What's cooking (2012/10 #01)
Junio C Hamano committed
Oct 2, 2012 at 16:14 UTC
d6f2734478ae670942d2412d015c29366374c816
1 file changed
+129
-169
whats-cooking.txt
+129
-169
@@ -1,17 +1,24 @@
1
To: git@vger.kernel.org
2
Bcc: lwn@lwn.net
3
-Subject: What's cooking in git.git (Sep 2012, #10; Sun, 30)
4
-X-master-at: 261b5119c7a7cb778dd812cfb006b511c5bc4065
5
-X-next-at: 8e89b3c30625b00571c4cbc5e5a74d8e3e442754
3
+Subject: What's cooking in git.git (Oct 2012, #01; Tue, 2)
4
+X-master-at: b65f30b6b3bf34831b32a1b209bc1955f2bf79df
5
+X-next-at: e57700a8bba97a7ca41f5703b8cd8de489f06c07
6
7
-What's cooking in git.git (Sep 2012, #10; Sun, 30)
7
+What's cooking in git.git (Oct 2012, #01; Tue, 2)
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
-The tip of 'master' will be tagged as 1.8.0-rc0 real soon now.
14
+The tip of 'master' is a bit past v1.8.0-rc0; we are more or less
15
+feature complete, except for a few topics marked as "Will merge to
16
+'master' in the list below.
17
+
18
+I do not expect the topics in the stalled category will be ready for
19
+the upcoming release; they may appear in the release after v1.8.0 if
20
+they are rerolled. Also some of the topics that have been cooking
21
+in 'next' have to be deferred to the next cycle.
22
23
I'm planning to keep this cycle reasonably short and aim for tagging
24
the result as 1.8.0 at the end of 9th week, on October 21st, after
@@ -24,123 +31,86 @@ repositories listed at
31
http://git-blame.blogspot.com/p/git-public-repositories.html
32
33
--------------------------------------------------
27
-[Graduated to "master"]
28
-
29
-* aw/rebase-i-edit-todo (2012-09-19) 5 commits
30
- (merged to 'next' on 2012-09-24 at 95625d7)
31
- + rebase -i: suggest using --edit-todo to fix an unknown instruction
32
- + rebase -i: Add tests for "--edit-todo"
33
- + rebase -i: Teach "--edit-todo" action
34
- + rebase -i: Refactor help messages for todo file
35
- + rebase usage: subcommands can not be combined with -i
36
-
37
- Teach an option to edit the insn sheet to "git rebase -i".
38
-
39
-
40
-* jk/smart-http-switch (2012-09-21) 2 commits
41
- (merged to 'next' on 2012-09-24 at b2a2777)
42
- + remote-curl: let users turn off smart http
43
- + remote-curl: rename is_http variable
44
- (this branch uses sp/maint-http-info-refs-no-retry.)
45
-
46
- Allows users to turn off smart-http when talking to dumb-only
47
- servers.
48
-
49
-
50
-* js/rebase-exec-command-not-found (2012-09-18) 1 commit
51
- (merged to 'next' on 2012-09-24 at ed5b048)
52
- + rebase -i: fix misleading error message after 'exec no-such' instruction
34
+[New Topics]
35
54
- When you misspell the command name you give to the "exec" action in
55
- the "git rebase -i" insn sheet, you are told that 'rebase' is not a
56
- git subcommand from "git rebase --continue".
36
+* bw/config-lift-variable-name-length-limit (2012-10-01) 1 commit
37
+ - Remove the hard coded length limit on variable names in config files
38
39
+ The configuration parser had an unnecessary hardcoded limit on
40
+ variable names that was not checked consistently. Lift the limit.
41
59
-* rr/maint-submodule-unknown-cmd (2012-09-25) 1 commit
60
- (merged to 'next' on 2012-09-25 at 4bb1bc4)
61
- + submodule: if $command was not matched, don't parse other args
42
+ Will merge to 'next'.
43
63
- "git submodule frotz" was not diagnosed as "frotz" being an unknown
64
- subcommand to "git submodule"; the user instead got a complaint
65
- that "git submodule status" was run with an unknown path "frotz".
44
45
+* jc/maint-t1450-fsck-order-fix (2012-10-02) 1 commit
46
+ - t1450: the order the objects are checked is undefined
47
68
-* rr/test-use-shell-path-not-shell (2012-09-25) 1 commit
69
- (merged to 'next' on 2012-09-27 at 3a1848f)
70
- + test-lib: use $SHELL_PATH, not $SHELL
48
+ The fsck test assumed too much on what kind of error it will
49
+ detect. The only important thing is the inconsistency is detected
50
+ as an error.
51
52
+ Will merge to 'next'.
53
73
-* rs/archive-zip-utf8 (2012-09-24) 4 commits
74
- (merged to 'next' on 2012-09-24 at 9b42b1c)
75
- + archive-zip: write extended timestamp
76
- + archive-zip: support UTF-8 paths
77
- + Revert "archive-zip: support UTF-8 paths"
78
- (merged to 'next' on 2012-09-11 at 3b1f071)
79
- + archive-zip: support UTF-8 paths
54
+--------------------------------------------------
55
+[Graduated to "master"]
56
81
- With another reroll, it looks like the series is as polished as it
82
- could be.
57
+* da/mergetool-custom (2012-09-25) 1 commit
58
+ (merged to 'next' on 2012-09-27 at f910851)
59
+ + mergetool--lib: Allow custom commands to override built-ins
60
61
+ The actual external command to run for mergetool backend can be
62
+ specified with difftool/mergetool.$name.cmd configuration
63
+ variables, but this mechanism was ignored for the backends we
64
+ natively support.
65
85
-* sp/maint-http-enable-gzip (2012-09-20) 1 commit
86
- (merged to 'next' on 2012-09-24 at 1d05142)
87
- + Enable info/refs gzip decompression in HTTP client
66
89
- Allows a more common 'gzip' Accept-Encoding to be used.
67
+* ep/malloc-check-perturb (2012-09-26) 1 commit
68
+ (merged to 'next' on 2012-09-27 at f115b8b)
69
+ + MALLOC_CHECK: enable it, unless disabled explicitly
70
71
+ Fixes a brown-paper bag bug.
72
92
-* sp/maint-http-info-refs-no-retry (2012-09-20) 1 commit
93
- (merged to 'next' on 2012-09-24 at ebcce8e)
94
- + Revert "retry request without query when info/refs?query fails"
95
- (this branch is used by jk/smart-http-switch.)
73
97
- Kills an old workaround for an unlikely server misconfiguration
98
- that hurts debuggability.
74
+* jc/blame-follows-renames (2012-09-21) 1 commit
75
+ (merged to 'next' on 2012-09-27 at 12634c1)
76
+ + git blame: document that it always follows origin across whole-file renames
77
+ (this branch is used by jc/blame-no-follow.)
78
100
---------------------------------------------------
101
-[New Topics]
79
+ Clarify the "blame" documentation to tell the users that there is
80
+ no need to ask for "--follow".
81
103
-* jl/submodule-add-by-name (2012-09-30) 2 commits
104
- - submodule add: Fail when .git/modules/<name> already exists unless forced
105
- - Teach "git submodule add" the --name option
106
-
107
- If you remove a submodule, in order to keep the repository so that
108
- "git checkout" to an older commit in the superproject history can
109
- resurrect the submodule, the real repository will stay in $GIT_DIR
110
- of the superproject. A later "git submodule add $path" to add a
111
- different submodule at the same path will fail. Diagnose this case
112
- a bit better, and if the user really wants to add an unrelated
113
- submodule at the same path, give the "--name" option to give it a
114
- place in $GIT_DIR of the superproject that does not conflict with
115
- the original submodule.
82
83
+* jk/completion-tests (2012-09-27) 2 commits
84
+ (merged to 'next' on 2012-09-27 at 5cd6968)
85
+ + t9902: add completion tests for "odd" filenames
86
+ + t9902: add a few basic completion tests
87
118
-* nd/grep-reflog (2012-09-29) 4 commits
119
- - revision: make --grep search in notes too if shown
120
- - log --grep-reflog: reject the option without -g
121
- - revision: add --grep-reflog to filter commits by reflog messages
122
- - grep: prepare for new header field filter
88
124
- Teach the commands from the "log" family the "--grep-reflog" option
125
- to limit output by string that appears in the reflog entry when the
126
- --walk-reflogs option is in effect.
89
+* jk/receive-pack-unpack-error-to-pusher (2012-09-21) 3 commits
90
+ (merged to 'next' on 2012-09-27 at 90f0c6f)
91
+ + receive-pack: drop "n/a" on unpacker errors
92
+ + receive-pack: send pack-processing stderr over sideband
93
+ + receive-pack: redirect unpack-objects stdout to /dev/null
94
95
+ Send errors from "unpack-objects" and "index-pack" back to the "git
96
+ push" over the git and smart-http protocols, just like it is done
97
+ for a push over the ssh protocol.
98
129
-* ep/malloc-check-perturb (2012-09-26) 1 commit
130
- (merged to 'next' on 2012-09-27 at f115b8b)
131
- + MALLOC_CHECK: enable it, unless disabled explicitly
99
133
- Fixes a brown-paper bag bug.
100
+* os/commit-submodule-ignore (2012-09-24) 1 commit
101
+ (merged to 'next' on 2012-09-27 at 9cd2cfd)
102
+ + commit: pay attention to submodule.$name.ignore in .gitmodules
103
135
- Will merge to 'master'.
104
+ "git status" honored the ignore=dirty settings in .gitmodules but
105
+ "git commit" didn't.
106
107
138
-* lt/mailinfo-handle-attachment-more-sanely (2012-09-30) 1 commit
139
- - mailinfo: don't require "text" mime type for attachments
108
+* rt/maint-clone-single (2012-09-20) 1 commit
109
+ (merged to 'next' on 2012-09-27 at a47d54d)
110
+ + clone --single: limit the fetch refspec to fetched branch
111
141
- A patch attached as application/octet-stream (e.g. not text/*) were
142
- mishandled,not correctly honoring Content-Transfer-Encoding
143
- (e.g. base64).
112
+ Running "git fetch" in a repository made with "git clone --single"
113
+ slurps all the branches, defeating the point of "--single".
114
115
--------------------------------------------------
116
[Stalled]
@@ -171,7 +141,7 @@ repositories listed at
141
- Improve documentation and comments regarding directory traversal API
142
- Update directory listing API doc to match code
143
174
- Expecting a further reroll to follow-up review comments.
144
+ Expecting a reroll.
145
146
147
* as/test-tweaks (2012-09-20) 7 commits
@@ -259,7 +229,6 @@ repositories listed at
229
* jc/blame-no-follow (2012-09-21) 2 commits
230
- blame: pay attention to --no-follow
231
- diff: accept --no-follow option
262
- (this branch uses jc/blame-follows-renames.)
232
233
Teaches "--no-follow" option to "git blame" to disable its
234
whole-file rename detection.
@@ -267,6 +236,16 @@ repositories listed at
236
Stalled mostly due to lack of responses.
237
238
239
+* mk/maint-graph-infinity-loop (2012-09-25) 1 commit
240
+ - graph.c: infinite loop in git whatchanged --graph -m
241
+
242
+ The --graph code fell into infinite loop when asked to do what the
243
+ code did not expect ;-)
244
+
245
+ Anybody who worked on "--graph" wants to comment?
246
+ Stalled mostly due to lack of responses.
247
+
248
+
249
* ph/credential-refactor (2012-09-02) 5 commits
250
- wincred: port to generic credential helper
251
- Merge branch 'ef/win32-cred-helper' into ph/credential-refactor
@@ -294,7 +273,7 @@ repositories listed at
273
link pointing at it.
274
275
Seems broken as it makes $(pwd) and TRASH_DIRECTORY inconsistent.
297
- Needs rerolling.
276
+ Will discard.
277
278
279
* jc/maint-push-refs-all (2012-08-27) 2 commits
@@ -382,46 +361,68 @@ repositories listed at
361
--------------------------------------------------
362
[Cooking]
363
385
-* da/mergetool-custom (2012-09-25) 1 commit
386
- (merged to 'next' on 2012-09-27 at f910851)
387
- + mergetool--lib: Allow custom commands to override built-ins
364
+* jl/submodule-add-by-name (2012-09-30) 2 commits
365
+ - submodule add: Fail when .git/modules/<name> already exists unless forced
366
+ - Teach "git submodule add" the --name option
367
389
- The actual external command to run for mergetool backend can be
390
- specified with difftool/mergetool.$name.cmd configuration
391
- variables, but this mechanism was ignored for the backends we
392
- natively support.
368
+ If you remove a submodule, in order to keep the repository so that
369
+ "git checkout" to an older commit in the superproject history can
370
+ resurrect the submodule, the real repository will stay in $GIT_DIR
371
+ of the superproject. A later "git submodule add $path" to add a
372
+ different submodule at the same path will fail. Diagnose this case
373
+ a bit better, and if the user really wants to add an unrelated
374
+ submodule at the same path, give the "--name" option to give it a
375
+ place in $GIT_DIR of the superproject that does not conflict with
376
+ the original submodule.
377
394
- Will merge to 'master'.
378
+ Will merge to 'next'.
379
380
397
-* mk/maint-graph-infinity-loop (2012-09-25) 1 commit
398
- - graph.c: infinite loop in git whatchanged --graph -m
381
+* nd/grep-reflog (2012-09-29) 4 commits
382
+ (merged to 'next' on 2012-10-01 at 57773a6)
383
+ + revision: make --grep search in notes too if shown
384
+ + log --grep-reflog: reject the option without -g
385
+ + revision: add --grep-reflog to filter commits by reflog messages
386
+ + grep: prepare for new header field filter
387
400
- The --graph code fell into infinite loop when asked to do what the
401
- code did not expect ;-)
388
+ Teach the commands from the "log" family the "--grep-reflog" option
389
+ to limit output by string that appears in the reflog entry when the
390
+ "--walk-reflogs" option is in effect.
391
403
- Anybody who worked on --graph wants to comment?
392
+ Will merge to 'master'.
393
394
406
-* tu/gc-auto-quiet (2012-09-27) 1 commit
407
- - silence git gc --auto --quiet output
395
+* lt/mailinfo-handle-attachment-more-sanely (2012-09-30) 1 commit
396
+ (merged to 'next' on 2012-10-01 at 2a1cecc)
397
+ + mailinfo: don't require "text" mime type for attachments
398
409
- Will merge to 'next'.
399
+ A patch attached as application/octet-stream (e.g. not text/*) were
400
+ mishandled, not correctly honoring Content-Transfer-Encoding
401
+ (e.g. base64).
402
403
+ Will merge to 'master'.
404
412
-* jk/completion-tests (2012-09-27) 2 commits
413
- (merged to 'next' on 2012-09-27 at 5cd6968)
414
- + t9902: add completion tests for "odd" filenames
415
- + t9902: add a few basic completion tests
405
+
406
+* tu/gc-auto-quiet (2012-09-27) 1 commit
407
+ (merged to 'next' on 2012-10-01 at ad8b91b)
408
+ + silence git gc --auto --quiet output
409
+
410
+ "gc --auto" notified the user that auto-packing has triggered even
411
+ under the "--quiet" option.
412
413
Will merge to 'master'.
414
415
420
-* jm/diff-context-config (2012-09-27) 2 commits
421
- - [SQUASH] test fixups
422
- - diff: diff.context configuration gives default to -U
416
+* jm/diff-context-config (2012-10-02) 2 commits
417
+ (merged to 'next' on 2012-10-02 at e57700a)
418
+ + t4055: avoid use of sed 'a' command
419
+ (merged to 'next' on 2012-10-01 at 509a558)
420
+ + diff: diff.context configuration gives default to -U
421
+
422
+ Teaches a new configuration variable to "git diff" Porcelain and
423
+ its friends.
424
424
- Will merge to 'next' after squashing.
425
+ Will defer to the next cycle.
426
427
428
* mh/ceiling (2012-09-29) 9 commits
@@ -440,10 +441,12 @@ repositories listed at
441
logic to escape the ceilings the user thought to have specified.
442
443
The solution felt a bit unnecessarily convoluted to me.
444
+ Expecting a reroll.
445
446
447
* jl/submodule-rm (2012-09-29) 1 commit
446
- - submodule: teach rm to remove submodules unless they contain a git directory
448
+ (merged to 'next' on 2012-10-01 at 4e5c4fc)
449
+ + submodule: teach rm to remove submodules unless they contain a git directory
450
451
"git rm submodule" cannot blindly remove a submodule directory as
452
its working tree may have local changes, and worse yet, it may even
@@ -453,7 +456,7 @@ repositories listed at
456
is not embedded in its working tree but is elsewhere and uses the
457
gitfile mechanism to point at it.
458
456
- Will merge to 'next'.
459
+ Will defer to the next cycle.
460
461
462
* nd/wildmatch (2012-09-27) 5 commits
@@ -470,15 +473,7 @@ repositories listed at
473
It was pointed out that some symbols that do not have to be global
474
are left global. I think this reroll fixed most of them.
475
473
-
474
-* os/commit-submodule-ignore (2012-09-24) 1 commit
475
- (merged to 'next' on 2012-09-27 at 9cd2cfd)
476
- + commit: pay attention to submodule.$name.ignore in .gitmodules
477
-
478
- "git status" honored the ignore=dirty settings in .gitmodules but
479
- "git commit" didn't.
480
-
481
- Will merge to 'master'.
476
+ Will merge to 'next'.
477
478
479
* nd/pretty-placeholder-with-color-option (2012-09-30) 9 commits
@@ -493,44 +488,9 @@ repositories listed at
488
- pretty: share code between format_decoration and show_decorations
489
490
496
-* jk/receive-pack-unpack-error-to-pusher (2012-09-21) 3 commits
497
- (merged to 'next' on 2012-09-27 at 90f0c6f)
498
- + receive-pack: drop "n/a" on unpacker errors
499
- + receive-pack: send pack-processing stderr over sideband
500
- + receive-pack: redirect unpack-objects stdout to /dev/null
501
-
502
- Send errors from "unpack-objects" and "index-pack" back to the "git
503
- push" over the git and smart-http protocols, just like it is done
504
- for a push over the ssh protocol.
505
-
506
- Will merge to 'master'.
507
-
508
-
509
-* jc/blame-follows-renames (2012-09-21) 1 commit
510
- (merged to 'next' on 2012-09-27 at 12634c1)
511
- + git blame: document that it always follows origin across whole-file renames
512
- (this branch is used by jc/blame-no-follow.)
513
-
514
- Clarify the "blame" documentation to tell the users that there is
515
- no need to ask for "--follow".
516
-
517
- Will merge to 'master'.
518
-
519
-
520
-* rt/maint-clone-single (2012-09-20) 1 commit
521
- (merged to 'next' on 2012-09-27 at a47d54d)
522
- + clone --single: limit the fetch refspec to fetched branch
523
-
524
- Running "git fetch" in a repository made with "git clone --single"
525
- slurps all the branches, defeating the point of "--single".
526
-
527
- Will merge to 'master'.
528
-
529
-
491
* jk/no-more-pre-exec-callback (2012-06-05) 1 commit
492
- pager: drop "wait for output to run less" hack
493
494
(Originally merged to 'next' on 2012-07-23)
495
535
- Will defer until the end of the 2012.
536
- while waiting for older "less" to go extinct.
496
+ Will defer to the next cycle.