs the intended "upstream" and
871
- turns it into "git checkout -t -b foo remotes/origin/foo" to
872
- correctly take existing remote definitions into account. The
873
- remote "origin" may be what uniquely map its own branch to
874
- remotes/some/where/foo but that some/where may not be "origin".
875
-
876
- Will merge to 'master'.
877
-
878
-
879
-* jc/prune-all (2013-04-25) 4 commits
880
- (merged to 'next' on 2013-04-26 at 97a7387)
881
- + prune: introduce OPT_EXPIRY_DATE() and use it
882
- (merged to 'next' on 2013-04-22 at b00ccf6)
883
- + api-parse-options.txt: document "no-" for non-boolean options
884
- + git-gc.txt, git-reflog.txt: document new expiry options
885
- + date.c: add parse_expiry_date()
886
- (this branch is used by mh/packed-refs-various.)
887
-
888
- We used the approxidate() parser for "--expire=<timestamp>" options
889
- of various commands, but it is better to treat --expire=all and
890
- --expire=now a bit more specially than using the current timestamp.
891
- Update "git gc" and "git reflog" with a new parsing function for
892
- expiry dates.
893
-
894
- Will merge to 'master'.
895
-
896
-
897
-* as/check-ignore (2013-04-29) 6 commits
898
- (merged to 'next' on 2013-04-30 at 646931f)
899
- + t0008: use named pipe (FIFO) to test check-ignore streaming
900
- (merged to 'next' on 2013-04-21 at 7515aa8)
901
- + Documentation: add caveats about I/O buffering for check-{attr,ignore}
902
- + check-ignore: allow incremental streaming of queries via --stdin
903
- + check-ignore: move setup into cmd_check_ignore()
904
- + check-ignore: add -n / --non-matching option
905
- + t0008: remove duplicated test fixture data
906
-
907
- Enhance "check-ignore" (1.8.2 update) to work more like "check-attr"
908
- over bidi-pipes.
909
-
910
- Will merge to 'master'.
911
-
912
-
913
-* mh/packed-refs-various (2013-05-01) 33 commits
914
- (merged to 'next' on 2013-05-01 at e527153)
915
- + refs: handle the main ref_cache specially
916
- + refs: change do_for_each_*() functions to take ref_cache arguments
917
- + pack_one_ref(): do some cheap tests before a more expensive one
918
- + pack_one_ref(): use write_packed_entry() to do the writing
919
- + pack_one_ref(): use function peel_entry()
920
- + refs: inline function do_not_prune()
921
- + pack_refs(): change to use do_for_each_entry()
922
- + refs: use same lock_file object for both ref-packing functions
923
- + pack_one_ref(): rename "path" parameter to "refname"
924
- + pack-refs: merge code from pack-refs.{c,h} into refs.{c,h}
925
- + pack-refs: rename handle_one_ref() to pack_one_ref()
926
- + refs: extract a function write_packed_entry()
927
- + repack_without_ref(): write peeled refs in the rewritten file
928
- + t3211: demonstrate loss of peeled refs if a packed ref is deleted
929
- + refs: change how packed refs are deleted
930
- + search_ref_dir(): return an index rather than a pointer
931
- + repack_without_ref(): silence errors for dangling packed refs
932
- + t3210: test for spurious error messages for dangling packed refs
933
- + refs: change the internal reference-iteration API
934
- + refs: extract a function peel_entry()
935
- + peel_ref(): fix return value for non-peelable, not-current reference
936
- + peel_object(): give more specific information in return value
937
- + refs: extract function peel_object()
938
- + refs: extract a function ref_resolves_to_object()
939
- + repack_without_ref(): use function get_packed_ref()
940
- + peel_ref(): use function get_packed_ref()
941
- + get_packed_ref(): return a ref_entry
942
- + do_for_each_ref_in_dirs(): remove dead code
943
- + refs: define constant PEELED_LINE_LENGTH
944
- + refs: document how current_ref is used
945
- + refs: document do_for_each_ref() and do_one_ref()
946
- + refs: document the fields of struct ref_value
947
- + refs: document flags constants REF_*
948
- (this branch uses jc/prune-all.)
949
-
950
- Updates reading and updating packed-refs file, correcting corner
951
- case bugs.
952
-
953
- Will merge to 'master'.
954
-
955
-
956
-* fc/transport-helper-error-reporting (2013-05-10) 12 commits
957
- (merged to 'next' on 2013-05-10 at 2a9af4b)
958
- + transport-helper: fix remote helper namespace regression
959
- + test: remote-helper: add missing and
960
- (merged to 'next' on 2013-04-25 at 3358f1a)
961
- + t5801: "VAR=VAL shell_func args" is forbidden
962
- (merged to 'next' on 2013-04-22 at 5ba6467)
963
- + transport-helper: update remote helper namespace
964
- + transport-helper: trivial code shuffle
965
- + transport-helper: warn when refspec is not used
966
- + transport-helper: clarify pushing without refspecs
967
- + transport-helper: update refspec documentation
968
- + transport-helper: clarify *:* refspec
969
- + transport-helper: improve push messages
970
- + transport-helper: mention helper name when it dies
971
- + transport-helper: report errors properly
972
- (this branch is tangled with js/transport-helper-error-reporting-fix.)
973
-
974
- Update transport helper to report errors and maintain ref hierarchy
975
- used to keep track of remote helper state better.
976
-
977
- Will merge to 'master'.
978
-
979
-
980
-* js/transport-helper-error-reporting-fix (2013-04-28) 13 commits
981
- (merged to 'next' on 2013-04-29 at 8cc4bb8)
982
- + git-remote-testgit: build it to run under $SHELL_PATH
983
- + git-remote-testgit: further remove some bashisms
984
- + git-remote-testgit: avoid process substitution
985
- (merged to 'next' on 2013-04-25 at 3358f1a)
986
- + t5801: "VAR=VAL shell_func args" is forbidden
987
- (merged to 'next' on 2013-04-22 at 5ba6467)
988
- + transport-helper: update remote helper namespace
989
- + transport-helper: trivial code shuffle
990
- + transport-helper: warn when refspec is not used
991
- + transport-helper: clarify pushing without refspecs
992
- + transport-helper: update refspec documentation
993
- + transport-helper: clarify *:* refspec
994
- + transport-helper: improve push messages
995
- + transport-helper: mention helper name when it dies
996
- + transport-helper: report errors properly
997
- (this branch is tangled with fc/transport-helper-error-reporting.)
998
-
999
- Finishing touches to fc/transport-helper-error-reporting topic.
1000
-
1001
- Will merge to 'master'.
1002
-
1003
-
1037
* jk/submodule-subdirectory-ok (2013-04-24) 3 commits
1038
(merged to 'next' on 2013-04-24 at 6306b29)
1039
+ submodule: fix quoting in relative_path()
@@ -1056,21 +1089,3 @@ of the repositories listed at
1089
aka "--no-ignore-removal".
1090
1091
Will cook in 'next' until Git 2.0.
1059
-
1060
---------------------------------------------------
1061
-[Discarded]
1062
-
1063
-* kb/full-history-compute-treesame-carefully (2013-05-06) 11 commits
1064
- . revision.c: treat A...B merge bases as if manually specified
1065
- . revision.c: discount side branches when computing TREESAME
1066
- . simplify-merges: drop merge from irrelevant side branch
1067
- . simplify-merges: never remove all TREESAME parents
1068
- . t6012: update test for tweaked full-history traversal
1069
- . revision.c: Make --full-history consider more merges
1070
- . rev-list-options.txt: correct TREESAME for P
1071
- . t6111: allow checking the parents as well
1072
- . t6111: new TREESAME test set
1073
- . t6019: test file dropped in -s ours merge
1074
- . decorate.c: compact table when growing
1075
-
1076
- Superseded by kb/full-history-compute-treesame-carefully-2