What's cooking (2015/07 #06)

Junio C Hamano committed Jul 21, 2015 at 14:59 UTC 163e88a64fd6b0026d58657841890152be98a026
1 file changed +241 -241
whats-cooking.txt
+241 -241
@@ -1,105 +1,182 @@
1 To: git@vger.kernel.org
2 Bcc: lwn@lwn.net
3 -Subject: What's cooking in git.git (Jul 2015, #05; Mon, 20)
4 -X-master-at: fbdeabf1f0992e41237caee8fc357849e0038a82
5 -X-next-at: 76e840bed084a99e9de166e429c25bc0b2d6a698
3 +Subject: What's cooking in git.git (Jul 2015, #06; Tue, 21)
4 +X-master-at: f99a38c0121456822f8a9dfb7928eefceaa98201
5 +X-next-at: 68aafd0cff0775f143dd13a8f2d95a3e65155388
6
7 -What's cooking in git.git (Jul 2015, #05; Mon, 20)
7 +What's cooking in git.git (Jul 2015, #06; Tue, 21)
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 +Tagged v2.5-rc3 today; hopefully we can have an uneventful 2.5 next
15 +week and then start the next cycle. I'll eject/drop the stalled
16 +topics and ask people to reroll/rebase if their topics are still
17 +viable.
18 +
19 You can find the changes described here in the integration branches
20 of the repositories listed at
21
22 http://git-blame.blogspot.com/p/git-public-repositories.html
23
24 --------------------------------------------------
20 -[New Topics]
25 +[Stalled]
26
22 -* zb/userdiff-fountain (2015-07-17) 1 commit
23 - - userdiff: add support for Fountain documents
27 +* jc/clone-bundle (2015-04-30) 1 commit
28 + - repack: optionally create a clone.bundle
29
25 - New userdiff pattern definition for fountain screenwriting markup
26 - format.
30 + Waiting for further work.
31 + Still an early WIP.
32
28 - Needs another reroll? ($gmane/274354)
33
34 +* jh/strbuf-read-use-read-in-full (2015-06-01) 1 commit
35 + - strbuf_read(): skip unnecessary strbuf_grow() at eof
36
31 -* as/sparse-checkout-removal (2015-07-17) 1 commit
32 - - unpack-trees: don't update files with CE_WT_REMOVE set
37 + Avoid one extra iteration and strbuf_grow() of 8kB in
38 + strbuf_read().
39
34 - Duy's replacement needs to be applied ($gmane/274158).
40 + Looked reasonable; perhaps a log message clarification is needed.
41
42 + Expecting a reroll.
43
37 -* cb/uname-in-untracked (2015-07-17) 1 commit
38 - - untracked: fix detection of uname(2) failure
44
40 - An experimental "untracked cache" feature used to use uname(2) in a
41 - slightly unportable way.
45 +* mg/index-read-error-messages (2015-06-01) 2 commits
46 + - messages: uniform error messages for index write
47 + - show-index: uniform error messages for index read
48
43 - Will merge to 'next'.
49 + The tip was RFC.
50 + Expecting a reroll.
51
52
46 -* jc/rerere-multi (2015-07-20) 5 commits
47 - - t4200: rerere a merge with two identical conflicts
48 - - rerere: delay the recording of preimage for the first time
49 - - rerere: handle leftover rr-cache/$ID directory and postimage files
50 - - rerere: scan $GIT_DIR/rr-cache/$ID when instantiating a rerere_id
51 - - rerere: split conflict ID further
52 - (this branch uses jc/rerere.)
53 +* hv/submodule-config (2015-06-15) 4 commits
54 + - do not die on error of parsing fetchrecursesubmodules option
55 + - use new config API for worktree configurations of submodules
56 + - extract functions for submodule config set and lookup
57 + - implement submodule config API for lookup of .gitmodules values
58
54 - This is a contination of jc/rerere topic.
59 + The gitmodules API accessed from the C code learned to cache stuff
60 + lazily.
61
62 + Needs another reroll? ($gmane/273743).
63
57 -* bw/portability-solaris (2015-07-20) 3 commits
58 - - tests: fix sed usage in tests to work around broken xpg4/sed on Solaris
59 - - tests: fix sed usage in tests to work around broken xpg4/sed on Solaris
60 - - tests: modify tr expressions so that xpg4/tr handles it on Solaris
64
62 - Needs another reroll? ($gmane/274296)
65 +* jk/log-missing-default-HEAD (2015-06-03) 1 commit
66 + - log: diagnose empty HEAD more clearly
67
68 + "git init empty && git -C empty log" said "bad default revision 'HEAD'",
69 + which was found to be a bit confusing to new users.
70
65 -* es/worktree-add (2015-07-20) 5 commits
66 - (merged to 'next' on 2015-07-20 at 76e840b)
67 - + config: rename "gc.pruneWorktreesExpire" to "gc.worktreePruneExpire"
68 - + Documentation/git-worktree: wordsmith worktree-related manpages
69 - + Documentation/config: fix stale "git prune --worktree" reference
70 - + Documentation/git-worktree: fix incorrect reference to file "locked"
71 - + Documentation/git-worktree: consistently use term "linked working tree"
72 - (this branch is used by es/worktree-add-cleanup.)
71 + What's the status of this one?
72
74 - Will merge to 'master'.
73
74 +* bc/object-id (2015-06-17) 10 commits
75 + . remote.c: use struct object_id in many functions
76 + . object-id: use struct object_id in struct object
77 + . remote.c: use struct object_id in ref_newer()
78 + . transport-helper.c: use struct object_id in push_refs_with_export()
79 + . connect.c: use struct object_id in get_remote_heads()
80 + . remote-curl: use struct object_id in parse_fetch()
81 + . fetch-pack: use struct object_id in add_sought_entry_mem()
82 + . object_id: convert struct ref to use object_id.
83 + . sha1_file: introduce has_object_file() helper
84 + . refs: convert some internal functions to use object_id
85
77 -* sg/bash-prompt-untracked-optim (2015-07-20) 2 commits
78 - - bash prompt: faster untracked status indicator with untracked directories
79 - - bash prompt: test untracked files status indicator with untracked dirs
86 + More transition from "unsigned char[40]" to "struct object_id".
87
81 - Will merge to 'next'.
88 + While GSoC and other topics are actively moving existing code
89 + around, this cannot go in; ejected from 'pu'.
90
83 ---------------------------------------------------
84 -[Graduated to "master"]
91
86 -* jc/diff-ws-error-highlight (2015-07-12) 1 commit
87 - (merged to 'next' on 2015-07-12 at 15b60ce)
88 - + diff: parse ws-error-highlight option more strictly
92 +* wp/sha1-name-negative-match (2015-06-08) 2 commits
93 + - sha1_name.c: introduce '^{/!-<negative pattern>}' notation
94 + - test for '!' handling in rev-parse's named commits
95
90 - A hotfix to a new feature in 2.5.0-rc.
96 + Introduce "branch^{/!-<pattern>}" notation to name a commit
97 + reachable from branch that does not match the given pattern.
98
99 + Expecting a reroll.
100
93 -* jk/still-interesting (2015-06-29) 1 commit
94 - (merged to 'next' on 2015-07-09 at e19fc0e)
95 - + revision.c: remove unneeded check for NULL
101
97 - Code clean-up.
102 +* mk/utf8-no-iconv-warn (2015-06-08) 1 commit
103 + - utf8.c: print warning about disabled iconv
104
99 - Will merge to 'master'.
105 + Warn when a reencoding is requested in a build without iconv
106 + support, as the end user is likely to get an unexpected result. I
107 + think the same level of safety should be added to a build with
108 + iconv support when the specified encoding is not available, but the
109 + patch does not go there.
110 +
111 + Expecting a reroll.
112 +
113 +
114 +* ak/format-patch-odir-config (2015-06-19) 1 commit
115 + - format-patch: introduce format.outputDirectory configuration
116 +
117 + Reroll exists but didn't pick it up as it seemed to be still
118 + collecting review comments.
119 +
120 + Expecting a reroll.
121 + ($gmane/272180).
122 +
123 +
124 +* mh/tempfile (2015-06-10) 14 commits
125 + - credential-cache--daemon: use tempfile module
126 + - credential-cache--daemon: delete socket from main()
127 + - gc: use tempfile module to handle gc.pid file
128 + - lock_repo_for_gc(): compute the path to "gc.pid" only once
129 + - diff: use tempfile module
130 + - setup_temporary_shallow(): use tempfile module
131 + - write_shared_index(): use tempfile module
132 + - register_tempfile(): new function to handle an existing temporary file
133 + - tempfile: add several functions for creating temporary files
134 + - register_tempfile_object(): new function, extracted from create_tempfile()
135 + - commit_lock_file(): use get_locked_file_path()
136 + - lockfile: remove some redundant functions
137 + - tempfile: a new module for handling temporary files
138 + - Move lockfile API documentation to lockfile.h
139 +
140 + Rebuild "lockfile" API on top of a new "tempfile" API.
141 + This needs rerolling, to include "tempfile.h" in "lockfile.h", at
142 + least.
143 +
144 + Expecting a reroll.
145 + ($gmane/271353)
146 +
147 +
148 +* ad/bisect-terms (2015-06-29) 4 commits
149 + - bisect: allow setting any user-specified in 'git bisect start'
150 + - bisect: add 'git bisect terms' to view the current terms
151 + - bisect: add the terms old/new
152 + - bisect: sanity check on terms
153 + (this branch uses ad/bisect-cleanup.)
154 +
155 + The use of 'good/bad' in "git bisect" made it confusing to use when
156 + hunting for a state change that is not a regression (e.g. bugfix).
157 + The command learned 'old/new' and then allows the end user to
158 + say e.g. "bisect start --term-old=fast --term=new=slow" to find a
159 + performance regression.
160 +
161 + Michael's idea to make 'good/bad' more intelligent does have
162 + certain attractiveness ($gname/272867), and makes some of the work
163 + on this topic a moot point.
164 +
165 + Will hold.
166 +
167 +
168 +* mh/notes-allow-reading-treeish (2015-07-13) 1 commit
169 + - notes: allow treeish expressions as notes ref
170 +
171 + Some "git notes" operations, e.g. "git log --notes=<note>", should
172 + be able to read notes from any tree-ish that is shaped like a notes
173 + tree, but the notes infrastructure required that the argument must
174 + be a ref under refs/notes/. Loosen it to require a valid ref only
175 + when the operation would update the notes (in which case we must
176 + have a place to store the updated notes tree, iow, a ref).
177 +
178 + Needs update to docs.
179
101 ---------------------------------------------------
102 -[Stalled]
180
181 * sg/config-name-only (2015-05-28) 3 commits
182 - completion: use new 'git config' options to reliably list variable names
@@ -165,19 +242,6 @@ of the repositories listed at
242 ($gmane/268058).
243
244
168 -* tf/gitweb-project-listing (2015-03-19) 5 commits
169 - - gitweb: make category headings into links when they are directories
170 - - gitweb: optionally set project category from its pathname
171 - - gitweb: add a link under the search box to clear a project filter
172 - - gitweb: if the PATH_INFO is incomplete, use it as a project_filter
173 - - gitweb: fix typo in man page
174 -
175 - Update gitweb to make it more pleasant to deal with a hierarchical
176 - forest of repositories.
177 -
178 - Any comments from those who use or have their own code in Gitweb?
179 -
180 -
245 * jc/a-lone-dash-stands-for-previous-branch (2015-03-16) 1 commit
246 - "-" and "@{-1}" on various programs
247
@@ -293,6 +357,64 @@ of the repositories listed at
357 --------------------------------------------------
358 [Cooking]
359
360 +* zb/userdiff-fountain (2015-07-21) 2 commits
361 + - SQUASH???
362 + - userdiff: add support for Fountain documents
363 +
364 + New userdiff pattern definition for fountain screenwriting markup
365 + format.
366 +
367 + Waiting for a response ($gmane/274403).
368 +
369 +
370 +* as/sparse-checkout-removal (2015-07-21) 1 commit
371 + (merged to 'next' on 2015-07-21 at ab94680)
372 + + unpack-trees: don't update files with CE_WT_REMOVE set
373 +
374 + "sparse checkout" misbehaved for a path that is excluded from the
375 + checkout when switching between branches that differ at the path.
376 +
377 + Will merge to 'master'.
378 +
379 +
380 +* cb/uname-in-untracked (2015-07-17) 1 commit
381 + (merged to 'next' on 2015-07-21 at d867af0)
382 + + untracked: fix detection of uname(2) failure
383 +
384 + An experimental "untracked cache" feature used uname(2) in a
385 + slightly unportable way.
386 +
387 + Will merge to 'master'.
388 +
389 +
390 +* bw/portability-solaris (2015-07-20) 3 commits
391 + - tests: fix sed usage in tests to work around broken xpg4/sed on Solaris
392 + - tests: fix sed usage in tests to work around broken xpg4/sed on Solaris
393 + - tests: modify tr expressions so that xpg4/tr handles it on Solaris
394 +
395 + Needs another reroll? ($gmane/274296)
396 +
397 +
398 +* es/worktree-add (2015-07-20) 5 commits
399 + (merged to 'next' on 2015-07-20 at 76e840b)
400 + + config: rename "gc.pruneWorktreesExpire" to "gc.worktreePruneExpire"
401 + + Documentation/git-worktree: wordsmith worktree-related manpages
402 + + Documentation/config: fix stale "git prune --worktree" reference
403 + + Documentation/git-worktree: fix incorrect reference to file "locked"
404 + + Documentation/git-worktree: consistently use term "linked working tree"
405 + (this branch is used by es/worktree-add-cleanup.)
406 +
407 + Will merge to 'master'.
408 +
409 +
410 +* sg/bash-prompt-untracked-optim (2015-07-20) 2 commits
411 + (merged to 'next' on 2015-07-21 at e151fb7)
412 + + bash prompt: faster untracked status indicator with untracked directories
413 + + bash prompt: test untracked files status indicator with untracked dirs
414 +
415 + Will merge to 'master'.
416 +
417 +
418 * es/worktree-add-cleanup (2015-07-20) 22 commits
419 - checkout: drop intimate knowledge of newly created worktree
420 - worktree: populate via "git reset --hard" rather than "git checkout"
@@ -353,19 +475,6 @@ of the repositories listed at
475 Will merge to 'master'.
476
477
356 -* mh/notes-allow-reading-treeish (2015-07-13) 1 commit
357 - - notes: allow treeish expressions as notes ref
358 -
359 - Some "git notes" operations, e.g. "git log --notes=<note>", should
360 - be able to read notes from any tree-ish that is shaped like a notes
361 - tree, but the notes infrastructure required that the argument must
362 - be a ref under refs/notes/. Loosen it to require a valid ref only
363 - when the operation would update the notes (in which case we must
364 - have a place to store the updated notes tree, iow, a ref).
365 -
366 - Needs update to docs.
367 -
368 -
478 * dt/log-follow-config (2015-07-09) 1 commit
479 (merged to 'next' on 2015-07-10 at b8fbb43)
480 + log: add "log.follow" configuration variable
@@ -384,7 +493,8 @@ of the repositories listed at
493
494
495 * se/doc-checkout-ours-theirs (2015-07-12) 1 commit
387 - - checkout: document subtlety around --ours/--theirs
496 + (merged to 'next' on 2015-07-21 at 5275f04)
497 + + checkout: document subtlety around --ours/--theirs
498
499 A "rebase" replays changes of the local branch on top of something
500 else, as such they are placed in stage #3 and referred to as
@@ -392,22 +502,25 @@ of the repositories listed at
502 work, are placed in stage #2 and referred to as "ours". Clarify
503 the "checkout --ours/--theirs".
504
505 + Will merge to 'master'.
506 +
507
508 * pt/am-tests (2015-07-20) 12 commits
397 - - t3901: test git-am encoding conversion
398 - - t3418: non-interactive rebase --continue with rerere enabled
399 - - t4150: tests for am --[no-]scissors
400 - - t4150: am with post-applypatch hook
401 - - t4150: am with pre-applypatch hook
402 - - t4150: am with applypatch-msg hook
403 - - t4150: am --resolved fails if index has unmerged entries
404 - - t4150: am --resolved fails if index has no changes
405 - - t4150: am refuses patches when paused
406 - - t4151: am --abort will keep dirty index intact
407 - - t4150: am fails if index is dirty
408 - - t4150: am.messageid really adds the message id
509 + (merged to 'next' on 2015-07-21 at 8d3d5f4)
510 + + t3901: test git-am encoding conversion
511 + + t3418: non-interactive rebase --continue with rerere enabled
512 + + t4150: tests for am --[no-]scissors
513 + + t4150: am with post-applypatch hook
514 + + t4150: am with pre-applypatch hook
515 + + t4150: am with applypatch-msg hook
516 + + t4150: am --resolved fails if index has unmerged entries
517 + + t4150: am --resolved fails if index has no changes
518 + + t4150: am refuses patches when paused
519 + + t4151: am --abort will keep dirty index intact
520 + + t4150: am fails if index is dirty
521 + + t4150: am.messageid really adds the message id
522
410 - Will merge to 'next'.
523 + Will merge to 'master'.
524
525
526 * kn/for-each-tag-branch (2015-07-13) 11 commits
@@ -484,6 +597,8 @@ of the repositories listed at
597
598 Rewrite "am" in "C".
599
600 + Will merge to 'next'.
601 +
602
603 * ad/bisect-cleanup (2015-06-29) 6 commits
604 (merged to 'next' on 2015-07-09 at 75e2a06)
@@ -514,13 +629,13 @@ of the repositories listed at
629 Will merge to 'master'.
630
631
517 -* dt/refs-backend-preamble (2015-07-13) 7 commits
632 +* dt/refs-backend-preamble (2015-07-21) 7 commits
633 - git-stash: use update-ref --create-reflog instead of creating files
634 - update-ref and tag: add --create-reflog arg
635 - refs: add REF_FORCE_CREATE_REFLOG flag
636 - git-reflog: add exists command
637 - refs: new public ref function: safe_create_reflog
523 - - refs: Break out check for reflog autocreation
638 + - refs: break out check for reflog autocreation
639 - refs.c: add err arguments to reflog functions
640
641 In preparation for allowing different "backends" to store the refs
@@ -529,12 +644,12 @@ of the repositories listed at
644 direct filesystem access to ref-like things like CHERRY_PICK_HEAD
645 from scripts and programs.
646
532 - Still under discussion.
533 -
534 - Will hold.
647 + Will merge to 'next'.
648
649
537 -* jk/date-mode-format (2015-06-29) 3 commits
650 +* jk/date-mode-format (2015-07-20) 4 commits
651 + (merged to 'next' on 2015-07-21 at 69f66be)
652 + + strbuf: make strbuf_addftime more robust
653 (merged to 'next' on 2015-07-09 at a01bfc3)
654 + introduce "format" date-mode
655 + convert "enum date_mode" into a struct
@@ -580,48 +695,15 @@ of the repositories listed at
695 Code clean-up and minor fixes (so far).
696
697
583 -* ad/bisect-terms (2015-06-29) 4 commits
584 - - bisect: allow setting any user-specified in 'git bisect start'
585 - - bisect: add 'git bisect terms' to view the current terms
586 - - bisect: add the terms old/new
587 - - bisect: sanity check on terms
588 - (this branch uses ad/bisect-cleanup.)
589 -
590 - The use of 'good/bad' in "git bisect" made it confusing to use when
591 - hunting for a state change that is not a regression (e.g. bugfix).
592 - The command learned 'old/new' and then allows the end user to
593 - say e.g. "bisect start --term-old=fast --term=new=slow" to find a
594 - performance regression.
595 -
596 - Michael's idea to make 'good/bad' more intelligent does have
597 - certain attractiveness ($gname/272867), and makes some of the work
598 - on this topic a moot point.
599 -
600 - Will hold.
601 -
602 -
603 -* mh/tempfile (2015-06-10) 14 commits
604 - - credential-cache--daemon: use tempfile module
605 - - credential-cache--daemon: delete socket from main()
606 - - gc: use tempfile module to handle gc.pid file
607 - - lock_repo_for_gc(): compute the path to "gc.pid" only once
608 - - diff: use tempfile module
609 - - setup_temporary_shallow(): use tempfile module
610 - - write_shared_index(): use tempfile module
611 - - register_tempfile(): new function to handle an existing temporary file
612 - - tempfile: add several functions for creating temporary files
613 - - register_tempfile_object(): new function, extracted from create_tempfile()
614 - - commit_lock_file(): use get_locked_file_path()
615 - - lockfile: remove some redundant functions
616 - - tempfile: a new module for handling temporary files
617 - - Move lockfile API documentation to lockfile.h
618 -
619 - Rebuild "lockfile" API on top of a new "tempfile" API.
620 - This needs rerolling, to include "tempfile.h" in "lockfile.h", at
621 - least.
698 +* jc/rerere-multi (2015-07-20) 5 commits
699 + - t4200: rerere a merge with two identical conflicts
700 + - rerere: delay the recording of preimage for the first time
701 + - rerere: handle leftover rr-cache/$ID directory and postimage files
702 + - rerere: scan $GIT_DIR/rr-cache/$ID when instantiating a rerere_id
703 + - rerere: split conflict ID further
704 + (this branch uses jc/rerere.)
705
623 - Expecting a reroll.
624 - ($gmane/271353)
706 + This is a contination of jc/rerere topic.
707
708
709 * js/fsck-opt (2015-06-23) 19 commits
@@ -653,16 +735,6 @@ of the repositories listed at
735 Will merge to 'master'.
736
737
656 -* ak/format-patch-odir-config (2015-06-19) 1 commit
657 - - format-patch: introduce format.outputDirectory configuration
658 -
659 - Reroll exists but didn't pick it up as it seems to be still
660 - collecting review comments.
661 -
662 - Expecting a reroll.
663 - ($gmane/272180).
664 -
665 -
738 * bc/gpg-verify-raw (2015-06-22) 7 commits
739 (merged to 'next' on 2015-06-24 at 08a1164)
740 + verify-tag: add option to print raw gpg status information
@@ -794,18 +866,6 @@ of the repositories listed at
866 Will merge to 'master'.
867
868
797 -* mk/utf8-no-iconv-warn (2015-06-08) 1 commit
798 - - utf8.c: print warning about disabled iconv
799 -
800 - Warn when a reencoding is requested in a build without iconv
801 - support, as the end user is likely to get an unexpected result. I
802 - think the same level of safety should be added to a build with
803 - iconv support when the specified encoding is not available, but the
804 - patch does not go there.
805 -
806 - Expecting a reroll.
807 -
808 -
869 * mr/rebase-i-customize-insn-sheet (2015-06-15) 1 commit
870 (merged to 'next' on 2015-07-10 at ec383cd)
871 + git-rebase--interactive.sh: add config option for custom instruction format
@@ -880,43 +940,6 @@ of the repositories listed at
940 Will merge to 'master'.
941
942
883 -* wp/sha1-name-negative-match (2015-06-08) 2 commits
884 - - sha1_name.c: introduce '^{/!-<negative pattern>}' notation
885 - - test for '!' handling in rev-parse's named commits
886 -
887 - Introduce "branch^{/!-<pattern>}" notation to name a commit
888 - reachable from branch that does not match the given pattern.
889 -
890 - Expecting a reroll.
891 -
892 -
893 -* bc/object-id (2015-06-17) 10 commits
894 - . remote.c: use struct object_id in many functions
895 - . object-id: use struct object_id in struct object
896 - . remote.c: use struct object_id in ref_newer()
897 - . transport-helper.c: use struct object_id in push_refs_with_export()
898 - . connect.c: use struct object_id in get_remote_heads()
899 - . remote-curl: use struct object_id in parse_fetch()
900 - . fetch-pack: use struct object_id in add_sought_entry_mem()
901 - . object_id: convert struct ref to use object_id.
902 - . sha1_file: introduce has_object_file() helper
903 - . refs: convert some internal functions to use object_id
904 -
905 - More transition from "unsigned char[40]" to "struct object_id".
906 -
907 - While GSoC and other topics are actively moving existing code
908 - around, this cannot go in; ejected from 'pu'.
909 -
910 -
911 -* jk/log-missing-default-HEAD (2015-06-03) 1 commit
912 - - log: diagnose empty HEAD more clearly
913 -
914 - "git init empty && git -C empty log" said "bad default revision 'HEAD'",
915 - which was found to be a bit confusing to new users.
916 -
917 - What's the status of this one?
918 -
919 -
943 * gr/rebase-i-drop-warn (2015-06-30) 3 commits
944 (merged to 'next' on 2015-07-15 at 6f4f2c8)
945 + git rebase -i: add static check for commands and SHA-1
@@ -931,37 +954,6 @@ of the repositories listed at
954 Will merge to 'master'.
955
956
934 -* jh/strbuf-read-use-read-in-full (2015-06-01) 1 commit
935 - - strbuf_read(): skip unnecessary strbuf_grow() at eof
936 -
937 - Avoid one extra iteration and strbuf_grow() of 8kB in
938 - strbuf_read().
939 -
940 - Looked reasonable; perhaps a log message clarification is needed.
941 -
942 - Expecting a reroll.
943 -
944 -
945 -* mg/index-read-error-messages (2015-06-01) 2 commits
946 - - messages: uniform error messages for index write
947 - - show-index: uniform error messages for index read
948 -
949 - The tip was RFC.
950 - Expecting a reroll.
951 -
952 -
953 -* hv/submodule-config (2015-06-15) 4 commits
954 - - do not die on error of parsing fetchrecursesubmodules option
955 - - use new config API for worktree configurations of submodules
956 - - extract functions for submodule config set and lookup
957 - - implement submodule config API for lookup of .gitmodules values
958 -
959 - The gitmodules API accessed from the C code learned to cache stuff
960 - lazily.
961 -
962 - Needs another reroll? ($gmane/273743).
963 -
964 -
957 * jc/push-tags-also (2015-05-29) 1 commit
958 - push --tags: push tags *in addition to* other stuff
959
@@ -984,13 +976,6 @@ of the repositories listed at
976 Will merge to 'master'.
977
978
987 -* jc/clone-bundle (2015-04-30) 1 commit
988 - - repack: optionally create a clone.bundle
989 -
990 - Waiting for further work.
991 - Still an early WIP.
992 -
993 -
979 * ee/clean-remove-dirs (2015-06-26) 6 commits
980 (merged to 'next' on 2015-06-29 at d595659)
981 + read_gitfile_gently: fix use-after-free
@@ -1016,3 +1001,18 @@ of the repositories listed at
1001 has been deprecated since October 2007.
1002
1003 Will keep in 'next' during the 2.5 cycle.
1004 +
1005 +--------------------------------------------------
1006 +[Discarded]
1007 +
1008 +* tf/gitweb-project-listing (2015-03-19) 4 commits
1009 + - gitweb: make category headings into links when they are directories
1010 + - gitweb: optionally set project category from its pathname
1011 + - gitweb: add a link under the search box to clear a project filter
1012 + - gitweb: if the PATH_INFO is incomplete, use it as a project_filter
1013 +
1014 + Update gitweb to make it more pleasant to deal with a hierarchical
1015 + forest of repositories.
1016 +
1017 + A fresh restart will not be rejected, but ejected from my tree for
1018 + now.