What's cooking (2013/07 #09)
Junio C Hamano committed
Jul 29, 2013 at 12:20 UTC
33f854ca9eb8251f5b7fa1c670d4abcdd76764ce
1 file changed
+286
-386
whats-cooking.txt
+286
-386
@@ -1,27 +1,20 @@
1
To: git@vger.kernel.org
2
Bcc: lwn@lwn.net
3
-Subject: What's cooking in git.git (Jul 2013, #08; Mon, 22)
4
-X-master-at: 0bde8c0c1e53e2b2001f4ced58d3e66865758cea
5
-X-next-at: c240c481b4c5d7c300800bb80174c0853d587c6d
3
+Subject: What's cooking in git.git (Jul 2013, #09; Mon, 29)
4
+X-master-at: 8f6a3e5d71fa1800c3f330bab2b6b6399ec7aea7
5
+X-next-at: cf3f07e67f28b0068199b16bc6974f4bbafb4e51
6
7
-What's cooking in git.git (Jul 2013, #08; Mon, 22)
7
+What's cooking in git.git (Jul 2013, #09; Mon, 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
-This again is one of those back-to-back issues of "What's cooking"
15
-report, but I expect that the rest of the week will be slower than
16
-usual due to it being the OSCON week, and I wanted to categorize the
17
-topics in flight into those that should be in -rc1 and those that
18
-should cook until the next cycle. Hopefully -rc0 will be tagged
19
-tomorrow with some of the former, but not all, as we would like to
20
-cook topics that are not trivial in 'next' for at least a week or
21
-so. The ones marked as "will merge to master" below are marked as
22
-such because I do not expect major issues in them to be found before
23
-we tag -rc1 (see http://tinyurl.com/gitCal), while ones marked as
24
-"will cook in next" may not be so trivial.
14
+The shape of the upcoming release is pretty much known by now; all
15
+the topics that are marked for 'master' in this issue will likely to
16
+be in the final, and others will cook during the pre-release freeze
17
+period.
18
19
You can find the changes described here in the integration branches
20
of the repositories listed at
@@ -31,230 +24,248 @@ of the repositories listed at
24
--------------------------------------------------
25
[Graduated to "master"]
26
34
-* db/show-ref-head (2013-07-17) 1 commit
35
- (merged to 'next' on 2013-07-18 at 73797d0)
36
- + show-ref: make --head always show the HEAD ref
37
-
38
- The "--head" option to "git show-ref" was only to add "HEAD" to the
39
- list of candidate refs to be filtered by the usual rules
40
- (e.g. "--heads" that only show refs under refs/heads). Change the
41
- meaning of the option to always show "HEAD" regardless of what
42
- filtering will be applied to any other ref (this is a backward
43
- incompatible change, so I may need to add an entry to the Release
44
- Notes).
27
+* ob/typofixes (2013-07-22) 3 commits
28
+ (merged to 'next' on 2013-07-22 at 8574f9f)
29
+ + typofix: in-code comments
30
+ + typofix: documentation
31
+ + typofix: release notes
32
33
47
-* dw/request-pull-diag (2013-07-17) 1 commit
48
- (merged to 'next' on 2013-07-18 at d4406de)
49
- + request-pull: improve error message for invalid revision args
34
+* es/contacts (2013-07-21) 5 commits
35
+ (merged to 'next' on 2013-07-22 at a78c3d6)
36
+ + contrib: contacts: add documentation
37
+ + contrib: contacts: add mailmap support
38
+ + contrib: contacts: interpret committish akin to format-patch
39
+ + contrib: contacts: add ability to parse from committish
40
+ + contrib: add git-contacts helper
41
51
-
52
-* es/blame-L-breakage (2013-07-17) 6 commits
53
- (merged to 'next' on 2013-07-18 at cfd871a)
54
- + blame-options.txt: explain that -L <start> and <end> are optional
55
- + blame-options.txt: place each -L option variation on its own line
56
- + t8001/t8002 (blame): add blame -L :funcname tests
57
- + t8001/t8002 (blame): add blame -L tests
58
- + t8001/t8002 (blame): modernize style
59
- + line-range: fix "blame -L X,-N" regression
60
-
61
- The refactoring made for parsing "-L" option recently to support
62
- "git log -L" seems to have broken "git blame -L X,-5" to show 5
63
- lines leading to X.
64
-
65
-
66
-* es/check-mailmap (2013-07-13) 2 commits
67
- (merged to 'next' on 2013-07-21 at 23c5363)
68
- + t4203: test check-mailmap command invocation
69
- + builtin: add git-check-mailmap command
70
- (this branch is used by es/contacts.)
71
-
72
- A new command to allow scripts to query the mailmap information.
42
+ A helper to read from a set of format-patch output files or a range
43
+ of commits and find those who may have insights to the code that
44
+ the changes touch by running a series of "git blame" commands.
45
46
75
-* hv/config-from-blob (2013-07-12) 5 commits
76
- (merged to 'next' on 2013-07-15 at 9ab7f3f)
77
- + do not die when error in config parsing of buf occurs
78
- + teach config --blob option to parse config from database
79
- + config: make parsing stack struct independent from actual data source
80
- + config: drop cf validity check in get_next_char()
81
- + config: factor out config file stack management
47
+* es/line-log-further-fixes (2013-07-23) 5 commits
48
+ + line-log: fix "log -LN" crash when N is last line of file
49
+ + range-set: satisfy non-empty ranges invariant
50
+ + t4211: demonstrate crash when first -L encountered is empty range
51
+ + t4211: demonstrate empty -L range crash
52
+ + range-set: fix sort_and_merge_range_set() corner case bug
53
+ (this branch is used by tr/line-log.)
54
83
- Allow configuration data to be read from in-tree blob objects,
84
- which would help working in a bare repository and submodule
85
- updates.
55
56
+* jk/cat-file-batch-optim (2013-07-18) 9 commits
57
+ (merged to 'next' on 2013-07-22 at 965897c)
58
+ + Fix some sparse warnings
59
+ + sha1_object_info_extended: pass object_info to helpers
60
+ + sha1_object_info_extended: make type calculation optional
61
+ + packed_object_info: make type lookup optional
62
+ + packed_object_info: hoist delta type resolution to helper
63
+ + sha1_loose_object_info: make type lookup optional
64
+ + sha1_object_info_extended: rename "status" to "type"
65
+ + cat-file: disable object/refname ambiguity check for batch mode
66
+ + Merge branch 'nd/warn-ambiguous-object-name' into jk/cat-file-batch-optim
67
88
-* jc/mailmap-case-insensitivity (2013-07-15) 9 commits
89
- (merged to 'next' on 2013-07-18 at 83341ac)
90
- + mailmap: style fixes
91
- + mailmap: debug: avoid passing NULL to fprintf() '%s' conversion specification
92
- + mailmap: debug: eliminate -Wformat field precision type warning
93
- + mailmap: debug: fix malformed fprintf() format conversion specification
94
- + mailmap: debug: fix out-of-order fprintf() arguments
95
- + mailmap: do not downcase mailmap entries
96
- + t4203: demonstrate loss of uppercase characters in canonical email
97
- + mailmap: do not lose single-letter names
98
- + t4203: demonstrate loss of single-character name in mailmap entry
68
+ If somebody wants to only know on-disk footprint of an object
69
+ without having to know its type or payload size, we can bypass a
70
+ lot of code to cheaply learn it.
71
100
- The mailmap mechanism unnecessarily downcased the e-mail addresses
101
- in the output, and also ignored the human name when it is a single
102
- character name.
72
73
+* jm/doc-ref-prune (2013-07-18) 2 commits
74
+ (merged to 'next' on 2013-07-22 at 414e6ea)
75
+ + Documentation: fix git-prune example usage
76
+ + Documentation: remove --prune from pack-refs examples
77
105
-* jc/name-rev-exact-ref (2013-07-18) 5 commits
106
- (merged to 'next' on 2013-07-18 at c8bc00c)
107
- + describe: fix --contains when a tag is given as input
108
- + name-rev: differentiate between tags and commits they point at
109
- (merged to 'next' on 2013-07-11 at fd5b30c)
110
- + describe: use argv-array
111
- + name-rev: allow converting the exact object name at the tip of a ref
112
- + name-ref: factor out name shortening logic from name_ref()
78
114
- Corrects the longstanding sloppiness in the implementation of
115
- name-rev that conflated "we take commit-ish" and "differences
116
- between tags and commits do not matter".
79
+* mh/multimail (2013-07-22) 2 commits
80
+ (merged to 'next' on 2013-07-22 at e27c933)
81
+ + post-receive-email: deprecate script in favor of git-multimail
82
+ + git-multimail: an improved replacement for post-receive-email
83
84
+ An enhanced "post-receive" hook to send e-mail messages.
85
119
-* jc/simple-add-must-be-a-no-op (2013-07-19) 1 commit
120
- (merged to 'next' on 2013-07-19 at a1543ee)
121
- + t2202: make sure "git add" (no args) stays a no-op
86
123
- This detected a mismerge of one of "add-2.0" topics to the 'jch'
124
- and 'pu' branches.
87
+* mh/ref-races-optim-invalidate-cached (2013-06-20) 1 commit
88
+ (merged to 'next' on 2013-07-22 at 144d135)
89
+ + refs: do not invalidate the packed-refs cache unnecessarily
90
91
+ This requires the platform lstat() to be correct to avoid false
92
+ negatives.
93
127
-* jk/gcc-function-attributes (2013-07-19) 4 commits
128
- (merged to 'next' on 2013-07-19 at 5f6ad91)
129
- + Add the LAST_ARG_MUST_BE_NULL macro
130
- (merged to 'next' on 2013-07-12 at 5a7081f)
131
- + wt-status: use "format" function attribute for status_printf
132
- + use "sentinel" function attribute for variadic lists
133
- + add missing "format" function attributes
94
135
- Use the function attributes extension to catch mistakes in use of
136
- our own variadic functions that use NULL sentinel at the end
137
- (i.e. like execl(3)) and format strings (i.e. like printf(3)).
95
+* ml/avoid-using-grep-on-crlf-files (2013-07-18) 1 commit
96
+ (merged to 'next' on 2013-07-22 at f861472)
97
+ + test-lib.sh - define and use GREP_STRIPS_CR
98
99
+ On systems that understand a CRLF as a line ending, tests in this
100
+ script that worked on files with CRLF line endings using "grep" to
101
+ extract matching lines may lose the CR at the end of lines that
102
+ match, causing the actual output not to match the expected output.
103
140
-* jk/t0008-sigpipe-fix (2013-07-12) 1 commit
141
- (merged to 'next' on 2013-07-15 at cead4b2)
142
- + t0008: avoid SIGPIPE race condition on fifo
104
144
- Fix for recent test breakage on 'master'.
105
+* ml/cygwin-updates (2013-07-21) 4 commits
106
+ (merged to 'next' on 2013-07-22 at e9c9872)
107
+ + cygwin: stop forcing core.filemode=false
108
+ + Cygwin 1.7 supports mmap
109
+ + Cygwin 1.7 has thread-safe pread
110
+ + Cygwin 1.7 needs compat/regex
111
112
+ The tip one does _not_ revert c869753e (Force core.filemode to
113
+ false on Cygwin., 2006-12-30) on purpose, so that people can
114
+ still retain the old behaviour if they wanted to.
115
147
-* jx/clean-interactive (2013-06-26) 16 commits
148
- (merged to 'next' on 2013-07-12 at 61dd42d)
149
- + test: run testcases with POSIX absolute paths on Windows
150
- + test: add t7301 for git-clean--interactive
151
- + git-clean: add documentation for interactive git-clean
152
- + git-clean: add ask each interactive action
153
- + git-clean: add select by numbers interactive action
154
- + git-clean: add filter by pattern interactive action
155
- + git-clean: use a git-add-interactive compatible UI
156
- + git-clean: add colors to interactive git-clean
157
- + git-clean: show items of del_list in columns
158
- + git-clean: add support for -i/--interactive
159
- (merged to 'next' on 2013-06-26 at f7be2d8)
160
- + git-clean: refactor git-clean into two phases
161
- + write_name{_quoted_relative,}(): remove redundant parameters
162
- + quote_path_relative(): remove redundant parameter
163
- + quote.c: substitute path_relative with relative_path
164
- + path.c: refactor relative_path(), not only strip prefix
165
- + test: add test cases for relative_path
116
167
- Add "interactive" mode to "git clean".
117
+* rh/template-updates (2013-07-15) 3 commits
118
+ (merged to 'next' on 2013-07-22 at 53dffdd)
119
+ + templates: spell ASCII in uppercase in pre-commit hook
120
+ + templates: Reformat pre-commit hook's message
121
+ + templates: Use heredoc in pre-commit hook
122
123
+ This is an earlier part of a 6 patch series, with log message
124
+ corrected.
125
170
-* mk/upload-pack-off-by-one-dead-code-removal (2013-07-15) 1 commit
171
- (merged to 'next' on 2013-07-15 at a8f55a2)
172
- + upload-pack: remove a piece of dead code
126
127
+* rj/sparse (2013-07-21) 1 commit
128
+ (merged to 'next' on 2013-07-22 at 24efece)
129
+ + Revert "compat/unsetenv.c: Fix a sparse warning"
130
175
-* mm/color-auto-default (2013-07-19) 1 commit
176
- (merged to 'next' on 2013-07-19 at 880d984)
177
- + git add -e: Explicitly specify that patch should have no color
131
179
- A finishing touch to fix breakage to "add -e" caused by defaulting
180
- ui.color to "auto".
132
+* sb/misc-fixes (2013-07-15) 3 commits
133
+ (merged to 'next' on 2013-07-21 at 880b08c)
134
+ + diff.c: Do not initialize a variable, which gets reassigned anyway.
135
+ + commit: Fix a memory leak in determine_author_info
136
+ + daemon.c:handle: Remove unneeded check for null pointer.
137
138
+ Assorted code cleanups and a minor fix.
139
183
-* mm/diff-no-patch-synonym-to-s (2013-07-17) 6 commits
184
- (merged to 'next' on 2013-07-18 at 3b893d6)
185
- + Documentation/git-log.txt: capitalize section names
186
- + Documentation: move description of -s, --no-patch to diff-options.txt
187
- + Documentation/git-show.txt: include common diff options, like git-log.txt
188
- + diff: allow --patch & cie to override -s/--no-patch
189
- + diff: allow --no-patch as synonym for -s
190
- + t4000-diff-format.sh: modernize style
140
192
- "git show -s" was less discoverable than it should have been. It
193
- now has a more natural looking "git show --no-patch" synonym.
141
+* sb/traverse-trees-bitmask-variable-name (2013-07-19) 1 commit
142
+ (merged to 'next' on 2013-07-22 at be3227c)
143
+ + traverse_trees(): clarify return value of the callback
144
145
196
-* nd/const-struct-cache-entry (2013-07-09) 1 commit
197
- (merged to 'next' on 2013-07-15 at 9ce0122)
198
- + Convert "struct cache_entry *" to "const ..." wherever possible
146
+* tr/line-log (2013-07-23) 1 commit
147
+ (merged to 'next' on 2013-07-23 at a0a7f9b)
148
+ + Merge branch 'es/line-log-further-fixes' into tr/line-log
149
+ (this branch uses es/line-log-further-fixes.)
150
151
+--------------------------------------------------
152
+[New Topics]
153
201
-* rr/send-email-ssl-verify (2013-07-18) 1 commit
202
- (merged to 'next' on 2013-07-21 at 62099ff)
203
- + send-email: be explicit with SSL certificate verification
154
+* ob/typofixes (2013-07-29) 1 commit
155
+ - many small typofixes
156
205
- Newer Net::SMTP::SSL module does not want the user programs to use
206
- the default behaviour to let server certificate go without
207
- verification, so by default enable the verification with a
208
- mechanism to turn it off if needed.
157
+ Will merge to 'next'.
158
159
211
-* sb/parse-object-buffer-eaten (2013-07-17) 1 commit
212
- (merged to 'next' on 2013-07-18 at 364689c)
213
- + parse_object_buffer: correct freeing the buffer
160
+* jh/remote-hg-fetch-fix (2013-07-25) 2 commits
161
+ (merged to 'next' on 2013-07-25 at 33161ad)
162
+ + Revert "remotes-hg: bugfix for fetching non local remotes"
163
+ (merged to 'next' on 2013-07-24 at 9c96641)
164
+ + remotes-hg: bugfix for fetching non local remotes
165
166
+ Waiting for the final patch after discussion settles.
167
216
-* tr/do-not-call-submodules-subprojects (2013-07-18) 2 commits
217
- (merged to 'next' on 2013-07-18 at 6386fd5)
218
- + show-branch: fix description of --date-order
219
- + apply, entry: speak of submodules instead of subprojects
168
169
+* ma/hg-to-git (2013-07-23) 1 commit
170
+ - hg-to-git: --allow-empty-message in git commit
171
222
-* tr/fd-gotcha-fixes (2013-07-12) 2 commits
223
- (merged to 'next' on 2013-07-18 at 30f4fdf)
224
- + run-command: dup_devnull(): guard against syscalls failing
225
- + git_mkstemps: correctly test return value of open()
172
+ Will merge to 'next'.
173
227
- Two places we did not check return value (expected to be a file
228
- descriptor) correctly.
174
175
+* nd/clone-connectivity-shortcut (2013-07-23) 1 commit
176
+ - smart http: use the same connectivity check on cloning
177
231
-* tr/protect-low-3-fds (2013-07-17) 2 commits
232
- (merged to 'next' on 2013-07-18 at 504d972)
233
- + git: ensure 0/1/2 are open in main()
234
- + daemon/shell: refactor redirection of 0/1/2 from /dev/null
178
+ Will merge to 'next'.
179
236
- When "git" is spawned in such a way that any of the low 3 file
237
- descriptors is closed, our first open() may yield file descriptor 2,
238
- and writing error message to it would screw things up in a big way.
180
240
---------------------------------------------------
241
-[New Topics]
181
+* bc/completion-for-bash-3.0 (2013-07-24) 1 commit
182
+ (merged to 'next' on 2013-07-24 at 913d89c)
183
+ + git-completion.bash: replace zsh notation that breaks bash 3.X
184
243
-* rj/doc-rev-parse (2013-07-22) 2 commits
244
- (merged to 'next' on 2013-07-22 at 8188667)
245
- + rev-parse(1): logically group options
246
- + rev-parse: remove restrictions on some options
185
+ Will merge to 'master'.
186
248
- Will cook in 'next'.
187
188
+* jk/capabilities-doc (2013-07-26) 5 commits
189
+ (merged to 'next' on 2013-07-29 at cf3f07e)
190
+ + document 'allow-tip-sha1-in-want' capability
191
+ (merged to 'next' on 2013-07-24 at 5af9a33)
192
+ + document 'quiet' receive-pack capability
193
+ + document 'agent' protocol capability
194
+ + docs: note that receive-pack knows side-band-64k capability
195
+ + docs: fix 'report-status' protocol capability thinko
196
251
-* ob/typofixes (2013-07-22) 3 commits
252
- - typofix: in-code comments
253
- - typofix: documentation
254
- - typofix: release notes
197
+ Will merge to 'master'.
198
+
199
+
200
+* jx/clean-interactive (2013-07-24) 2 commits
201
+ - git-clean: implement partial matching for selection
202
+ (merged to 'next' on 2013-07-24 at 1e63bb9)
203
+ + Documentation/git-clean: fix description for range
204
+
205
+
206
+* sb/mailmap-updates (2013-07-24) 1 commit
207
+ (merged to 'next' on 2013-07-24 at 9f8e681)
208
+ + .mailmap: combine more (email, name) to individual persons
209
+
210
+ Will merge to 'master'.
211
+
212
+
213
+* tr/log-full-diff-keep-true-parents (2013-07-29) 2 commits
214
+ - SQUASH??? free saved-parents slab when done
215
+ - [PERHAPS LIKE THIS] log: use true parents for diff even when rewriting
216
+
217
+
218
+* db/http-savecookies (2013-07-24) 1 commit
219
+ - http: Add http.savecookies option to write out HTTP cookies
220
+
221
+
222
+* ds/doc-two-kinds-of-tags (2013-07-29) 1 commit
223
+ - docs/git-tag: explain lightweight versus annotated tags
224
225
Will merge to 'next' and then to 'master'.
226
227
+
228
+* jk/commit-how-to-abort-cherry-pick (2013-07-29) 1 commit
229
+ - commit: tweak empty cherry pick advice for sequencer
230
+
231
+ Will merge to 'next' and then to 'master'.
232
+
233
+
234
+* lf/echo-n-is-not-portable (2013-07-29) 1 commit
235
+ - Avoid using `echo -n` anywhere
236
+
237
+ Will merge to 'next'.
238
+
239
+
240
+* rj/commit-slab-fix (2013-07-29) 1 commit
241
+ - commit-slab.h: Fix memory allocation and addressing
242
+
243
+ Will merge to 'next' and then to 'master'.
244
+
245
+
246
+* rr/maint-tilde-markup-in-doc (2013-07-26) 1 commit
247
+ - config doc: quote paths, fixing tilde-interpretation
248
+
249
+ Will merge to 'next' and then to 'master'.
250
+
251
+
252
+* rr/rebase-autostash (2013-07-29) 1 commit
253
+ - git-rebase: fix typo
254
+
255
+ Will merge to 'next' and then to 'master'.
256
+
257
+
258
+* sb/fsck-opt-bool (2013-07-29) 1 commit
259
+ - fsck: Replace deprecated OPT_BOOLEAN by OPT_BOOL
260
+
261
+ Will merge to 'next'.
262
+
263
+
264
+* rr/feed-real-path-to-editor (2013-07-29) 1 commit
265
+ - editor: use canonicalized absolute path
266
+
267
+ Will merge to 'next'.
268
+
269
--------------------------------------------------
270
[Stalled]
271
@@ -377,47 +388,22 @@ of the repositories listed at
388
--------------------------------------------------
389
[Cooking]
390
380
-* jk/fast-import-empty-ls (2013-06-23) 4 commits
381
- - fast-import: allow moving the root tree
382
- - fast-import: allow ls or filecopy of the root tree
383
- - fast-import: set valid mode on root tree in "ls"
384
- - t9300: document fast-import empty path issues
385
-
386
- Will merge to 'next'.
387
-
388
-
389
-* jm/doc-ref-prune (2013-07-18) 2 commits
390
- (merged to 'next' on 2013-07-22 at 414e6ea)
391
- + Documentation: fix git-prune example usage
392
- + Documentation: remove --prune from pack-refs examples
393
-
394
- Will merge to 'master'.
395
-
396
-
397
-* rj/sparse (2013-07-21) 1 commit
398
- (merged to 'next' on 2013-07-22 at 24efece)
399
- + Revert "compat/unsetenv.c: Fix a sparse warning"
400
-
401
- Will merge to 'master'.
402
-
403
-
404
-* sb/traverse-trees-bitmask-variable-name (2013-07-19) 1 commit
405
- (merged to 'next' on 2013-07-22 at be3227c)
406
- + traverse_trees(): clarify return value of the callback
407
-
408
- Will merge to 'master'.
391
+* rj/doc-rev-parse (2013-07-22) 2 commits
392
+ (merged to 'next' on 2013-07-22 at 8188667)
393
+ + rev-parse(1): logically group options
394
+ + rev-parse: remove restrictions on some options
395
396
+ Will cook in 'next'.
397
411
-* mh/multimail (2013-07-15) 2 commits
412
- - post-receive-email: deprecate script in favor of git-multimail
413
- - git-multimail: an improved replacement for post-receive-email
398
415
- An enhanced "post-receive" hook to send e-mail messages.
399
+* jk/fast-import-empty-ls (2013-06-23) 4 commits
400
+ (merged to 'next' on 2013-07-22 at d26dab6)
401
+ + fast-import: allow moving the root tree
402
+ + fast-import: allow ls or filecopy of the root tree
403
+ + fast-import: set valid mode on root tree in "ls"
404
+ + t9300: document fast-import empty path issues
405
417
- I tend to agree that the wording of the tip one is a bit too
418
- strong. An old script does not necessarily have to be avoided by
419
- its existing users only because it has not got updates for a long
420
- time.
406
+ Will cook in 'next'.
407
408
409
* jc/diff-filter-negation (2013-07-19) 6 commits
@@ -442,112 +428,69 @@ of the repositories listed at
428
* mh/packed-refs-do-one-ref-recursion (2013-07-17) 1 commit
429
- do_one_ref(): save and restore value of current_ref
430
445
- Needs a bit better description to be placed in the Release Notes.
446
-
447
-
448
-* ml/avoid-using-grep-on-crlf-files (2013-07-18) 1 commit
449
- (merged to 'next' on 2013-07-22 at f861472)
450
- + test-lib.sh - define and use GREP_STRIPS_CR
451
-
452
- On systems that understand a CRLF as a line ending, tests in this
453
- script that worked on files with CRLF line endings using "grep" to
454
- extract matching lines may lose the CR at the end of lines that
455
- match, causing the actual output not to match the expected output.
456
-
457
- Will merge to 'master'.
458
-
459
-
460
-* ml/cygwin-updates (2013-07-21) 4 commits
461
- (merged to 'next' on 2013-07-22 at e9c9872)
462
- + cygwin: stop forcing core.filemode=false
463
- + Cygwin 1.7 supports mmap
464
- + Cygwin 1.7 has thread-safe pread
465
- + Cygwin 1.7 needs compat/regex
466
-
467
- The tip one does _not_ revert c869753e (Force core.filemode to
468
- false on Cygwin., 2006-12-30) on purpose, so that people can
469
- still retain the old behaviour if they wanted to.
431
+ Fix a NULL-pointer dereference during nested iterations over
432
+ references (for example, when replace references are being used).
433
471
- Will merge to 'master'.
434
+ Will merge to 'next' and then to 'master'.
435
436
474
-* nd/magic-pathspec (2013-07-17) 47 commits
475
- - t6131 - skip tests if on case-insensitive file system
476
- - parse_pathspec: accept :(icase)path syntax
477
- - pathspec: support :(glob) syntax
478
- - pathspec: make --literal-pathspecs disable pathspec magic
479
- - pathspec: support :(literal) syntax for noglob pathspec
480
- - kill limit_pathspec_to_literal() as it's only used by parse_pathspec()
481
- - parse_pathspec: preserve prefix length via PATHSPEC_PREFIX_ORIGIN
482
- - parse_pathspec: make sure the prefix part is wildcard-free
483
- - rename field "raw" to "_raw" in struct pathspec
484
- - tree-diff: remove the use of pathspec's raw[] in follow-rename codepath
485
- - remove match_pathspec() in favor of match_pathspec_depth()
486
- - remove init_pathspec() in favor of parse_pathspec()
487
- - remove diff_tree_{setup,release}_paths
488
- - convert common_prefix() to use struct pathspec
489
- - convert add_files_to_cache to take struct pathspec
490
- - convert {read,fill}_directory to take struct pathspec
491
- - convert refresh_index to take struct pathspec
492
- - convert report_path_error to take struct pathspec
493
- - checkout: convert read_tree_some to take struct pathspec
494
- - convert unmerge_cache to take struct pathspec
495
- - convert run_add_interactive to use struct pathspec
496
- - convert read_cache_preload() to take struct pathspec
497
- - line-log: convert to use parse_pathspec
498
- - reset: convert to use parse_pathspec
499
- - add: convert to use parse_pathspec
500
- - check-ignore: convert to use parse_pathspec
501
- - archive: convert to use parse_pathspec
502
- - ls-files: convert to use parse_pathspec
503
- - rm: convert to use parse_pathspec
504
- - checkout: convert to use parse_pathspec
505
- - rerere: convert to use parse_pathspec
506
- - status: convert to use parse_pathspec
507
- - commit: convert to use parse_pathspec
508
- - clean: convert to use parse_pathspec
509
- - guard against new pathspec magic in pathspec matching code
510
- - parse_pathspec: support prefixing original patterns
511
- - parse_pathspec: support stripping/checking submodule paths
512
- - parse_pathspec: support stripping submodule trailing slashes
513
- - parse_pathspec: add special flag for max_depth feature
514
- - convert some get_pathspec() calls to parse_pathspec()
515
- - parse_pathspec: add PATHSPEC_PREFER_{CWD,FULL} flags
516
- - parse_pathspec: save original pathspec for reporting
517
- - add parse_pathspec() that converts cmdline args to struct pathspec
518
- - pathspec: add copy_pathspec
519
- - pathspec: i18n-ize error strings in pathspec parsing code
520
- - move struct pathspec and related functions to pathspec.[ch]
521
- - clean: remove unused variable "seen"
437
+* nd/magic-pathspec (2013-07-23) 48 commits
438
+ (merged to 'next' on 2013-07-23 at 784c3a1)
439
+ + rm: do not set a variable twice without intermediate reading.
440
+ (merged to 'next' on 2013-07-22 at 39f7194)
441
+ + t6131 - skip tests if on case-insensitive file system
442
+ + parse_pathspec: accept :(icase)path syntax
443
+ + pathspec: support :(glob) syntax
444
+ + pathspec: make --literal-pathspecs disable pathspec magic
445
+ + pathspec: support :(literal) syntax for noglob pathspec
446
+ + kill limit_pathspec_to_literal() as it's only used by parse_pathspec()
447
+ + parse_pathspec: preserve prefix length via PATHSPEC_PREFIX_ORIGIN
448
+ + parse_pathspec: make sure the prefix part is wildcard-free
449
+ + rename field "raw" to "_raw" in struct pathspec
450
+ + tree-diff: remove the use of pathspec's raw[] in follow-rename codepath
451
+ + remove match_pathspec() in favor of match_pathspec_depth()
452
+ + remove init_pathspec() in favor of parse_pathspec()
453
+ + remove diff_tree_{setup,release}_paths
454
+ + convert common_prefix() to use struct pathspec
455
+ + convert add_files_to_cache to take struct pathspec
456
+ + convert {read,fill}_directory to take struct pathspec
457
+ + convert refresh_index to take struct pathspec
458
+ + convert report_path_error to take struct pathspec
459
+ + checkout: convert read_tree_some to take struct pathspec
460
+ + convert unmerge_cache to take struct pathspec
461
+ + convert run_add_interactive to use struct pathspec
462
+ + convert read_cache_preload() to take struct pathspec
463
+ + line-log: convert to use parse_pathspec
464
+ + reset: convert to use parse_pathspec
465
+ + add: convert to use parse_pathspec
466
+ + check-ignore: convert to use parse_pathspec
467
+ + archive: convert to use parse_pathspec
468
+ + ls-files: convert to use parse_pathspec
469
+ + rm: convert to use parse_pathspec
470
+ + checkout: convert to use parse_pathspec
471
+ + rerere: convert to use parse_pathspec
472
+ + status: convert to use parse_pathspec
473
+ + commit: convert to use parse_pathspec
474
+ + clean: convert to use parse_pathspec
475
+ + guard against new pathspec magic in pathspec matching code
476
+ + parse_pathspec: support prefixing original patterns
477
+ + parse_pathspec: support stripping/checking submodule paths
478
+ + parse_pathspec: support stripping submodule trailing slashes
479
+ + parse_pathspec: add special flag for max_depth feature
480
+ + convert some get_pathspec() calls to parse_pathspec()
481
+ + parse_pathspec: add PATHSPEC_PREFER_{CWD,FULL} flags
482
+ + parse_pathspec: save original pathspec for reporting
483
+ + add parse_pathspec() that converts cmdline args to struct pathspec
484
+ + pathspec: add copy_pathspec
485
+ + pathspec: i18n-ize error strings in pathspec parsing code
486
+ + move struct pathspec and related functions to pathspec.[ch]
487
+ + clean: remove unused variable "seen"
488
489
Use "struct pathspec" interface in more places, instead of array of
490
characters, the latter of which cannot express magic pathspecs
491
(e.g. ":(icase)makefile" that matches both Makefile and makefile).
492
527
- Will merge to 'next'.
528
-
529
-
530
-* rh/template-updates (2013-07-15) 3 commits
531
- (merged to 'next' on 2013-07-22 at 53dffdd)
532
- + templates: spell ASCII in uppercase in pre-commit hook
533
- + templates: Reformat pre-commit hook's message
534
- + templates: Use heredoc in pre-commit hook
535
-
536
- This is an earlier part of a 6 patch series, with log message
537
- corrected.
538
-
539
- Will merge to 'master'.
540
-
541
-
542
-* sb/misc-fixes (2013-07-15) 3 commits
543
- (merged to 'next' on 2013-07-21 at 880b08c)
544
- + diff.c: Do not initialize a variable, which gets reassigned anyway.
545
- + commit: Fix a memory leak in determine_author_info
546
- + daemon.c:handle: Remove unneeded check for null pointer.
547
-
548
- Assorted code cleanups and a minor fix.
549
-
550
- Will merge to 'master'.
493
+ Will cook in 'next'.
494
495
496
* jc/check-x-z (2013-07-11) 4 commits
@@ -569,25 +512,6 @@ of the repositories listed at
512
Will cook in 'next'.
513
514
572
-* jk/cat-file-batch-optim (2013-07-18) 9 commits
573
- (merged to 'next' on 2013-07-22 at 965897c)
574
- + Fix some sparse warnings
575
- + sha1_object_info_extended: pass object_info to helpers
576
- + sha1_object_info_extended: make type calculation optional
577
- + packed_object_info: make type lookup optional
578
- + packed_object_info: hoist delta type resolution to helper
579
- + sha1_loose_object_info: make type lookup optional
580
- + sha1_object_info_extended: rename "status" to "type"
581
- + cat-file: disable object/refname ambiguity check for batch mode
582
- + Merge branch 'nd/warn-ambiguous-object-name' into jk/cat-file-batch-optim
583
-
584
- If somebody wants to only know on-disk footprint of an object
585
- without having to know its type or payload size, we can bypass a
586
- lot of code to cheaply learn it.
587
-
588
- Will merge to 'master'.
589
-
590
-
515
* ms/fetch-prune-configuration (2013-07-18) 1 commit
516
(merged to 'next' on 2013-07-21 at 7ffb690)
517
+ fetch: make --prune configurable
@@ -603,16 +527,19 @@ of the repositories listed at
527
Will cook in 'next'.
528
529
606
-* km/http-curl-config-per-url (2013-07-22) 5 commits
607
- - config: allow http.<url>.* any user matching
608
- - tests: add new test for the url_normalize function
609
- - config: improve support for http.<url>.* settings
610
- - config: add support for http.<url>.* settings
611
- - http.c: fix parsing of http.sslCertPasswordProtected variable
530
+* km/http-curl-config-per-url (2013-07-25) 6 commits
531
+ (merged to 'next' on 2013-07-25 at 531e8cb)
532
+ + docs: update http.<url>.* options documentation
533
+ (merged to 'next' on 2013-07-23 at f84689b)
534
+ + config: allow http.<url>.* any user matching
535
+ + tests: add new test for the url_normalize function
536
+ + config: improve support for http.<url>.* settings
537
+ + config: add support for http.<url>.* settings
538
+ + http.c: fix parsing of http.sslCertPasswordProtected variable
539
540
Allows custom curl configuration per URL when talking over http.
541
615
- Will merge to 'next'.
542
+ Will cook in 'next'.
543
544
545
* km/svn-1.8-serf-only (2013-07-18) 3 commits
@@ -636,13 +563,14 @@ of the repositories listed at
563
Will hold, until we get any user.
564
565
639
-* jc/push-cas (2013-07-17) 6 commits
640
- - t5533: test "push --lockref"
641
- - push --lockref: tie it all together
642
- - push --lockref: implement logic to populate old_sha1_expect[]
643
- - remote.c: add command line option parser for "--lockref"
644
- - builtin/push.c: use OPT_BOOL, not OPT_BOOLEAN
645
- - cache.h: move remote/connect API out of it
566
+* jc/push-cas (2013-07-22) 6 commits
567
+ (merged to 'next' on 2013-07-23 at 72371b0)
568
+ + t5533: test "push --force-with-lease"
569
+ + push --force-with-lease: tie it all together
570
+ + push --force-with-lease: implement logic to populate old_sha1_expect[]
571
+ + remote.c: add command line option parser for "--force-with-lease"
572
+ + builtin/push.c: use OPT_BOOL, not OPT_BOOLEAN
573
+ + cache.h: move remote/connect API out of it
574
575
Allow a safer "rewind of the remote tip" push than blind "--force",
576
by requiring that the overwritten remote ref to be unchanged since
@@ -652,31 +580,13 @@ of the repositories listed at
580
the big red button to override any safety, thanks to J6t's sanity
581
(the original round allowed --lockref to defeat --force).
582
655
- I have strong doubts on the merit of the logic to choose the
656
- default for "--lockref" implemented here. It is suitable only for
657
- the simplest workflows, and it may hurt users more than it helps
658
- them.
659
-
660
- Also the option may be called something with "--force" prefix, as
661
- it is a weaker and more controlled form of "--force" that is used
662
- to defeat the usual "must fast-forward" safety.
663
-
664
- Will redo one more time.
665
-
666
-
667
-* es/contacts (2013-07-21) 5 commits
668
- (merged to 'next' on 2013-07-22 at a78c3d6)
669
- + contrib: contacts: add documentation
670
- + contrib: contacts: add mailmap support
671
- + contrib: contacts: interpret committish akin to format-patch
672
- + contrib: contacts: add ability to parse from committish
673
- + contrib: add git-contacts helper
674
-
675
- A helper to read from a set of format-patch output files or a range
676
- of commits and find those who may have insights to the code that
677
- the changes touch by running a series of "git blame" commands.
583
+ The logic to choose the default implemented here is fragile
584
+ (e.g. "git fetch" after seeing a failure will update the
585
+ remote-tracking branch and will make the next "push" pass,
586
+ defeating the safety pretty easily). It is suitable only for the
587
+ simplest workflows, and it may hurt users more than it helps them.
588
679
- Will merge to 'master'.
589
+ Will cook in 'next'.
590
591
592
* jc/pull-training-wheel (2013-07-19) 1 commit
@@ -718,16 +628,6 @@ of the repositories listed at
628
scripts, so fixing it is not all that urgent.
629
630
721
-* mh/ref-races-optim-invalidate-cached (2013-06-20) 1 commit
722
- (merged to 'next' on 2013-07-22 at 144d135)
723
- + refs: do not invalidate the packed-refs cache unnecessarily
724
-
725
- This requires the platform lstat() to be correct to avoid false
726
- negatives.
727
-
728
- Will merge to 'master'.
729
-
730
-
631
* jn/add-2.0-u-A-sans-pathspec (2013-04-26) 1 commit
632
- git add: -u/-A now affects the entire working tree
633