What's cooking (2013/03 #06)
Junio C Hamano committed
Mar 21, 2013 at 16:04 UTC
c34d0e1e1baeb60f7af75143107a9c0d6644b7ea
1 file changed
+805
-648
whats-cooking.txt
+805
-648
@@ -1,32 +1,18 @@
1
To: git@vger.kernel.org
2
Bcc: lwn@lwn.net
3
-Subject: What's cooking in git.git (Mar 2013, #05; Mon, 18)
4
-X-master-at: 1c71541ddd51968ca2957063b00da42941d13d98
5
-X-next-at: 4a4606479e7f1a5f817dd3a6582cd086aa0e70c2
3
+Subject: What's cooking in git.git (Mar 2013, #06; Thu, 21)
4
+X-master-at: 7b592fadf1e23b10b913e0771b9f711770597266
5
+X-next-at: ddc39398152011b6a071e28b0f18f69ba427d0c0
6
7
-What's cooking in git.git (Mar 2013, #05; Mon, 18)
7
+What's cooking in git.git (Mar 2013, #06; Thu, 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
-We have been scheduling each cycle to last for 8-10 weeks, but at
15
-the end of 1.8.2 cycle we already have quite a many topics already
16
-in flight. I am wondering perhaps we should cut this cycle a bit
17
-short by tagging -rc0 shortly after we have those we already have
18
-been cooking graduated to 'master' (we already have 200 non-merge
19
-commits on 'next').
20
-
21
-This issue classifies these topics in flight into three categories:
22
-Trivially Safe, Safe and Risky. "Risky" does not mean the topic is
23
-with known breakages, but just that my gut feeling tells it would
24
-not be surprising if there were unintended consequences, either
25
-these topics touch fairly deep part of the codepath, their design
26
-may not match some obscure but still valid use cases, or for some
27
-other unforeseen reason. Feeding this message to "cook -w" (found
28
-in 'todo' branch) script may make the overall picture easier to
29
-review.
14
+Quite a few topics have graduated to 'master'; they were all cooking
15
+in 'next' during the feature-freeze from the previous cycle.
16
17
You can find the changes described here in the integration branches
18
of the repositories listed at
@@ -34,856 +20,1025 @@ of the repositories listed at
20
http://git-blame.blogspot.com/p/git-public-repositories.html
21
22
--------------------------------------------------
37
-[New Topics]
23
+[Graduated to "master"]
24
39
-* ap/combine-diff-ignore-whitespace (2013-03-14) 1 commit
40
- (merged to 'next' on 2013-03-18 at 32435a6)
41
- + Allow combined diff to ignore white-spaces
25
+* ap/maint-update-index-h-is-for-help (2013-02-23) 1 commit
26
+ (merged to 'next' on 2013-03-19 at 1d7d180)
27
+ + update-index: allow "-h" to also display options
28
43
- Teach "diff --cc" output to honor options to ignore various forms
44
- of whitespace changes.
29
+ Originally merged to 'next' on 2013-02-25
30
46
- Will merge to 'master' in the 4th batch (Safe).
31
+ "git update-index -h" did not do the usual "-h(elp)" thing.
32
33
49
-* jk/checkout-attribute-lookup (2013-03-14) 2 commits
50
- (merged to 'next' on 2013-03-15 at 28bd515)
51
- + entry: fix filter lookup
52
- + t2003: modernize style
34
+* bc/commit-complete-lines-given-via-m-option (2013-02-19) 4 commits
35
+ (merged to 'next' on 2013-03-19 at 1609f9d)
36
+ + Documentation/git-commit.txt: rework the --cleanup section
37
+ + git-commit: only append a newline to -m mesg if necessary
38
+ + t7502: demonstrate breakage with a commit message with trailing newlines
39
+ + t/t7502: compare entire commit message with what was expected
40
54
- Codepath to stream blob object contents directly from the object
55
- store to filesystem did not use the correct path to find conversion
56
- filters when writing to temporary files.
41
+ Originally merged to 'next' on 2013-02-19
42
58
- Will merge to 'master' in the 4th batch (Safe).
43
+ 'git commit -m "$str"' when $str was already terminated with a LF
44
+ now avoids adding an extra LF to the message.
45
46
61
-* jk/difftool-dir-diff-edit-fix (2013-03-14) 3 commits
62
- (merged to 'next' on 2013-03-15 at add7b77)
63
- + difftool --dir-diff: symlink all files matching the working tree
64
- + difftool: avoid double slashes in symlink targets
65
- + git-difftool(1): fix formatting of --symlink description
47
+* da/difftool-fixes (2013-02-21) 4 commits
48
+ (merged to 'next' on 2013-03-19 at ffcb23f)
49
+ + t7800: "defaults" is no longer a builtin tool name
50
+ + t7800: modernize tests
51
+ + t7800: update copyright notice
52
+ + difftool: silence uninitialized variable warning
53
67
- "git difftool --dir-diff" made symlinks to working tree files when
68
- preparing a temporary directory structure, so that accidental edits
69
- of these files in the difftool are reflected back to the working
70
- tree, but the logic to decide when to do so was not quite right.
54
+ Originally merged to 'next' on 2013-02-25
55
72
- Will merge to 'master' in the 4th batch (Safe).
56
+ Minor maintenance updates to difftool, and updates to its tests.
57
58
75
-* lf/setup-prefix-pathspec (2013-03-14) 2 commits
76
- (merged to 'next' on 2013-03-14 at 7552204)
77
- + setup.c: check that the pathspec magic ends with ")"
78
- + setup.c: stop prefix_pathspec() from looping past the end of string
59
+* da/downcase-u-in-usage (2013-02-24) 20 commits
60
+ (merged to 'next' on 2013-02-26 at 977b67e)
61
+ + contrib/mw-to-git/t/install-wiki.sh: use a lowercase "usage:" string
62
+ + contrib/examples/git-remote.perl: use a lowercase "usage:" string
63
+ + tests: use a lowercase "usage:" string
64
+ + git-svn: use a lowercase "usage:" string
65
+ + Documentation/user-manual.txt: use a lowercase "usage:" string
66
+ + templates/hooks--update.sample: use a lowercase "usage:" string
67
+ + contrib/hooks/setgitperms.perl: use a lowercase "usage:" string
68
+ + contrib/examples: use a lowercase "usage:" string
69
+ + contrib/fast-import/import-zips.py: use spaces instead of tabs
70
+ + contrib/fast-import/import-zips.py: fix broken error message
71
+ + contrib/fast-import: use a lowercase "usage:" string
72
+ + contrib/credential: use a lowercase "usage:" string
73
+ + git-cvsimport: use a lowercase "usage:" string
74
+ + git-cvsimport: use a lowercase "usage:" string
75
+ + git-cvsexportcommit: use a lowercase "usage:" string
76
+ + git-archimport: use a lowercase "usage:" string
77
+ + git-merge-one-file: use a lowercase "usage:" string
78
+ + git-relink: use a lowercase "usage:" string
79
+ + git-svn: use a lowercase "usage:" string
80
+ + git-sh-setup: use a lowercase "usage:" string
81
80
- Rerolls aw/setup-prefix-pathspec.
82
+ Originally merged to 'next' on 2013-02-26
83
82
- Will merge to 'master' in the 3rd batch (Safe).
84
85
+* dm/ni-maxhost-may-be-missing (2013-02-25) 1 commit
86
+ (merged to 'next' on 2013-02-26 at 93ec2c9)
87
+ + git-compat-util.h: Provide missing netdb.h definitions
88
85
-* tb/document-status-u-tradeoff (2013-03-16) 2 commits
86
- (merged to 'next' on 2013-03-18 at 2fc53b0)
87
- + status: advise to consider use of -u when read_directory takes too long
88
- + git status: document trade-offs in choosing parameters to the -u option
89
+ Originally merged to 'next' on 2013-02-26
90
90
- Suggest users to look into using--untracked=no option when "git
91
- status" takes too long.
91
93
- Will merge to 'master' in the 2nd batch (Trivially Safe).
92
+* gp/avoid-explicit-mention-of-dot-git-refs (2013-02-24) 1 commit
93
+ (merged to 'next' on 2013-02-26 at ec42d98)
94
+ + Fix ".git/refs" stragglers
95
96
+ Originally merged to 'next' on 2013-02-26
97
96
-* jk/fully-peeled-packed-ref (2013-03-18) 4 commits
97
- (merged to 'next' on 2013-03-18 at cbcfa32)
98
- + pack-refs: add fully-peeled trait
99
- + pack-refs: write peeled entry for non-tags
100
- + use parse_object_or_die instead of die("bad object")
101
- + avoid segfaults on parse_object failure
98
103
- Not that we do not actively encourage having annotated tags outside
104
- refs/tags/ hierarchy, but they were not advertised correctly to the
105
- ls-remote and fetch with recent version of Git.
99
+* gp/describe-match-uses-glob-pattern (2013-02-24) 1 commit
100
+ (merged to 'next' on 2013-02-26 at c9cc789)
101
+ + describe: Document --match pattern format
102
107
- Will merge to 'master' in the 3rd batch (Safe).
103
+ Originally merged to 'next' on 2013-02-26
104
105
110
-* jk/peel-ref (2013-03-16) 3 commits
111
- (merged to 'next' on 2013-03-18 at 77f0e4e)
112
- + upload-pack: load non-tip "want" objects from disk
113
- + upload-pack: make sure "want" objects are parsed
114
- + upload-pack: drop lookup-before-parse optimization
106
+* jc/color-diff-doc (2013-02-22) 1 commit
107
+ (merged to 'next' on 2013-02-25 at c37541c)
108
+ + diff-options: unconfuse description of --color
109
116
- Recent optimization broke shallow clones.
110
+ Originally merged to 'next' on 2013-02-25
111
118
- Will merge to 'master' in the 3rd batch (Safe).
112
113
+* jc/fetch-raw-sha1 (2013-02-07) 4 commits
114
+ (merged to 'next' on 2013-03-19 at d5363f6)
115
+ + fetch: fetch objects by their exact SHA-1 object names
116
+ + upload-pack: optionally allow fetching from the tips of hidden refs
117
+ + fetch: use struct ref to represent refs to be fetched
118
+ + parse_fetch_refspec(): clarify the codeflow a bit
119
121
-* nd/index-pack-l10n-buf-overflow (2013-03-16) 1 commit
122
- (merged to 'next' on 2013-03-18 at b7a4a8e)
123
- + index-pack: fix buffer overflow caused by translations
120
+ Originally merged to 'next' on 2013-02-14
121
125
- Will merge to 'master' in the 2nd batch (Trivially Safe).
122
+ Allows requests to fetch objects at any tip of refs (including
123
+ hidden ones). It seems that there may be use cases even outside
124
+ Gerrit (e.g. $gmane/215701).
125
126
128
-* nd/magic-pathspecs (2013-03-15) 45 commits
129
- - Rename field "raw" to "_raw" in struct pathspec
130
- - pathspec: support :(glob) syntax
131
- - pathspec: make --literal-pathspecs disable pathspec magic
132
- - pathspec: support :(literal) syntax for noglob pathspec
133
- - Kill limit_pathspec_to_literal() as it's only used by parse_pathspec()
134
- - parse_pathspec: preserve prefix length via PATHSPEC_PREFIX_ORIGIN
135
- - parse_pathspec: make sure the prefix part is wildcard-free
136
- - tree-diff: remove the use of pathspec's raw[] in follow-rename codepath
137
- - Remove match_pathspec() in favor of match_pathspec_depth()
138
- - Remove init_pathspec() in favor of parse_pathspec()
139
- - Remove diff_tree_{setup,release}_paths
140
- - Convert common_prefix() to use struct pathspec
141
- - Convert add_files_to_cache to take struct pathspec
142
- - Convert {read,fill}_directory to take struct pathspec
143
- - Convert refresh_index to take struct pathspec
144
- - Convert report_path_error to take struct pathspec
145
- - checkout: convert read_tree_some to take struct pathspec
146
- - Convert unmerge_cache to take struct pathspec
147
- - Convert run_add_interactive to use struct pathspec
148
- - Convert read_cache_preload() to take struct pathspec
149
- - reset: convert to use parse_pathspec
150
- - add: convert to use parse_pathspec
151
- - check-ignore: convert to use parse_pathspec
152
- - archive: convert to use parse_pathspec
153
- - ls-files: convert to use parse_pathspec
154
- - rm: convert to use parse_pathspec
155
- - checkout: convert to use parse_pathspec
156
- - rerere: convert to use parse_pathspec
157
- - status: convert to use parse_pathspec
158
- - commit: convert to use parse_pathspec
159
- - clean: convert to use parse_pathspec
160
- - Guard against new pathspec magic in pathspec matching code
161
- - parse_pathspec: support prefixing original patterns
162
- - parse_pathspec: support stripping/checking submodule paths
163
- - parse_pathspec: support stripping submodule trailing slashes
164
- - parse_pathspec: a special flag for max_depth feature
165
- - Convert some get_pathspec() calls to parse_pathspec()
166
- - parse_pathspec: add PATHSPEC_PREFER_{CWD,FULL}
167
- - parse_pathspec: save original pathspec for reporting
168
- - Add parse_pathspec() that converts cmdline args to struct pathspec
169
- - pathspec: add copy_pathspec
170
- - pathspec: i18n-ize error strings in pathspec parsing code
171
- - Move struct pathspec and related functions to pathspec.[ch]
172
- - clean: remove unused variable "seen"
173
- - setup.c: check that the pathspec magic ends with ")"
127
+* jc/maint-push-refspec-default-doc (2013-03-08) 1 commit
128
+ (merged to 'next' on 2013-03-19 at 0b2fa41)
129
+ + Documentation/git-push: clarify the description of defaults
130
175
- Migrate the rest of codebase to use "struct pathspec" more.
131
+ Originally merged to 'next' on 2013-03-14
132
+
133
+ Clarify in the documentation "what" gets pushed to "where" when the
134
+ command line to "git push" does not say these explicitly.
135
+
136
+
137
+* jc/perl-cat-blob (2013-02-22) 1 commit
138
+ (merged to 'next' on 2013-03-19 at ad3aae3)
139
+ + Git.pm: fix cat_blob crashes on large files
140
+
141
+ Originally merged to 'next' on 2013-02-25
142
177
- Expecting a reroll, but this is overall going in the right direction.
178
- $gmane/218385
143
+ perl/Git.pm::cat_blob slurped everything in core only to write it
144
+ out to a file descriptor, which was not a very smart thing to do.
145
+
146
+
147
+* jk/graph-c-expose-symbols-for-cgit (2013-03-03) 1 commit
148
+ (merged to 'next' on 2013-03-04 at be35b12)
149
+ + Revert "graph.c: mark private file-scope symbols as static"
150
+
151
+ Originally merged to 'next' on 2013-03-04
152
+
153
+ In the v1.8.0 era, we changed symbols that do not have to be global
154
+ to file scope static, but a few functions in graph.c were used by
155
+ CGit from sideways bypassing the entry points of the API the
156
+ in-tree users use.
157
+
158
+
159
+* jk/mailsplit-maildir-muttsort (2013-03-02) 1 commit
160
+ (merged to 'next' on 2013-03-19 at 26f457c)
161
+ + mailsplit: sort maildir filenames more cleverly
162
+
163
+ Originally merged to 'next' on 2013-03-03
164
+
165
+ Sort filenames read from the maildir/ in a way that is more likely
166
+ to sort messages in the order the writing MUA meant to, by sorting
167
+ numeric segment in numeric order and non-numeric segment in
168
+ alphabetical order.
169
+
170
+
171
+* jk/utf-8-can-be-spelled-differently (2013-02-25) 1 commit
172
+ (merged to 'next' on 2013-03-19 at 18ebb43)
173
+ + utf8: accept alternate spellings of UTF-8
174
+
175
+ Originally merged to 'next' on 2013-02-26
176
+
177
+ Some platforms and users spell UTF-8 differently; retry with the
178
+ most official "UTF-8" when the system does not understand the
179
+ user-supplied encoding name that are the common alternative
180
+ spellings of UTF-8.
181
+
182
+
183
+* jn/shell-disable-interactive (2013-03-09) 2 commits
184
+ (merged to 'next' on 2013-03-19 at 4696604)
185
+ + shell: new no-interactive-login command to print a custom message
186
+ + shell doc: emphasize purpose and security model
187
+
188
+ Originally merged to 'next' on 2013-03-14
189
+
190
+ When the interactive access to git-shell is not enabled, we issue a
191
+ message meant to help the system admininstrator to enable it. Add
192
+ an explicit way to help the end users who connect to the service by
193
+ issuing custom messages to refuse such an access.
194
+
195
+
196
+* lf/bundle-verify-list-prereqs (2013-03-08) 2 commits
197
+ (merged to 'next' on 2013-03-08 at 9e55d6d)
198
+ + bundle: Add colons to list headings in "verify"
199
+ + bundle: Fix "verify" output if history is complete
200
+
201
+ Originally merged to 'next' on 2013-03-08
202
+
203
+
204
+* mg/gpg-interface-using-status (2013-02-14) 5 commits
205
+ (merged to 'next' on 2013-03-19 at bd54c7a)
206
+ + pretty: make %GK output the signing key for signed commits
207
+ + pretty: parse the gpg status lines rather than the output
208
+ + gpg_interface: allow to request status return
209
+ + log-tree: rely upon the check in the gpg_interface
210
+ + gpg-interface: check good signature in a reliable way
211
+
212
+ Originally merged to 'next' on 2013-02-26
213
+
214
+ Call "gpg" using the right API when validating the signature on
215
+ tags.
216
+
217
+
218
+* mn/send-email-works-with-credential (2013-02-27) 6 commits
219
+ (merged to 'next' on 2013-03-19 at 2daad63)
220
+ + git-send-email: use git credential to obtain password
221
+ + Git.pm: add interface for git credential command
222
+ + Git.pm: allow pipes to be closed prior to calling command_close_bidi_pipe
223
+ + Git.pm: refactor command_close_bidi_pipe to use _cmd_close
224
+ + Git.pm: fix example in command_close_bidi_pipe documentation
225
+ + Git.pm: allow command_close_bidi_pipe to be called as method
226
+
227
+ Originally merged to 'next' on 2013-02-27
228
+
229
+ Hooks the credential system to send-email.
230
+
231
+
232
+* nd/branch-error-cases (2013-02-23) 1 commit
233
+ (merged to 'next' on 2013-03-19 at 17834dd)
234
+ + branch: segfault fixes and validation
235
+
236
+ Originally merged to 'next' on 2013-02-25
237
+
238
+ "git branch" had more cases where it did not bother to check
239
+ nonsense command line parameters.
240
+
241
+
242
+* nd/count-garbage (2013-02-15) 4 commits
243
+ (merged to 'next' on 2013-03-19 at cf6fa47)
244
+ + count-objects: report how much disk space taken by garbage files
245
+ + count-objects: report garbage files in pack directory too
246
+ + sha1_file: reorder code in prepare_packed_git_one()
247
+ + git-count-objects.txt: describe each line in -v output
248
+
249
+ Originally merged to 'next' on 2013-02-17
250
+
251
+ "git count-objects -v" did not count leftover temporary packfiles
252
+ and other kinds of garbage.
253
+
254
+
255
+* nd/doc-index-format (2013-02-23) 3 commits
256
+ (merged to 'next' on 2013-02-26 at 4d3caea)
257
+ + update-index: list supported idx versions and their features
258
+ + read-cache.c: use INDEX_FORMAT_{LB,UB} in verify_hdr()
259
+ + index-format.txt: mention of v4 is missing in some places
260
+
261
+ Originally merged to 'next' on 2013-02-26
262
+
263
+ Update the index format documentation to mention the v4 format.
264
+
265
+
266
+* nd/index-pack-l10n-buf-overflow (2013-03-16) 1 commit
267
+ (merged to 'next' on 2013-03-19 at 01f0223)
268
+ + index-pack: fix buffer overflow caused by translations
269
+
270
+ Originally merged to 'next' on 2013-03-18
271
272
273
* nd/preallocate-hash (2013-03-16) 1 commit
182
- (merged to 'next' on 2013-03-18 at 0b928c5)
274
+ (merged to 'next' on 2013-03-19 at b92f720)
275
+ Preallocate hash tables when the number of inserts are known in advance
276
277
+ Originally merged to 'next' on 2013-03-18
278
+
279
When we know approximately how many entries we will have in the
280
hash-table, it makes sense to size the hash table to that number
281
from the beginning to avoid unnecessary rehashing.
282
189
- Will merge to 'master' in the 2nd batch (Trivially Safe).
283
284
+* ob/imap-send-ssl-verify (2013-02-20) 1 commit
285
+ (merged to 'next' on 2013-03-19 at 81aaed2)
286
+ + imap-send: support Server Name Indication (RFC4366)
287
192
-* pe/pull-rebase-v-q (2013-03-16) 1 commit
193
- (merged to 'next' on 2013-03-18 at 8e17883)
194
- + pull: Apply -q and -v options to rebase mode as well
288
+ Originally merged to 'next' on 2013-02-25
289
196
- Will merge to 'master' in the 2nd batch (Safe).
290
+ Correctly connect to SSL/TLS sites that serve multiple hostnames on
291
+ a single IP by including Server Name Indication in the client-hello.
292
293
199
-* rs/archive-zip-raw-compression (2013-03-16) 1 commit
200
- (merged to 'next' on 2013-03-18 at 229897a)
201
- + archive-zip: use deflateInit2() to ask for raw compressed data
202
- (this branch uses rs/zip-compresssed-size-with-export-subst.)
294
+* rj/msvc-build (2013-02-25) 5 commits
295
+ (merged to 'next' on 2013-02-26 at 7493068)
296
+ + msvc: avoid collisions between "tags" and "TAGS"
297
+ + msvc: test-svn-fe: Fix linker "unresolved external" error
298
+ + msvc: Fix build by adding missing symbol defines
299
+ + msvc: git-daemon: Fix linker "unresolved external" errors
300
+ + msvc: Fix compilation errors caused by poll.h emulation
301
204
- Will merge to 'master' in the 4th batch (Safe).
302
+ Originally merged to 'next' on 2013-02-26
303
304
207
-* jk/fast-export-object-lookup (2013-03-17) 2 commits
208
- (merged to 'next' on 2013-03-18 at 98c436d)
209
- + fast-export: do not load blob objects twice
210
- + fast-export: rename handle_object function
305
+* rs/zip-compresssed-size-with-export-subst (2013-02-27) 1 commit
306
+ (merged to 'next' on 2013-03-19 at 3d2ff88)
307
+ + archive-zip: fix compressed size for stored export-subst files
308
+ (this branch is used by rs/archive-zip-raw-compression.)
309
212
- Will merge to 'master' in the 3rd batch (Safe).
310
+ Originally merged to 'next' on 2013-03-03
311
312
+ When export-subst is used, "zip" output recorded incorrect
313
+ size of the file.
314
215
-* sw/safe-create-leading-dir-race (2013-03-17) 2 commits
216
- - SQUASH???
217
- - safe_create_leading_directories: fix race that could give a false negative
315
219
- Waiting for response to review comments.
316
+* rt/commit-cleanup-config (2013-02-23) 1 commit
317
+ (merged to 'next' on 2013-03-19 at 846cf01)
318
+ + t7502: perform commits using alternate editor in a subshell
319
221
---------------------------------------------------
222
-[Stalled]
320
+ Originally merged to 'next' on 2013-02-25
321
224
-* po/help-guides (2013-03-03) 5 commits
225
- - help doc: include --guide option description
226
- - help.c: add list_common_guides_help() function
227
- - help.c: add --guide option
228
- - help.c: use OPT_COUNTUP
229
- - show 'git help <guide>' usage, with examples
322
+ Fix tests that contaminated their environments and affected new
323
+ tests introduced later in the sequence by containing their effects
324
+ in their own subshells.
325
231
- Give more visibility to "concept guides" to help "git help" users.
326
233
- Expecting a reroll.
234
- $gmane/217384
327
+* tb/document-status-u-tradeoff (2013-03-16) 2 commits
328
+ (merged to 'next' on 2013-03-19 at 01573b6)
329
+ + status: advise to consider use of -u when read_directory takes too long
330
+ + git status: document trade-offs in choosing parameters to the -u option
331
332
+ Originally merged to 'next' on 2013-03-18
333
237
-* hv/config-from-strbuf (2013-03-10) 4 commits
238
- - teach config parsing to read from strbuf
239
- - config: make parsing stack struct independent from actual data source
240
- - config: drop file pointer validity check in get_next_char()
241
- - config: factor out config file stack management
334
+ Suggest users to look into using--untracked=no option when "git
335
+ status" takes too long.
336
243
- Expecting a reroll.
244
- $gmane/217811
337
338
+* tk/doc-filter-branch (2013-02-26) 2 commits
339
+ (merged to 'next' on 2013-02-26 at bd4638b)
340
+ + Documentation: filter-branch env-filter example
341
+ + git-filter-branch.txt: clarify ident variables usage
342
247
-* mb/gitweb-highlight-link-target (2012-12-20) 1 commit
248
- - Highlight the link target line in Gitweb using CSS
343
+ Originally merged to 'next' on 2013-02-26
344
250
- Expecting a reroll.
251
- $gmane/211935
345
346
+* tz/credential-authinfo (2013-02-25) 1 commit
347
+ (merged to 'next' on 2013-03-19 at 1a68953)
348
+ + Add contrib/credentials/netrc with GPG support
349
254
-* mb/remote-default-nn-origin (2012-07-11) 6 commits
255
- - Teach get_default_remote to respect remote.default.
256
- - Test that plain "git fetch" uses remote.default when on a detached HEAD.
257
- - Teach clone to set remote.default.
258
- - Teach "git remote" about remote.default.
259
- - Teach remote.c about the remote.default configuration setting.
260
- - Rename remote.c's default_remote_name static variables.
350
+ Originally merged to 'next' on 2013-02-27
351
262
- When the user does not specify what remote to interact with, we
263
- often attempt to use 'origin'. This can now be customized via a
264
- configuration variable.
352
+ A new read-only credential helper (in contrib/) to interact with
353
+ the .netrc/.authinfo files. Hopefully mn/send-email-authinfo topic
354
+ can rebuild on top of something like this.
355
266
- Expecting a reroll.
267
- $gmane/210151
356
269
- "The first remote becomes the default" bit is better done as a
270
- separate step.
357
+* we/submodule-update-prefix-output (2013-03-03) 1 commit
358
+ (merged to 'next' on 2013-03-19 at a818063)
359
+ + submodule update: when using recursion, show full path
360
+
361
+ Originally merged to 'next' on 2013-03-04
362
+
363
+ "git submodule update", when recursed into sub-submodules, did not
364
+ acccumulate the prefix paths.
365
+
366
+
367
+* wk/doc-pre-rebase (2013-02-24) 1 commit
368
+ (merged to 'next' on 2013-02-25 at a6ec310)
369
+ + Documentation/githooks: Explain pre-rebase parameters
370
+
371
+ Originally merged to 'next' on 2013-02-25
372
+
373
+
374
+* wk/user-manual-literal-format (2013-02-25) 1 commit
375
+ (merged to 'next' on 2013-02-26 at d59ce38)
376
+ + user-manual: Standardize backtick quoting
377
+
378
+ Originally merged to 'next' on 2013-02-26
379
380
--------------------------------------------------
273
-[Cooking]
381
+[New Topics]
382
275
-* ph/tag-force-no-warn-on-creation (2013-03-13) 1 commit
276
- (merged to 'next' on 2013-03-14 at 809e9ae)
277
- + tag: --force does not have to warn when creating tags
383
+* jc/merge-tag-object (2013-03-19) 1 commit
384
+ - merge: a random object may not necssarily be a commit
385
279
- Will merge to 'master' in the 3rd batch (Safe).
386
+ "git merge $(git rev-parse v1.8.2)" behaved quite differently from
387
+ "git merge v1.8.2" as if v1.8.2 were written as v1.8.2^0 and did
388
+ not pay much attention to the annotated tag payload.
389
390
+ This makes the code notice the type of the tag object, in addition
391
+ to the dwim_ref() based classification the current code uses
392
+ (i.e. the name appears in refs/tags/) to decide when to special
393
+ case merging of tags.
394
282
-* nd/branch-show-rebase-bisect-state (2013-03-16) 6 commits
283
- (merged to 'next' on 2013-03-18 at 4a46064)
284
- + branch: show more information when HEAD is detached
285
- + status: show more info than "currently not on any branch"
286
- + wt-status: move wt_status_get_state() out to wt_status_print()
287
- + wt-status: split wt_status_state parsing function out
288
- + wt-status: move strbuf into read_and_strip_branch()
289
- + Merge branch 'jc/reflog-reverse-walk' into nd/branch-show-rebase-bisect-state
290
- (this branch uses jc/reflog-reverse-walk.)
395
+ We may actually want to call the current behaviour a feature,
396
+ though. After all, the user went to great length to feed something
397
+ different from the normal "v1.8.2" to the command.
398
292
- Will merge to 'master' in the 4th batch (Safe).
399
400
+* jk/index-pack-correct-depth-fix (2013-03-20) 1 commit
401
+ - index-pack: always zero-initialize object_entry list
402
295
-* kb/p4merge (2013-03-13) 2 commits
296
- (merged to 'next' on 2013-03-14 at 6d390d4)
297
- + mergetools/p4merge: create a base if none available
298
- + mergetools/p4merge: swap LOCAL and REMOTE
403
+ "index-pack --fix-thin" used uninitialize value to compute delta
404
+ depths of objects it appends to the resulting pack.
405
300
- Adjust the order mergetools feeds the files to the p4merge backend
301
- to match the p4 convention.
406
303
- Will merge to 'master' in the 3rd batch (Safe).
407
+* jn/push-tests (2013-03-19) 3 commits
408
+ - push test: rely on &&-chaining instead of 'if bad; then echo Oops; fi'
409
+ - push test: simplify check of push result
410
+ - push test: use test_config when appropriate
411
412
+ Update t5516 with style fixes.
413
306
-* jc/maint-push-refspec-default-doc (2013-03-08) 1 commit
307
- (merged to 'next' on 2013-03-14 at 912eb14)
308
- + Documentation/git-push: clarify the description of defaults
414
310
- Clarify in the documentation "what" gets pushed to "where" when the
311
- command line to "git push" does not say these explicitly.
415
+* nd/index-pack-threaded-fixes (2013-03-19) 2 commits
416
+ - index-pack: guard nr_resolved_deltas reads by lock
417
+ - index-pack: protect deepest_delta in multithread code
418
313
- Will merge to 'master' in the 2nd batch (Trivially Safe).
419
+ "index-pack --verify-stat" used a few counters outside protection
420
+ of mutex, possibly showing incorrect numbers.
421
422
316
-* jc/reflog-reverse-walk (2013-03-08) 3 commits
317
- (merged to 'next' on 2013-03-14 at 3f2ed32)
318
- + reflog: add for_each_reflog_ent_reverse() API
319
- + for_each_recent_reflog_ent(): simplify opening of a reflog file
320
- + for_each_reflog_ent(): extract a helper to process a single entry
321
- (this branch is used by nd/branch-show-rebase-bisect-state.)
423
+* kb/status-ignored-optim (2013-03-19) 8 commits
424
+ - dir.c: git-status: avoid is_excluded checks for tracked files
425
+ - dir.c: replace is_path_excluded with now equivalent is_excluded API
426
+ - dir.c: unify is_excluded and is_path_excluded APIs
427
+ - dir.c: move prep_exclude and factor out parts of last_exclude_matching
428
+ - dir.c: git-status --ignored: don't list empty directories as ignored
429
+ - dir.c: git-status --ignored: don't list empty ignored directories
430
+ - dir.c: git-status --ignored: don't list files in ignored directories
431
+ - dir.c: git-status --ignored: don't drop ignored directories
432
323
- An internal function used to implement "git checkout @{-1}" was
324
- hard to use correctly.
433
+ "git status --ignored" had many corner case bugs. Also the command
434
+ has been optimized by taking advantage of the fact that paths that
435
+ are already known to the index do not have to be checked against
436
+ the .gitignore mechanism most of the time.
437
326
- Will merge to 'master' in the 3rd batch (Safe).
438
+ Expecting a review from others; this codepath is used not only by
439
+ status but also in add and clean, so a bug in it can result in lost
440
+ data.
441
442
329
-* jk/alias-in-bare (2013-03-08) 3 commits
330
- (merged to 'next' on 2013-03-09 at 2f9d72a)
331
- + setup: suppress implicit "." work-tree for bare repos
332
- + environment: add GIT_PREFIX to local_repo_env
333
- + cache.h: drop LOCAL_REPO_ENV_SIZE
443
+* rr/tests-dedup-test-config (2013-03-19) 1 commit
444
+ - t4018,7810,7811: remove test_config() redefinition
445
335
- An aliased command spawned from a bare repository that does not say
336
- it is bare with "core.bare = yes" is treated as non-bare by mistake.
446
+ Will merge to 'next'.
447
338
- Will merge to 'master' in the 3rd batch (Safe).
448
449
+* ja/directory-attrs (2013-03-20) 1 commit
450
+ - trailing slash required in .gitattributes
451
341
-* pw/p4-symlinked-root (2013-03-11) 3 commits
342
- (merged to 'next' on 2013-03-12 at 559b926)
343
- + git p4: avoid expanding client paths in chdir
344
- + git p4 test: should honor symlink in p4 client root
345
- + git p4 test: make sure P4CONFIG relative path works
452
+ Expose but not fix a recent regression that started not matching a
453
+ directory "sub" with gitattributes pattern "sub" (without trailing
454
+ slash).
455
347
- Will merge to 'master' in the 3rd batch (Safe).
456
457
+* jn/add-2.0-u-A-sans-pathspec (2013-03-20) 5 commits
458
+ - git add: -u/-A now affects the entire working tree
459
+ - add -A: only show pathless 'add -A' warning when changes exist outside cwd
460
+ - add -u: only show pathless 'add -u' warning when changes exist outside cwd
461
+ - add: make warn_pathless_add() a no-op after first call
462
+ - add: make pathless 'add [-u|-A]' warning a file-global function
463
350
-* jc/add-2.0-delete-default (2013-03-08) 3 commits
351
- (merged to 'next' on 2013-03-14 at 4684748)
352
- + git add <pathspec>... defaults to "-A"
353
- (merged to 'next' on 2013-03-14 at 2b7a7a2)
354
- + git add: start preparing for "git add <pathspec>..." to default to "-A"
355
- + builtin/add.c: simplify boolean variables
464
+ Replaces jc/add-2.0-u-A-sans-pathspec topic by not warning against
465
+ "add -u/-A" that is ran without pathspec when there is no change
466
+ outside the current directory.
467
357
- "git add dir/" updated modified files and added new files, but does
358
- not notice removed files, which may be "Huh?" to some users. They
359
- can of course use "git add -A dir/", but why should they?
468
361
- There seemed to be some interest in this topic, so resurrected and
362
- rebased on top of recent documentation updates to propose a
363
- possible transition plan.
469
+* rr/test-3200-style (2013-03-20) 1 commit
470
+ - t3200 (branch): modernize style
471
365
- Will cook in 'next' until Git 2.0.
472
+ Churns.
473
474
+ Will merge to 'next'.
475
368
-* jc/add-2.0-u-A-sans-pathspec (2013-03-14) 2 commits
369
- (merged to 'next' on 2013-03-14 at 9538716)
370
- + git add: -u/-A now affects the entire working tree
371
- + t2200: check that "add -u" limits itself to subdirectory
476
373
- "git add -u/-A" without pathspec has traditonally limited its
374
- operation to the current directory and its subdirectories, but in
375
- Git 1.8.2 we started encouraging users to be more explicit to
376
- specify "." when they mean it (and use ":/" to make it affect the
377
- entire working tree). With this, we finally change the behaviour
378
- and make it affect the entire working tree in Git 2.0.
477
+* tr/packed-object-info-wo-recursion (2013-03-20) 1 commit
478
+ - sha1_file: remove recursion in packed_object_info
479
380
- Will cook in 'next' until Git 2.0.
480
+ Attempts to reduce the stack footprint of sha1_object_info()
481
+ codepath.
482
483
383
-* jk/empty-archive (2013-03-10) 2 commits
384
- (merged to 'next' on 2013-03-12 at 942e592)
385
- + archive: handle commits with an empty tree
386
- + test-lib: factor out $GIT_UNZIP setup
484
+* sr/am-show-final-message-in-applying-indicator (2013-03-21) 1 commit
485
+ - git-am: show the final log message on "Applying:" indicator
486
388
- "git archive" reports a failure when asked to create an archive out
389
- of an empty tree. It would be more intuitive to give an empty
390
- archive back in such a case.
487
392
- Will merge to 'master' in the 3rd batch (Safe).
488
+* yd/doc-merge-annotated-tag (2013-03-21) 1 commit
489
+ - Documentation: merging a tag is a special case
490
491
395
-* jk/graph-c-expose-symbols-for-cgit (2013-03-03) 1 commit
396
- (merged to 'next' on 2013-03-04 at be35b12)
397
- + Revert "graph.c: mark private file-scope symbols as static"
492
+* yd/doc-is-in-asciidoc (2013-03-21) 1 commit
493
+ - CodingGuidelines: our documents are in AsciiDoc
494
399
- In the v1.8.0 era, we changed symbols that do not have to be global
400
- to file scope static, but a few functions in graph.c were used by
401
- CGit from sideways bypassing the entry points of the API the
402
- in-tree users use.
495
+--------------------------------------------------
496
+[Stalled]
497
404
- Will merge to 'master' in the 1st batch (Trivially Safe).
498
+* jc/format-patch (2013-02-21) 2 commits
499
+ - format-patch: --inline-single
500
+ - format-patch: rename "no_inline" field
501
502
+ A new option to send a single patch to the standard output to be
503
+ appended at the bottom of a message. I personally have no need for
504
+ this, but it was easy enough to cobble together. Tests, docs and
505
+ stripping out more MIMEy stuff are left as exercises to interested
506
+ parties.
507
407
-* we/submodule-update-prefix-output (2013-03-03) 1 commit
408
- (merged to 'next' on 2013-03-04 at 908df73)
409
- + submodule update: when using recursion, show full path
508
+ Not ready for inclusion.
509
+
510
+
511
+* mg/qnx6 (2013-02-25) 1 commit
512
+ - QNX: newer QNX 6.x.x is not so crippled
513
+
514
+ Not ready for inclusion.
515
411
- Will merge to 'master' in the 2nd batch (Safe).
516
517
+* po/help-guides (2013-03-03) 5 commits
518
+ - help doc: include --guide option description
519
+ - help.c: add list_common_guides_help() function
520
+ - help.c: add --guide option
521
+ - help.c: use OPT_COUNTUP
522
+ - show 'git help <guide>' usage, with examples
523
414
-* jc/nobody-sets-src-peer-ref (2013-03-04) 1 commit
415
- (merged to 'next' on 2013-03-07 at 1910e21)
416
- + match_push_refs(): nobody sets src->peer_ref anymore
524
+ Give more visibility to "concept guides" to help "git help" users.
525
418
- Dead code removal.
526
+ Expecting a reroll.
527
+ $gmane/217384
528
420
- Even though I think this change is correct, please report
421
- immediately if you find any unintended side effect.
529
423
- Will merge to 'master' in the 4th batch (Risky).
530
+* hv/config-from-strbuf (2013-03-10) 4 commits
531
+ - teach config parsing to read from strbuf
532
+ - config: make parsing stack struct independent from actual data source
533
+ - config: drop file pointer validity check in get_next_char()
534
+ - config: factor out config file stack management
535
536
+ Expecting a reroll.
537
+ $gmane/217811
538
426
-* jc/push-follow-tag (2013-03-05) 4 commits
427
- (merged to 'next' on 2013-03-09 at 748fbed)
428
- + push: --follow-tags
429
- + commit.c: use clear_commit_marks_many() in in_merge_bases_many()
430
- + commit.c: add in_merge_bases_many()
431
- + commit.c: add clear_commit_marks_many()
539
433
- The new "--follow-tags" option tells "git push" to push relevant
434
- annotated tags when pushing branches out.
540
+* mb/gitweb-highlight-link-target (2012-12-20) 1 commit
541
+ - Highlight the link target line in Gitweb using CSS
542
436
- Will merge to 'master' in the 3rd batch (Safe).
543
+ Expecting a reroll.
544
+ $gmane/211935
545
546
439
-* jc/maint-reflog-expire-clean-mark-typofix (2013-03-05) 1 commit
440
- (merged to 'next' on 2013-03-07 at 0a61cbb)
441
- + reflog: fix typo in "reflog expire" clean-up codepath
547
+* mb/remote-default-nn-origin (2012-07-11) 6 commits
548
+ - Teach get_default_remote to respect remote.default.
549
+ - Test that plain "git fetch" uses remote.default when on a detached HEAD.
550
+ - Teach clone to set remote.default.
551
+ - Teach "git remote" about remote.default.
552
+ - Teach remote.c about the remote.default configuration setting.
553
+ - Rename remote.c's default_remote_name static variables.
554
443
- Will merge to 'master' in the 3rd batch (Safe).
555
+ When the user does not specify what remote to interact with, we
556
+ often attempt to use 'origin'. This can now be customized via a
557
+ configuration variable.
558
559
+ Expecting a reroll.
560
+ $gmane/210151
561
446
-* lf/bundle-verify-list-prereqs (2013-03-08) 2 commits
447
- (merged to 'next' on 2013-03-08 at 9e55d6d)
448
- + bundle: Add colons to list headings in "verify"
449
- + bundle: Fix "verify" output if history is complete
562
+ "The first remote becomes the default" bit is better done as a
563
+ separate step.
564
451
- Will merge to 'master' in the 1st batch (Trivially Safe).
565
+--------------------------------------------------
566
+[Cooking]
567
568
+* ap/combine-diff-ignore-whitespace (2013-03-14) 1 commit
569
+ (merged to 'next' on 2013-03-19 at dfb2c98)
570
+ + Allow combined diff to ignore white-spaces
571
454
-* ks/rfc2047-one-char-at-a-time (2013-03-09) 1 commit
455
- (merged to 'next' on 2013-03-09 at a12465e)
456
- + format-patch: RFC 2047 says multi-octet character may not be split
572
+ Originally merged to 'next' on 2013-03-18
573
458
- When "format-patch" quoted a non-ascii strings on the header files,
459
- it incorrectly applied rfc2047 and chopped a single character in
460
- the middle of it.
574
+ Teach "diff --cc" output to honor options to ignore various forms
575
+ of whitespace changes.
576
462
- Will merge to 'master' in the 3rd batch (Safe).
577
+ Will merge to 'master' in the 4th batch (Safe).
578
579
465
-* kb/name-hash (2013-02-27) 1 commit
466
- (merged to 'next' on 2013-03-05 at 7f7e5d4)
467
- + name-hash.c: fix endless loop with core.ignorecase=true
580
+* jk/checkout-attribute-lookup (2013-03-20) 3 commits
581
+ (merged to 'next' on 2013-03-20 at 43a89e8)
582
+ + t2003: work around path mangling issue on Windows
583
+ (merged to 'next' on 2013-03-19 at b063a55)
584
+ + entry: fix filter lookup
585
+ + t2003: modernize style
586
469
- The code to keep track of what directory names are known to Git on
470
- platforms with case insensitive filesystems can get confused upon
471
- a hash collision between these pathnames and looped forever.
587
+ Originally merged to 'next' on 2013-03-15
588
473
- Even though I think this change is correct, please report
474
- immediately if you find any unintended side effect.
589
+ Codepath to stream blob object contents directly from the object
590
+ store to filesystem did not use the correct path to find conversion
591
+ filters when writing to temporary files.
592
476
- Will merge to 'master' in the 4th batch (Risky).
593
+ Will merge to 'master' in the 4th batch (Safe).
594
595
479
-* rs/zip-compresssed-size-with-export-subst (2013-02-27) 1 commit
480
- (merged to 'next' on 2013-03-03 at c1ac6d8)
481
- + archive-zip: fix compressed size for stored export-subst files
482
- (this branch is used by rs/archive-zip-raw-compression.)
596
+* jk/difftool-dir-diff-edit-fix (2013-03-14) 3 commits
597
+ (merged to 'next' on 2013-03-19 at e68014a)
598
+ + difftool --dir-diff: symlink all files matching the working tree
599
+ + difftool: avoid double slashes in symlink targets
600
+ + git-difftool(1): fix formatting of --symlink description
601
484
- When export-subst is used, "zip" output recorded incorrect
485
- size of the file.
602
+ Originally merged to 'next' on 2013-03-15
603
487
- Will merge to 'master' in the 2nd batch (Safe).
604
+ "git difftool --dir-diff" made symlinks to working tree files when
605
+ preparing a temporary directory structure, so that accidental edits
606
+ of these files in the difftool are reflected back to the working
607
+ tree, but the logic to decide when to do so was not quite right.
608
+
609
+ Will merge to 'master' in the 4th batch (Safe).
610
611
490
-* jc/describe (2013-02-28) 1 commit
491
- (merged to 'next' on 2013-03-05 at 6b353f3)
492
- + describe: --match=<pattern> must limit the refs even when used with --all
612
+* lf/setup-prefix-pathspec (2013-03-14) 2 commits
613
+ (merged to 'next' on 2013-03-19 at 0017e9e)
614
+ + setup.c: check that the pathspec magic ends with ")"
615
+ + setup.c: stop prefix_pathspec() from looping past the end of string
616
494
- The "--match=<pattern>" option of "git describe", when used with
495
- "--all" to allow refs that are not annotated tags to be used as a
496
- base of description, did not restrict the output from the command
497
- to those that match the given pattern.
617
+ Originally merged to 'next' on 2013-03-14
618
499
- We may want to have a looser matching that does not restrict to tags,
500
- but that can be done as a follow-up topic; this step is purely a bugfix.
619
+ "git cmd -- ':(top'" was not diagnosed as an invalid syntax, and
620
+ instead the parser kept reading beyond the end of the string.
621
622
Will merge to 'master' in the 3rd batch (Safe).
623
624
505
-* jk/mailsplit-maildir-muttsort (2013-03-02) 1 commit
506
- (merged to 'next' on 2013-03-03 at d5f7735)
507
- + mailsplit: sort maildir filenames more cleverly
625
+* jk/fully-peeled-packed-ref (2013-03-18) 4 commits
626
+ (merged to 'next' on 2013-03-19 at fa92bc7)
627
+ + pack-refs: add fully-peeled trait
628
+ + pack-refs: write peeled entry for non-tags
629
+ + use parse_object_or_die instead of die("bad object")
630
+ + avoid segfaults on parse_object failure
631
509
- Will merge to 'master' in the 2nd batch (Safe).
632
+ Originally merged to 'next' on 2013-03-18
633
634
+ Not that we do not actively encourage having annotated tags outside
635
+ refs/tags/ hierarchy, but they were not advertised correctly to the
636
+ ls-remote and fetch with recent version of Git.
637
512
-* tr/line-log (2013-02-28) 5 commits
513
- . log -L: :pattern:file syntax to find by funcname
514
- . Implement line-history search (git log -L)
515
- . Export rewrite_parents() for 'log -L'
516
- . blame: introduce $ as "end of file" in -L syntax
517
- . Refactor parse_loc
638
+ Will merge to 'master' in the 3rd batch (Safe).
639
519
- Expecting a reroll.
520
- $gmane/217278
640
522
- Kept outside 'pu' for now as it collides with nd/magic-pathspecs.
641
+* jk/peel-ref (2013-03-16) 3 commits
642
+ (merged to 'next' on 2013-03-19 at f0d4c16)
643
+ + upload-pack: load non-tip "want" objects from disk
644
+ + upload-pack: make sure "want" objects are parsed
645
+ + upload-pack: drop lookup-before-parse optimization
646
647
+ Originally merged to 'next' on 2013-03-18
648
525
-* jc/perl-cat-blob (2013-02-22) 1 commit
526
- (merged to 'next' on 2013-02-25 at 7c0079a)
527
- + Git.pm: fix cat_blob crashes on large files
649
+ Recent optimization broke shallow clones.
650
529
- perl/Git.pm::cat_blob slurped everything in core only to write it
530
- out to a file descriptor, which was not a very smart thing to do.
651
+ Will merge to 'master' in the 3rd batch (Safe).
652
532
- Will merge to 'master' in the 2nd batch (Safe).
653
654
+* nd/magic-pathspecs (2013-03-21) 46 commits
655
+ - fixup
656
+ - Rename field "raw" to "_raw" in struct pathspec
657
+ - pathspec: support :(glob) syntax
658
+ - pathspec: make --literal-pathspecs disable pathspec magic
659
+ - pathspec: support :(literal) syntax for noglob pathspec
660
+ - Kill limit_pathspec_to_literal() as it's only used by parse_pathspec()
661
+ - parse_pathspec: preserve prefix length via PATHSPEC_PREFIX_ORIGIN
662
+ - parse_pathspec: make sure the prefix part is wildcard-free
663
+ - tree-diff: remove the use of pathspec's raw[] in follow-rename codepath
664
+ - Remove match_pathspec() in favor of match_pathspec_depth()
665
+ - Remove init_pathspec() in favor of parse_pathspec()
666
+ - Remove diff_tree_{setup,release}_paths
667
+ - Convert common_prefix() to use struct pathspec
668
+ - Convert add_files_to_cache to take struct pathspec
669
+ - Convert {read,fill}_directory to take struct pathspec
670
+ - Convert refresh_index to take struct pathspec
671
+ - Convert report_path_error to take struct pathspec
672
+ - checkout: convert read_tree_some to take struct pathspec
673
+ - Convert unmerge_cache to take struct pathspec
674
+ - Convert run_add_interactive to use struct pathspec
675
+ - Convert read_cache_preload() to take struct pathspec
676
+ - reset: convert to use parse_pathspec
677
+ - add: convert to use parse_pathspec
678
+ - check-ignore: convert to use parse_pathspec
679
+ - archive: convert to use parse_pathspec
680
+ - ls-files: convert to use parse_pathspec
681
+ - rm: convert to use parse_pathspec
682
+ - checkout: convert to use parse_pathspec
683
+ - rerere: convert to use parse_pathspec
684
+ - status: convert to use parse_pathspec
685
+ - commit: convert to use parse_pathspec
686
+ - clean: convert to use parse_pathspec
687
+ - Guard against new pathspec magic in pathspec matching code
688
+ - parse_pathspec: support prefixing original patterns
689
+ - parse_pathspec: support stripping/checking submodule paths
690
+ - parse_pathspec: support stripping submodule trailing slashes
691
+ - parse_pathspec: a special flag for max_depth feature
692
+ - Convert some get_pathspec() calls to parse_pathspec()
693
+ - parse_pathspec: add PATHSPEC_PREFER_{CWD,FULL}
694
+ - parse_pathspec: save original pathspec for reporting
695
+ - Add parse_pathspec() that converts cmdline args to struct pathspec
696
+ - pathspec: add copy_pathspec
697
+ - pathspec: i18n-ize error strings in pathspec parsing code
698
+ - Move struct pathspec and related functions to pathspec.[ch]
699
+ - clean: remove unused variable "seen"
700
+ - setup.c: check that the pathspec magic ends with ")"
701
535
-* nd/doc-index-format (2013-02-23) 3 commits
536
- (merged to 'next' on 2013-02-26 at 4d3caea)
537
- + update-index: list supported idx versions and their features
538
- + read-cache.c: use INDEX_FORMAT_{LB,UB} in verify_hdr()
539
- + index-format.txt: mention of v4 is missing in some places
702
+ Migrate the rest of codebase to use "struct pathspec" more.
703
541
- Update the index format documentation to mention the v4 format.
704
543
- Will merge to 'master' in the 1st batch (Trivially Safe).
705
+* pe/pull-rebase-v-q (2013-03-16) 1 commit
706
+ (merged to 'next' on 2013-03-19 at 10b311c)
707
+ + pull: Apply -q and -v options to rebase mode as well
708
709
+ Originally merged to 'next' on 2013-03-18
710
546
-* ap/maint-diff-rename-avoid-overlap (2013-03-06) 3 commits
547
- (merged to 'next' on 2013-03-06 at 3bc8dda)
548
- + tests: make sure rename pretty print works
549
- (merged to 'next' on 2013-02-26 at 19d70bf)
550
- + diff: prevent pprint_rename from underrunning input
551
- (merged to 'next' on 2013-02-25 at c9bd6d3)
552
- + diff: Fix rename pretty-print when suffix and prefix overlap
711
+ Teach "git pull --rebase" to pass "-v/-q" command line options to
712
+ underlying "git rebase".
713
554
- The logic used by "git diff -M --stat" to shorten the names of
555
- files before and after a rename did not work correctly when the
556
- common prefix and suffix between the two filenames overlapped.
714
+ Will merge to 'master' in the 2nd batch (Safe).
715
558
- Will merge to 'master' in the 3rd batch (Safe).
716
717
+* rs/archive-zip-raw-compression (2013-03-16) 1 commit
718
+ (merged to 'next' on 2013-03-19 at 8cc1cb3)
719
+ + archive-zip: use deflateInit2() to ask for raw compressed data
720
561
-* ap/maint-update-index-h-is-for-help (2013-02-23) 1 commit
562
- (merged to 'next' on 2013-02-25 at f5f767c)
563
- + update-index: allow "-h" to also display options
721
+ Originally merged to 'next' on 2013-03-18
722
565
- Will merge to 'master' in the 2nd batch (Safe).
723
+ Will merge to 'master' in the 4th batch (Safe).
724
725
568
-* jc/color-diff-doc (2013-02-22) 1 commit
569
- (merged to 'next' on 2013-02-25 at c37541c)
570
- + diff-options: unconfuse description of --color
726
+* jk/fast-export-object-lookup (2013-03-17) 2 commits
727
+ (merged to 'next' on 2013-03-19 at 026ac3d)
728
+ + fast-export: do not load blob objects twice
729
+ + fast-export: rename handle_object function
730
572
- Will merge to 'master' in the 1st batch (Trivially Safe).
731
+ Originally merged to 'next' on 2013-03-18
732
733
+ Will merge to 'master' in the 3rd batch (Safe).
734
575
-* nd/branch-error-cases (2013-02-23) 1 commit
576
- (merged to 'next' on 2013-02-25 at 1d0289f)
577
- + branch: segfault fixes and validation
735
579
- "git branch" had more cases where it did not bother to check
580
- nonsense command line parameters.
736
+* sw/safe-create-leading-dir-race (2013-03-17) 2 commits
737
+ - SQUASH???
738
+ - safe_create_leading_directories: fix race that could give a false negative
739
582
- Will merge to 'master' in the 2nd batch (Safe).
740
+ Waiting for response to review comments.
741
742
585
-* rt/commit-cleanup-config (2013-02-23) 1 commit
586
- (merged to 'next' on 2013-02-25 at 8249b61)
587
- + t7502: perform commits using alternate editor in a subshell
743
+* ph/tag-force-no-warn-on-creation (2013-03-13) 1 commit
744
+ (merged to 'next' on 2013-03-19 at b273fd3)
745
+ + tag: --force does not have to warn when creating tags
746
589
- Fix tests that contaminated their environments and affected new
590
- tests introduced later in the sequence by containing their effects
591
- in their own subshells.
747
+ Originally merged to 'next' on 2013-03-14
748
593
- Will merge to 'master' in the 2nd batch (Safe).
749
+ "git tag -f <tag>" always said "Updated tag '<tag>'" even when
750
+ creating a new tag (i.e. not overwriting nor updating).
751
752
+ Will merge to 'master' in the 3rd batch (Safe).
753
596
-* wk/doc-pre-rebase (2013-02-24) 1 commit
597
- (merged to 'next' on 2013-02-25 at a6ec310)
598
- + Documentation/githooks: Explain pre-rebase parameters
754
600
- Will merge to 'master' in the 1st batch (Trivially Safe).
755
+* nd/branch-show-rebase-bisect-state (2013-03-16) 6 commits
756
+ - branch: show more information when HEAD is detached
757
+ - status: show more info than "currently not on any branch"
758
+ - wt-status: move wt_status_get_state() out to wt_status_print()
759
+ - wt-status: split wt_status_state parsing function out
760
+ - wt-status: move strbuf into read_and_strip_branch()
761
+ - Merge branch 'jc/reflog-reverse-walk' into nd/branch-show-rebase-bisect-state
762
+ (this branch uses jc/reflog-reverse-walk.)
763
764
+ Originally merged to 'next' on 2013-03-18
765
+
766
+ The "biseting X" display is iffy.
767
603
-* da/downcase-u-in-usage (2013-02-24) 20 commits
604
- (merged to 'next' on 2013-02-26 at 977b67e)
605
- + contrib/mw-to-git/t/install-wiki.sh: use a lowercase "usage:" string
606
- + contrib/examples/git-remote.perl: use a lowercase "usage:" string
607
- + tests: use a lowercase "usage:" string
608
- + git-svn: use a lowercase "usage:" string
609
- + Documentation/user-manual.txt: use a lowercase "usage:" string
610
- + templates/hooks--update.sample: use a lowercase "usage:" string
611
- + contrib/hooks/setgitperms.perl: use a lowercase "usage:" string
612
- + contrib/examples: use a lowercase "usage:" string
613
- + contrib/fast-import/import-zips.py: use spaces instead of tabs
614
- + contrib/fast-import/import-zips.py: fix broken error message
615
- + contrib/fast-import: use a lowercase "usage:" string
616
- + contrib/credential: use a lowercase "usage:" string
617
- + git-cvsimport: use a lowercase "usage:" string
618
- + git-cvsimport: use a lowercase "usage:" string
619
- + git-cvsexportcommit: use a lowercase "usage:" string
620
- + git-archimport: use a lowercase "usage:" string
621
- + git-merge-one-file: use a lowercase "usage:" string
622
- + git-relink: use a lowercase "usage:" string
623
- + git-svn: use a lowercase "usage:" string
624
- + git-sh-setup: use a lowercase "usage:" string
768
626
- Will merge to 'master' in the 1st batch (Trivially Safe).
769
+* kb/p4merge (2013-03-13) 2 commits
770
+ (merged to 'next' on 2013-03-19 at 43cc599)
771
+ + mergetools/p4merge: create a base if none available
772
+ + mergetools/p4merge: swap LOCAL and REMOTE
773
774
+ Originally merged to 'next' on 2013-03-14
775
629
-* dm/ni-maxhost-may-be-missing (2013-02-25) 1 commit
630
- (merged to 'next' on 2013-02-26 at 93ec2c9)
631
- + git-compat-util.h: Provide missing netdb.h definitions
776
+ Adjust the order mergetools feeds the files to the p4merge backend
777
+ to match the p4 convention.
778
633
- Will merge to 'master' in the 1st batch (Trivially Safe).
779
+ Will merge to 'master' in the 3rd batch (Safe).
780
781
636
-* gp/avoid-explicit-mention-of-dot-git-refs (2013-02-24) 1 commit
637
- (merged to 'next' on 2013-02-26 at ec42d98)
638
- + Fix ".git/refs" stragglers
782
+* jc/reflog-reverse-walk (2013-03-08) 3 commits
783
+ (merged to 'next' on 2013-03-19 at 25beb2a)
784
+ + reflog: add for_each_reflog_ent_reverse() API
785
+ + for_each_recent_reflog_ent(): simplify opening of a reflog file
786
+ + for_each_reflog_ent(): extract a helper to process a single entry
787
+ (this branch is used by nd/branch-show-rebase-bisect-state.)
788
640
- Will merge to 'master' in the 1st batch (Trivially Safe).
789
+ Originally merged to 'next' on 2013-03-14
790
791
+ An internal function used to implement "git checkout @{-1}" was
792
+ hard to use correctly.
793
643
-* gp/describe-match-uses-glob-pattern (2013-02-24) 1 commit
644
- (merged to 'next' on 2013-02-26 at c9cc789)
645
- + describe: Document --match pattern format
646
- (this branch is used by gp/forbid-describe-all-match.)
794
+ Will merge to 'master' in the 3rd batch (Safe).
795
648
- Will merge to 'master' in the 1st batch (Trivially Safe).
796
797
+* jk/alias-in-bare (2013-03-08) 3 commits
798
+ (merged to 'next' on 2013-03-19 at d2b4227)
799
+ + setup: suppress implicit "." work-tree for bare repos
800
+ + environment: add GIT_PREFIX to local_repo_env
801
+ + cache.h: drop LOCAL_REPO_ENV_SIZE
802
651
-* jk/common-make-variables-export-safety (2013-02-25) 1 commit
652
- (merged to 'next' on 2013-03-05 at 084ae43)
653
- + Makefile: make mandir, htmldir and infodir absolute
803
+ Originally merged to 'next' on 2013-03-09
804
655
- Make the three variables safer to be exported to submakes by
656
- ensuring that they are full paths so that they can be used as
657
- installation location.
805
+ An aliased command spawned from a bare repository that does not say
806
+ it is bare with "core.bare = yes" is treated as non-bare by mistake.
807
659
- Even though I think this change is correct, please report
660
- immediately if you find any unintended side effect.
808
+ Will merge to 'master' in the 3rd batch (Safe).
809
662
- Will merge to 'master' in the 4th batch (Risky).
810
811
+* pw/p4-symlinked-root (2013-03-11) 3 commits
812
+ (merged to 'next' on 2013-03-19 at 9127e82)
813
+ + git p4: avoid expanding client paths in chdir
814
+ + git p4 test: should honor symlink in p4 client root
815
+ + git p4 test: make sure P4CONFIG relative path works
816
+
817
+ Originally merged to 'next' on 2013-03-12
818
665
-* jk/suppress-clang-warning (2013-02-25) 1 commit
666
- (merged to 'next' on 2013-03-14 at f30489a)
667
- + fix clang -Wtautological-compare with unsigned enum
819
+ "git p4" did not behave well when the path to the root of the P4
820
+ client was not its real path.
821
822
Will merge to 'master' in the 3rd batch (Safe).
823
824
672
-* mg/qnx6 (2013-02-25) 1 commit
673
- - QNX: newer QNX 6.x.x is not so crippled
825
+* jc/add-2.0-delete-default (2013-03-08) 3 commits
826
+ - git add <pathspec>... defaults to "-A"
827
+ - git add: start preparing for "git add <pathspec>..." to default to "-A"
828
+ - builtin/add.c: simplify boolean variables
829
675
- Still under discussion.
676
- Not ready for inclusion.
830
+ Originally merged to 'next' on 2013-03-14
831
832
+ "git add dir/" updated modified files and added new files, but does
833
+ not notice removed files, which may be "Huh?" to some users. They
834
+ can of course use "git add -A dir/", but why should they?
835
679
-* mg/unsigned-time-t (2013-02-25) 2 commits
680
- (merged to 'next' on 2013-03-14 at 815f297)
681
- + Fix time offset calculation in case of unsigned time_t
682
- + date.c: fix unsigned time_t comparison
836
+ There seemed to be some interest in this topic, so resurrected and
837
+ rebased on top of recent documentation updates to propose a
838
+ possible transition plan.
839
684
- A few workarounds for systems with unsigned time_t.
840
+ Will cook in 'next' until Git 2.0.
841
+
842
+
843
+* jk/empty-archive (2013-03-10) 2 commits
844
+ (merged to 'next' on 2013-03-19 at bb4eb61)
845
+ + archive: handle commits with an empty tree
846
+ + test-lib: factor out $GIT_UNZIP setup
847
+
848
+ Originally merged to 'next' on 2013-03-12
849
+
850
+ "git archive" reports a failure when asked to create an archive out
851
+ of an empty tree. It would be more intuitive to give an empty
852
+ archive back in such a case.
853
854
Will merge to 'master' in the 3rd batch (Safe).
855
856
689
-* rj/msvc-build (2013-02-25) 5 commits
690
- (merged to 'next' on 2013-02-26 at 7493068)
691
- + msvc: avoid collisions between "tags" and "TAGS"
692
- + msvc: test-svn-fe: Fix linker "unresolved external" error
693
- + msvc: Fix build by adding missing symbol defines
694
- + msvc: git-daemon: Fix linker "unresolved external" errors
695
- + msvc: Fix compilation errors caused by poll.h emulation
857
+* jc/nobody-sets-src-peer-ref (2013-03-04) 1 commit
858
+ - match_push_refs(): nobody sets src->peer_ref anymore
859
697
- Will merge to 'master' in the 1st batch (Trivially Safe).
860
+ Originally merged to 'next' on 2013-03-07
861
862
+ Dead code removal.
863
700
-* wk/user-manual-literal-format (2013-02-25) 1 commit
701
- (merged to 'next' on 2013-02-26 at d59ce38)
702
- + user-manual: Standardize backtick quoting
864
+ Even though I think this change is correct, please report
865
+ immediately if you find any unintended side effect.
866
704
- Will merge to 'master' in the 1st batch (Trivially Safe).
867
+ Will merge to 'master' in the 4th batch (Risky).
868
869
707
-* jk/utf-8-can-be-spelled-differently (2013-02-25) 1 commit
708
- (merged to 'next' on 2013-02-26 at c079525)
709
- + utf8: accept alternate spellings of UTF-8
870
+* jc/push-follow-tag (2013-03-05) 4 commits
871
+ (merged to 'next' on 2013-03-19 at d302a10)
872
+ + push: --follow-tags
873
+ + commit.c: use clear_commit_marks_many() in in_merge_bases_many()
874
+ + commit.c: add in_merge_bases_many()
875
+ + commit.c: add clear_commit_marks_many()
876
711
- Will merge to 'master' in the 2nd batch (Safe).
877
+ Originally merged to 'next' on 2013-03-09
878
879
+ The new "--follow-tags" option tells "git push" to push relevant
880
+ annotated tags when pushing branches out.
881
714
-* jk/pkt-line-cleanup (2013-02-24) 19 commits
715
- (merged to 'next' on 2013-02-25 at d83e970)
716
- + remote-curl: always parse incoming refs
717
- + remote-curl: move ref-parsing code up in file
718
- + remote-curl: pass buffer straight to get_remote_heads
719
- + teach get_remote_heads to read from a memory buffer
720
- + pkt-line: share buffer/descriptor reading implementation
721
- + pkt-line: provide a LARGE_PACKET_MAX static buffer
722
- + pkt-line: move LARGE_PACKET_MAX definition from sideband
723
- + pkt-line: teach packet_read_line to chomp newlines
724
- + pkt-line: provide a generic reading function with options
725
- + pkt-line: drop safe_write function
726
- + pkt-line: move a misplaced comment
727
- + write_or_die: raise SIGPIPE when we get EPIPE
728
- + upload-archive: use argv_array to store client arguments
729
- + upload-archive: do not copy repo name
730
- + send-pack: prefer prefixcmp over memcmp in receive_status
731
- + fetch-pack: fix out-of-bounds buffer offset in get_ack
732
- + upload-pack: remove packet debugging harness
733
- + upload-pack: do not add duplicate objects to shallow list
734
- + upload-pack: use get_sha1_hex to parse "shallow" lines
882
+ Will merge to 'master' in the 3rd batch (Safe).
883
736
- Cleans up pkt-line API, implementation and its callers to make them
737
- more robust. Even though I think this change is correct, please
738
- report immediately if you find any unintended side effect.
884
740
- Will merge to 'master' in the 3rd batch (Risky).
885
+* jc/maint-reflog-expire-clean-mark-typofix (2013-03-05) 1 commit
886
+ (merged to 'next' on 2013-03-19 at a4f9eac)
887
+ + reflog: fix typo in "reflog expire" clean-up codepath
888
889
+ Originally merged to 'next' on 2013-03-07
890
743
-* ob/imap-send-ssl-verify (2013-02-20) 1 commit
744
- (merged to 'next' on 2013-02-25 at e897609)
745
- + imap-send: support Server Name Indication (RFC4366)
891
+ In "git reflog expire", REACHABLE bit was not cleared from the
892
+ correct objects.
893
747
- Correctly connect to SSL/TLS sites that serve multiple hostnames on
748
- a single IP by including Server Name Indication in the client-hello.
894
+ Will merge to 'master' in the 3rd batch (Safe).
895
750
- Will merge to 'master' in the 2nd batch (Safe).
896
897
+* ks/rfc2047-one-char-at-a-time (2013-03-09) 1 commit
898
+ (merged to 'next' on 2013-03-19 at 577ddf2)
899
+ + format-patch: RFC 2047 says multi-octet character may not be split
900
753
-* jc/format-patch (2013-02-21) 2 commits
754
- - format-patch: --inline-single
755
- - format-patch: rename "no_inline" field
901
+ Originally merged to 'next' on 2013-03-09
902
757
- A new option to send a single patch to the standard output to be
758
- appended at the bottom of a message. I personally have no need for
759
- this, but it was easy enough to cobble together. Tests, docs and
760
- stripping out more MIMEy stuff are left as exercises to interested
761
- parties.
903
+ When "format-patch" quoted a non-ascii strings on the header files,
904
+ it incorrectly applied rfc2047 and chopped a single character in
905
+ the middle of it.
906
763
- Not ready for inclusion.
907
+ Will merge to 'master' in the 3rd batch (Safe).
908
909
766
-* tk/doc-filter-branch (2013-02-26) 2 commits
767
- (merged to 'next' on 2013-02-26 at bd4638b)
768
- + Documentation: filter-branch env-filter example
769
- + git-filter-branch.txt: clarify ident variables usage
910
+* kb/name-hash (2013-02-27) 1 commit
911
+ - name-hash.c: fix endless loop with core.ignorecase=true
912
+
913
+ Originally merged to 'next' on 2013-03-05
914
771
- Will merge to 'master' in the 1st batch (Trivially Safe).
915
+ The code to keep track of what directory names are known to Git on
916
+ platforms with case insensitive filesystems can get confused upon
917
+ a hash collision between these pathnames and looped forever.
918
919
+ Even though I think this change is correct, please report
920
+ immediately if you find any unintended side effect.
921
774
-* bc/commit-complete-lines-given-via-m-option (2013-02-19) 4 commits
775
- (merged to 'next' on 2013-02-19 at cf622b7)
776
- + Documentation/git-commit.txt: rework the --cleanup section
777
- + git-commit: only append a newline to -m mesg if necessary
778
- + t7502: demonstrate breakage with a commit message with trailing newlines
779
- + t/t7502: compare entire commit message with what was expected
922
+ Will merge to 'master' in the 4th batch (Risky).
923
781
- 'git commit -m "$str"' when $str was already terminated with a LF
782
- now avoids adding an extra LF to the message.
924
784
- Will merge to 'master' in the 2nd batch (Safe).
925
+* jc/describe (2013-02-28) 1 commit
926
+ (merged to 'next' on 2013-03-19 at 89e6e47)
927
+ + describe: --match=<pattern> must limit the refs even when used with --all
928
929
+ Originally merged to 'next' on 2013-03-05
930
787
-* da/difftool-fixes (2013-02-21) 4 commits
788
- (merged to 'next' on 2013-02-25 at 687db1f)
789
- + t7800: "defaults" is no longer a builtin tool name
790
- + t7800: modernize tests
791
- + t7800: update copyright notice
792
- + difftool: silence uninitialized variable warning
931
+ The "--match=<pattern>" option of "git describe", when used with
932
+ "--all" to allow refs that are not annotated tags to be used as a
933
+ base of description, did not restrict the output from the command
934
+ to those that match the given pattern.
935
794
- Minor maintenance updates to difftool, and updates to its tests.
936
+ We may want to have a looser matching that does not restrict to tags,
937
+ but that can be done as a follow-up topic; this step is purely a bugfix.
938
796
- Will merge to 'master' in the 2nd batch (Safe).
939
+ Will merge to 'master' in the 3rd batch (Safe).
940
941
799
-* nd/read-directory-recursive-optim (2013-02-17) 1 commit
800
- (merged to 'next' on 2013-02-17 at 36ba9f4)
801
- + read_directory: avoid invoking exclude machinery on tracked files
942
+* tr/line-log (2013-03-21) 6 commits
943
+ - Speed up log -L... -M
944
+ - log -L: :pattern:file syntax to find by funcname
945
+ - Implement line-history search (git log -L)
946
+ - Export rewrite_parents() for 'log -L'
947
+ - fixup
948
+ - Refactor parse_loc
949
803
- "git status" has been optimized by taking advantage of the fact
804
- that paths that are already known to the index do not have to be
805
- checked against the .gitignore mechanism under some conditions.
806
- Even though I think this change is correct, please report
807
- immediately if you find any unintended side effect.
950
+ Rerolled; collides with nd/magic-pathspecs.
951
+ Does not seem to pass its own test (t4211-line-log.sh).
952
809
- Will merge to 'master' in the 2nd batch (Risky).
953
954
+* ap/maint-diff-rename-avoid-overlap (2013-03-06) 3 commits
955
+ (merged to 'next' on 2013-03-19 at c3276cf)
956
+ + tests: make sure rename pretty print works
957
+ + diff: prevent pprint_rename from underrunning input
958
+ + diff: Fix rename pretty-print when suffix and prefix overlap
959
812
-* mg/gpg-interface-using-status (2013-02-14) 5 commits
813
- (merged to 'next' on 2013-02-26 at 93f0e72)
814
- + pretty: make %GK output the signing key for signed commits
815
- + pretty: parse the gpg status lines rather than the output
816
- + gpg_interface: allow to request status return
817
- + log-tree: rely upon the check in the gpg_interface
818
- + gpg-interface: check good signature in a reliable way
960
+ Originally merged to 'next' on 2013-03-06
961
820
- Call "gpg" using the right API when validating the signature on
821
- tags.
962
+ The logic used by "git diff -M --stat" to shorten the names of
963
+ files before and after a rename did not work correctly when the
964
+ common prefix and suffix between the two filenames overlapped.
965
823
- Will merge to 'master' in the 2nd batch (Safe).
966
+ Will merge to 'master' in the 3rd batch (Safe).
967
968
826
-* jn/shell-disable-interactive (2013-03-09) 2 commits
827
- (merged to 'next' on 2013-03-14 at 3121f9f)
828
- + shell: new no-interactive-login command to print a custom message
829
- + shell doc: emphasize purpose and security model
969
+* jk/common-make-variables-export-safety (2013-02-25) 1 commit
970
+ - Makefile: make mandir, htmldir and infodir absolute
971
831
- Will merge to 'master' in the 2nd batch (Trivially Safe).
972
+ Originally merged to 'next' on 2013-03-05
973
974
+ Make the three variables safer to be exported to submakes by
975
+ ensuring that they are full paths so that they can be used as
976
+ installation location.
977
834
-* jc/fetch-raw-sha1 (2013-02-07) 4 commits
835
- (merged to 'next' on 2013-02-14 at ffa3c65)
836
- + fetch: fetch objects by their exact SHA-1 object names
837
- + upload-pack: optionally allow fetching from the tips of hidden refs
838
- + fetch: use struct ref to represent refs to be fetched
839
- + parse_fetch_refspec(): clarify the codeflow a bit
978
+ Even though I think this change is correct, please report
979
+ immediately if you find any unintended side effect.
980
841
- Allows requests to fetch objects at any tip of refs (including
842
- hidden ones). It seems that there may be use cases even outside
843
- Gerrit (e.g. $gmane/215701).
981
+ Will merge to 'master' in the 4th batch (Risky).
982
845
- Will merge to 'master' in the 2nd batch (Safe).
983
984
+* jk/suppress-clang-warning (2013-02-25) 1 commit
985
+ (merged to 'next' on 2013-03-19 at 1fd6858)
986
+ + fix clang -Wtautological-compare with unsigned enum
987
848
-* mn/send-email-works-with-credential (2013-02-27) 6 commits
849
- (merged to 'next' on 2013-02-27 at ee7ae0e)
850
- + git-send-email: use git credential to obtain password
851
- + Git.pm: add interface for git credential command
852
- + Git.pm: allow pipes to be closed prior to calling command_close_bidi_pipe
853
- + Git.pm: refactor command_close_bidi_pipe to use _cmd_close
854
- + Git.pm: fix example in command_close_bidi_pipe documentation
855
- + Git.pm: allow command_close_bidi_pipe to be called as method
988
+ Originally merged to 'next' on 2013-03-14
989
857
- Hooks the credential system to send-email.
990
+ Will merge to 'master' in the 3rd batch (Safe).
991
859
- Will merge to 'master' in the 2nd batch (Safe).
992
993
+* mg/unsigned-time-t (2013-02-25) 2 commits
994
+ (merged to 'next' on 2013-03-19 at 1fc1ad6)
995
+ + Fix time offset calculation in case of unsigned time_t
996
+ + date.c: fix unsigned time_t comparison
997
+
998
+ Originally merged to 'next' on 2013-03-14
999
862
-* nd/count-garbage (2013-02-15) 4 commits
863
- (merged to 'next' on 2013-02-17 at b2af923)
864
- + count-objects: report how much disk space taken by garbage files
865
- + count-objects: report garbage files in pack directory too
866
- + sha1_file: reorder code in prepare_packed_git_one()
867
- + git-count-objects.txt: describe each line in -v output
1000
+ A few workarounds for systems with unsigned time_t.
1001
869
- Will merge to 'master' in the 2nd batch (Safe).
1002
+ Will merge to 'master' in the 3rd batch (Safe).
1003
1004
872
-* tz/credential-authinfo (2013-02-25) 1 commit
873
- (merged to 'next' on 2013-02-27 at 7a261cb)
874
- + Add contrib/credentials/netrc with GPG support
1005
+* jk/pkt-line-cleanup (2013-03-21) 20 commits
1006
+ - do not use GIT_TRACE_PACKET=3 in tests
1007
+ - remote-curl: always parse incoming refs
1008
+ - remote-curl: move ref-parsing code up in file
1009
+ - remote-curl: pass buffer straight to get_remote_heads
1010
+ - teach get_remote_heads to read from a memory buffer
1011
+ - pkt-line: share buffer/descriptor reading implementation
1012
+ - pkt-line: provide a LARGE_PACKET_MAX static buffer
1013
+ - pkt-line: move LARGE_PACKET_MAX definition from sideband
1014
+ - pkt-line: teach packet_read_line to chomp newlines
1015
+ - pkt-line: provide a generic reading function with options
1016
+ - pkt-line: drop safe_write function
1017
+ - pkt-line: move a misplaced comment
1018
+ - write_or_die: raise SIGPIPE when we get EPIPE
1019
+ - upload-archive: use argv_array to store client arguments
1020
+ - upload-archive: do not copy repo name
1021
+ - send-pack: prefer prefixcmp over memcmp in receive_status
1022
+ - fetch-pack: fix out-of-bounds buffer offset in get_ack
1023
+ - upload-pack: remove packet debugging harness
1024
+ - upload-pack: do not add duplicate objects to shallow list
1025
+ - upload-pack: use get_sha1_hex to parse "shallow" lines
1026
+
1027
+ Originally merged to 'next' on 2013-02-25
1028
876
- A new read-only credential helper (in contrib/) to interact with
877
- the .netrc/.authinfo files. Hopefully mn/send-email-authinfo topic
878
- can rebuild on top of something like this.
1029
+ Cleans up pkt-line API, implementation and its callers to make them
1030
+ more robust. Even though I think this change is correct, please
1031
+ report immediately if you find any unintended side effect.
1032
880
- Will merge to 'master' in the 2nd batch (Safe).
1033
+ Will merge to 'master' in the 3rd batch (Risky).
1034
1035
1036
* jl/submodule-deinit (2013-03-04) 1 commit
884
- (merged to 'next' on 2013-03-05 at 097164e)
1037
+ (merged to 'next' on 2013-03-19 at d8367c5)
1038
+ submodule: add 'deinit' command
1039
1040
+ Originally merged to 'next' on 2013-03-05
1041
+
1042
There was no Porcelain way to say "I no longer am interested in
1043
this submodule", once you express your interest in a submodule with
1044
"submodule init". "submodule deinit" is the way to do so.
@@ -892,12 +1047,11 @@ of the repositories listed at
1047
1048
1049
* jc/remove-export-from-config-mak-in (2013-03-05) 3 commits
895
- (merged to 'next' on 2013-03-05 at abaa3cb)
896
- + Fix `make install` when configured with autoconf
897
- (merged to 'next' on 2013-02-12 at eb8af04)
898
- + Makefile: do not export mandir/htmldir/infodir
899
- (merged to 'next' on 2013-02-07 at 33f7d4f)
900
- + config.mak.in: remove unused definitions
1050
+ - Fix `make install` when configured with autoconf
1051
+ - Makefile: do not export mandir/htmldir/infodir
1052
+ - config.mak.in: remove unused definitions
1053
+
1054
+ Originally merged to 'next' on 2013-03-05
1055
1056
config.mak.in template had an "export" line to cause a few
1057
common makefile variables to be exported; if they need to be
@@ -915,8 +1069,9 @@ of the repositories listed at
1069
1070
1071
* jc/remove-treesame-parent-in-simplify-merges (2013-01-17) 1 commit
918
- (merged to 'next' on 2013-01-30 at b639b47)
919
- + simplify-merges: drop merge from irrelevant side branch
1072
+ - simplify-merges: drop merge from irrelevant side branch
1073
+
1074
+ Originally merged to 'next' on 2013-01-30
1075
1076
The --simplify-merges logic did not cull irrelevant parents from a
1077
merge that is otherwise not interesting with respect to the paths
@@ -929,42 +1084,44 @@ of the repositories listed at
1084
Will merge to 'master' in the 2nd batch (Risky).
1085
1086
932
-* jc/push-2.0-default-to-simple (2013-01-16) 14 commits
933
- (merged to 'next' on 2013-01-16 at 23f5df2)
934
- + t5570: do not assume the "matching" push is the default
935
- + t5551: do not assume the "matching" push is the default
936
- + t5550: do not assume the "matching" push is the default
937
- (merged to 'next' on 2013-01-09 at 74c3498)
938
- + doc: push.default is no longer "matching"
939
- + push: switch default from "matching" to "simple"
940
- + t9401: do not assume the "matching" push is the default
941
- + t9400: do not assume the "matching" push is the default
942
- + t7406: do not assume the "matching" push is the default
943
- + t5531: do not assume the "matching" push is the default
944
- + t5519: do not assume the "matching" push is the default
945
- + t5517: do not assume the "matching" push is the default
946
- + t5516: do not assume the "matching" push is the default
947
- + t5505: do not assume the "matching" push is the default
948
- + t5404: do not assume the "matching" push is the default
1087
+* jc/push-2.0-default-to-simple (2013-03-18) 15 commits
1088
+ - advice: Remove unused advice_push_non_ff_default
1089
+ - t5570: do not assume the "matching" push is the default
1090
+ - t5551: do not assume the "matching" push is the default
1091
+ - t5550: do not assume the "matching" push is the default
1092
+ - doc: push.default is no longer "matching"
1093
+ - push: switch default from "matching" to "simple"
1094
+ - t9401: do not assume the "matching" push is the default
1095
+ - t9400: do not assume the "matching" push is the default
1096
+ - t7406: do not assume the "matching" push is the default
1097
+ - t5531: do not assume the "matching" push is the default
1098
+ - t5519: do not assume the "matching" push is the default
1099
+ - t5517: do not assume the "matching" push is the default
1100
+ - t5516: do not assume the "matching" push is the default
1101
+ - t5505: do not assume the "matching" push is the default
1102
+ - t5404: do not assume the "matching" push is the default
1103
+
1104
+ Originally merged to 'next' on 2013-03-18
1105
1106
Will cook in 'next' until Git 2.0.
1107
1108
1109
* bc/append-signed-off-by (2013-02-23) 13 commits
954
- (merged to 'next' on 2013-02-25 at 32f7ac2)
955
- + git-commit: populate the edit buffer with 2 blank lines before s-o-b
956
- + Unify appending signoff in format-patch, commit and sequencer
957
- + format-patch: update append_signoff prototype
958
- + t4014: more tests about appending s-o-b lines
959
- + sequencer.c: teach append_signoff to avoid adding a duplicate newline
960
- + sequencer.c: teach append_signoff how to detect duplicate s-o-b
961
- + sequencer.c: always separate "(cherry picked from" from commit body
962
- + sequencer.c: require a conforming footer to be preceded by a blank line
963
- + sequencer.c: recognize "(cherry picked from ..." as part of s-o-b footer
964
- + t/t3511: add some tests of 'cherry-pick -s' functionality
965
- + t/test-lib-functions.sh: allow to specify the tag name to test_commit
966
- + commit, cherry-pick -s: remove broken support for multiline rfc2822 fields
967
- + sequencer.c: rework search for start of footer to improve clarity
1110
+ - git-commit: populate the edit buffer with 2 blank lines before s-o-b
1111
+ - Unify appending signoff in format-patch, commit and sequencer
1112
+ - format-patch: update append_signoff prototype
1113
+ - t4014: more tests about appending s-o-b lines
1114
+ - sequencer.c: teach append_signoff to avoid adding a duplicate newline
1115
+ - sequencer.c: teach append_signoff how to detect duplicate s-o-b
1116
+ - sequencer.c: always separate "(cherry picked from" from commit body
1117
+ - sequencer.c: require a conforming footer to be preceded by a blank line
1118
+ - sequencer.c: recognize "(cherry picked from ..." as part of s-o-b footer
1119
+ - t/t3511: add some tests of 'cherry-pick -s' functionality
1120
+ - t/test-lib-functions.sh: allow to specify the tag name to test_commit
1121
+ - commit, cherry-pick -s: remove broken support for multiline rfc2822 fields
1122
+ - sequencer.c: rework search for start of footer to improve clarity
1123
+
1124
+ Originally merged to 'next' on 2013-02-25
1125
1126
Consolidates codepaths that inspect log-message-to-be and decide to
1127
add a new Signed-off-by line in various commands. I think there is
@@ -976,26 +1133,26 @@ of the repositories listed at
1133
--------------------------------------------------
1134
[Discarded]
1135
979
-* pc/subtree-add-before-fetch (2013-02-28) 1 commit
980
- . contrib/subtree: allow addition of remote branch with name not locally present
981
-
982
- I'll leave subtree updates to the area maintainer.
983
-
1136
+* nd/read-directory-recursive-optim (2013-02-17) 1 commit
1137
+ . read_directory: avoid invoking exclude machinery on tracked files
1138
985
-* gp/forbid-describe-all-match (2013-02-24) 1 commit
986
- . describe: make --all and --match=PATTERN mutually incompatible
987
- (this branch uses gp/describe-match-uses-glob-pattern.)
1139
+ "git status" has been optimized by taking advantage of the fact
1140
+ that paths that are already known to the index do not have to be
1141
+ checked against the .gitignore mechanism under some conditions.
1142
+ Even though I think this change is correct, please report
1143
+ immediately if you find any unintended side effect.
1144
989
- jc/describe topic fixes the semantics of --match used with --all
990
- option.
1145
+ kb/status-ignored-optim may supersede this.
1146
1147
993
-* aw/setup-prefix-pathspec (2013-03-09) 2 commits
994
- (merged to 'next' on 2013-03-12 at 31a4a5e)
995
- + setup.c: check that the pathspec magic ends with ")"
996
- + setup.c: stop prefix_pathspec() from looping past the end of string
1148
+* jc/add-2.0-u-A-sans-pathspec (2013-03-14) 1 commit
1149
+ . git add: -u/-A now affects the entire working tree
1150
998
- "git cmd -- ':(top'" was not diagnosed as an invalid syntax, and
999
- instead the parser kept reading beyond the end of the string.
1151
+ Originally merged to 'next' on 2013-03-14
1152
1001
- Replaced by lf/setup-prefix-pathspec.
1153
+ "git add -u/-A" without pathspec has traditonally limited its
1154
+ operation to the current directory and its subdirectories, but in
1155
+ Git 1.8.2 we started encouraging users to be more explicit to
1156
+ specify "." when they mean it (and use ":/" to make it affect the
1157
+ entire working tree). With this, we finally change the behaviour
1158
+ and make it affect the entire working tree in Git 2.0.