What's cooking (2016/04 #05)
Junio C Hamano committed
Apr 18, 2016 at 15:38 UTC
b7f80a358cb93079b4b29f595e821ac13e474646
1 file changed
+462
-476
whats-cooking.txt
+462
-476
@@ -1,10 +1,10 @@
1
To: git@vger.kernel.org
2
Bcc: lwn@lwn.net
3
-Subject: What's cooking in git.git (Apr 2016, #04; Tue, 12)
4
-X-master-at: 7b0d47b3b6b5b64e02a5aa06b0452cadcdb18355
5
-X-next-at: f4fac6784293189f836875527464b8e7b234cb96
3
+Subject: What's cooking in git.git (Apr 2016, #05; Mon, 18)
4
+X-master-at: e6ac6e1f7d54584c2b03f073b5f329a37f4a9561
5
+X-next-at: c82888d429ec9129c13a7b821f0fb8d1c6d77bff
6
7
-What's cooking in git.git (Apr 2016, #04; Tue, 12)
7
+What's cooking in git.git (Apr 2016, #05; Mon, 18)
8
--------------------------------------------------
9
10
Here are the topics that have been cooking. Commits prefixed with
@@ -26,190 +26,287 @@ of the repositories listed at
26
--------------------------------------------------
27
[Graduated to "master"]
28
29
-* jc/merge-refuse-new-root (2016-03-23) 1 commit
30
- (merged to 'next' on 2016-04-04 at cd70fd6)
31
- + merge: refuse to create too cool a merge by default
29
+* ak/use-hashmap-iter-first-in-submodule-config (2016-03-23) 1 commit
30
+ (merged to 'next' on 2016-04-06 at 2aab890)
31
+ + submodule-config: use hashmap_iter_first()
32
33
- Originally merged to 'next' on 2016-03-23
33
+ Minor code cleanup.
34
35
- "git merge" used to allow merging two branches that have no common
36
- base by default, which led to a brand new history of an existing
37
- project created and then get pulled by an unsuspecting maintainer,
38
- which allowed an unnecessary parallel history merged into the
39
- existing project. The command has been taught not to allow this by
40
- default, with an escape hatch "--allow-unrelated-histories" option
41
- to be used in a rare event that merges histories of two projects
42
- that started their lives independently.
35
36
+* cc/apply (2016-04-01) 4 commits
37
+ (merged to 'next' on 2016-04-06 at 2e23c44)
38
+ + builtin/apply: free patch when parse_chunk() fails
39
+ + builtin/apply: handle parse_binary() failure
40
+ + apply: remove unused call to free() in gitdiff_{old,new}name()
41
+ + builtin/apply: get rid of useless 'name' variable
42
45
-* nd/apply-doc (2016-03-24) 2 commits
46
- (merged to 'next' on 2016-04-06 at f9bd355)
47
- + git-apply.txt: mention the behavior inside a subdir
48
- + git-apply.txt: remove a space
43
+ Minor code clean-up.
44
50
- A minor documentation update.
45
46
+* ep/trace-doc-sample-fix (2016-04-05) 1 commit
47
+ (merged to 'next' on 2016-04-06 at 0df7357)
48
+ + api-trace.txt: fix typo
49
+
50
+ Fix a typo in an example in the trace API documentation.
51
+
52
+
53
+* es/format-patch-doc-hide-no-patch (2016-04-04) 1 commit
54
+ (merged to 'next' on 2016-04-06 at 25d79bb)
55
+ + git-format-patch.txt: don't show -s as shorthand for multiple options
56
+
57
+ "git format-patch --help" showed `-s` and `--no-patch` as if these
58
+ are valid options to the command. We already hide `--patch` option
59
+ from the documentation, because format-patch is about showing the
60
+ diff, and the documentation now hides these options as well.
61
+
62
+
63
+* jc/makefile-redirection-stderr (2016-04-05) 1 commit
64
+ (merged to 'next' on 2016-04-06 at e3f2ded)
65
+ + Makefile: fix misdirected redirections
66
+
67
+ A minor fix in the Makefile.
68
+
69
+
70
+* jk/branch-shortening-funny-symrefs (2016-04-04) 1 commit
71
+ (merged to 'next' on 2016-04-06 at 1a3f8be)
72
+ + branch: fix shortening of non-remote symrefs
73
+
74
+ A change back in version 2.7 to "git branch" broke display of a
75
+ symbolic ref in a non-standard place in the refs/ hierarchy (we
76
+ expect symbolic refs to appear in refs/remotes/*/HEAD to point at
77
+ the primary branch the remote has, and as .git/HEAD to point at the
78
+ branch we locally checked out).
79
+
80
+ Will merge down to maint-2.7.
81
+
82
+
83
+* jk/check-repository-format (2016-03-11) 10 commits
84
+ (merged to 'next' on 2016-04-06 at a0dada0)
85
+ + verify_repository_format: mark messages for translation
86
+ + setup: drop repository_format_version global
87
+ + setup: unify repository version callbacks
88
+ + init: use setup.c's repo version verification
89
+ + setup: refactor repo format reading and verification
90
+ + config: drop git_config_early
91
+ + check_repository_format_gently: stop using git_config_early
92
+ + lazily load core.sharedrepository
93
+ + wrap shared_repository global in get/set accessors
94
+ + setup: document check_repository_format()
95
+ (this branch is used by dt/pre-refs-backend.)
96
+
97
+ The repository set-up sequence has been streamlined (the biggest
98
+ change is that there is no longer git_config_early()), so that we
99
+ do not attempt to look into refs/* when we know we do not have a
100
+ Git repository.
101
+
102
+
103
+* jn/mergetools-examdiff (2016-04-04) 2 commits
104
+ (merged to 'next' on 2016-04-06 at 819e858)
105
+ + mergetools: add support for ExamDiff
106
+ + mergetools: create mergetool_find_win32_cmd() helper function for winmerge
107
53
-* nd/apply-report-skip (2016-03-24) 1 commit
54
- (merged to 'next' on 2016-04-06 at ae2c824)
55
- + apply: report patch skipping in verbose mode
108
+ "git mergetools" learned to drive ExamDiff.
109
+
110
+
111
+* js/mingw-tests-2.8 (2016-04-04) 1 commit
112
+ (merged to 'next' on 2016-04-06 at f85a013)
113
+ + Windows: shorten code by re-using convert_slashes()
114
57
- "git apply -v" learned to report paths in the patch that were
58
- skipped via --include/--exclude mechanism or being outside the
59
- current working directory.
115
+ Code clean-up.
116
117
62
-* oa/doc-diff-check (2016-03-29) 1 commit
63
- (merged to 'next' on 2016-04-06 at e3d6e8d)
64
- + Documentation: git diff --check detects conflict markers
118
+* kn/for-each-tag-branch (2016-03-30) 1 commit
119
+ (merged to 'next' on 2016-04-06 at 4595ad3)
120
+ + for-each-ref: fix description of '--contains' in manpage
121
122
A minor documentation update.
123
124
69
-* pb/opt-cmdmode-doc (2016-03-25) 1 commit
70
- (merged to 'next' on 2016-04-06 at a5f3835)
71
- + api-parse-options.txt: document OPT_CMDMODE()
125
+* ky/branch-d-worktree (2016-03-29) 1 commit
126
+ (merged to 'next' on 2016-04-06 at 00f9bff)
127
+ + branch -d: refuse deleting a branch which is currently checked out
128
73
- Minor API documentation update.
129
+ When "git worktree" feature is in use, "git branch -d" allowed
130
+ deletion of a branch that is checked out in another worktree
131
132
76
-* ss/msvc (2016-03-30) 2 commits
77
- (merged to 'next' on 2016-04-06 at 4b53bce)
78
- + MSVC: use shipped headers instead of fallback definitions
79
- + MSVC: vsnprintf in Visual Studio 2015 doesn't need SNPRINTF_SIZE_CORR any more
133
+* ky/branch-m-worktree (2016-04-08) 3 commits
134
+ (merged to 'next' on 2016-04-08 at b673b5e)
135
+ + set_worktree_head_symref(): fix error message
136
+ (merged to 'next' on 2016-04-06 at e7b285c)
137
+ + branch -m: update all per-worktree HEADs
138
+ + refs: add a new function set_worktree_head_symref
139
81
- Build updates for MSVC.
140
+ When "git worktree" feature is in use, "git branch -m" renamed a
141
+ branch that is checked out in another worktree without adjusting
142
+ the HEAD symbolic ref for the worktree.
143
83
---------------------------------------------------
84
-[New Topics]
144
86
-* ad/commit-have-m-option (2016-04-07) 2 commits
87
- - commit: do not ignore an empty message given by -m ''
88
- - commit: --amend -m '' silently fails to wipe message
145
+* lt/pretty-expand-tabs (2016-04-04) 4 commits
146
+ (merged to 'next' on 2016-04-06 at 186ac2a)
147
+ + pretty: test --expand-tabs
148
+ + pretty: allow tweaking tabwidth in --expand-tabs
149
+ + pretty: enable --expand-tabs by default for selected pretty formats
150
+ + pretty: expand tabs in indented logs to make things line up properly
151
90
- "git commit" misbehaved in a few minor ways when an empty message
91
- is given via -m '', all of which has been corrected.
152
+ When "git log" shows the log message indented by 4-spaces, the
153
+ remainder of a line after a HT does not align in the way the author
154
+ originally intended. The command now expands tabs by default in
155
+ such a case, and allows the users to override it with a new option,
156
+ '--no-expand-tabs'.
157
93
- Will merge to 'next'.
158
159
+* mg/complete-cherry-mark-to-log (2016-04-05) 1 commit
160
+ (merged to 'next' on 2016-04-06 at 3002be6)
161
+ + completion: complete --cherry-mark for git log
162
96
-* jc/xstrfmt-null-with-prec-0 (2016-04-07) 1 commit
97
- - setup.c: do not feed NULL to "%.*s" even with precision 0
163
+ The completion scripts (in contrib/) did not include the
164
+ "--cherry-mark" option when completing "git log <HT>".
165
99
- Will merge to 'next'.
166
167
+* mj/pull-rebase-autostash (2016-04-04) 9 commits
168
+ (merged to 'next' on 2016-04-06 at b4e4f31)
169
+ + t5520: test --[no-]autostash with pull.rebase=true
170
+ + t5520: reduce commom lines of code
171
+ + t5520: factor out common "failing autostash" code
172
+ + t5520: factor out common "successful autostash" code
173
+ + t5520: use better test to check stderr output
174
+ + t5520: ensure consistent test conditions
175
+ + t5520: use consistent capitalization in test titles
176
+ + pull --rebase: add --[no-]autostash flag
177
+ + git-pull.c: introduce git_pull_config()
178
102
-* dt/pre-refs-backend (2016-04-10) 24 commits
103
- - refs: on symref reflog expire, lock symref not referrent
104
- - refs: move resolve_ref_unsafe into common code
105
- - show_head_ref(): check the result of resolve_ref_namespace()
106
- - check_aliased_update(): check that dst_name is non-NULL
107
- - checkout_paths(): remove unneeded flag variable
108
- - cmd_merge(): remove unneeded flag variable
109
- - fsck_head_link(): remove unneeded flag variable
110
- - read_raw_ref(): change flags parameter to unsigned int
111
- - files-backend: inline resolve_ref_1() into resolve_ref_unsafe()
112
- - read_raw_ref(): manage own scratch space
113
- - files-backend: break out ref reading
114
- - resolve_ref_1(): eliminate local variable "bad_name"
115
- - resolve_ref_1(): reorder code
116
- - resolve_ref_1(): eliminate local variable
117
- - resolve_ref_unsafe(): ensure flags is always set
118
- - resolve_ref_unsafe(): use for loop to count up to MAXDEPTH
119
- - resolve_missing_loose_ref(): simplify semantics
120
- - t1430: improve test coverage of deletion of badly-named refs
121
- - t1430: test for-each-ref in the presence of badly-named refs
122
- - t1430: don't rely on symbolic-ref for creating broken symrefs
123
- - t1430: clean up broken refs/tags/shadow
124
- - t1430: test the output and error of some commands more carefully
125
- - refs: move for_each_*ref* functions into common code
126
- - refs: move head_ref{,_submodule} to the common code
127
- (this branch uses jk/check-repository-format.)
179
+ "git pull --rebase" learned "--[no-]autostash" option, so that
180
+ the rebase.autostash configuration variable set to true can be
181
+ overridden from the command line.
182
129
- Code restructuring around the "refs" area to prepare for pluggable
130
- refs backends.
183
132
- Will merge to 'next'.
184
+* rt/completion-help (2016-03-24) 2 commits
185
+ (merged to 'next' on 2016-04-06 at 8c3ee08)
186
+ + completion: add 'revisions' and 'everyday' to 'git help'
187
+ + completion: add option '--guides' to 'git help'
188
189
+ Shell completion (in contrib/) updates.
190
135
-* ky/imap-send (2016-04-08) 2 commits
136
- - imap-send: fix CRAM-MD5 response calculation
137
- - imap-send: check for NOLOGIN capability only when using LOGIN command
191
139
- Support for CRAM-MD5 authentication method in "git imap-send" did
140
- not work well.
192
+* rt/rebase-i-shorten-stop-report (2016-03-28) 1 commit
193
+ (merged to 'next' on 2016-04-06 at 7a766b7)
194
+ + rebase-i: print an abbreviated hash when stop for editing
195
142
- Will merge to 'next'.
196
+ The commit object name reported when "rebase -i" stops has been
197
+ shortened.
198
199
145
-* ky/imap-send-openssl-1.1.0 (2016-04-08) 4 commits
146
- - configure: remove checking for HMAC_CTX_cleanup
147
- - imap-send: avoid deprecated TLSv1_method()
148
- - imap-send: check NULL return of SSL_CTX_new()
149
- - imap-send: use HMAC() function provided by OpenSSL
200
+* rz/worktree-no-checkout (2016-03-29) 1 commit
201
+ (merged to 'next' on 2016-04-06 at e725216)
202
+ + worktree: add: introduce --checkout option
203
151
- Upcoming OpenSSL 1.1.0 will break compilation b updating a few APIs
152
- we use in imap-send, which has been adjusted for the change.
204
+ "git worktree add" can be given "--no-checkout" option to only
205
+ create an empty worktree without checking out the files.
206
154
- Will merge to 'next'.
207
208
+* sb/misc-cleanups (2016-04-01) 4 commits
209
+ (merged to 'next' on 2016-04-06 at 4e63691)
210
+ + credential-cache, send_request: close fd when done
211
+ + bundle: don't leak an fd in case of early return
212
+ + abbrev_sha1_in_line: don't leak memory
213
+ + notes: don't leak memory in git_config_get_notes_strategy
214
157
-* jc/http-socks5h (2016-04-10) 1 commit
158
- - http: differentiate socks5:// and socks5h://
215
+ Assorted minor clean-ups.
216
160
- The socks5:// proxy support added back in 2.6.4 days was not aware
161
- that socks5h:// proxies behave differently.
217
163
- Will merge to 'next'.
218
+* sg/diff-multiple-identical-renames (2016-03-30) 1 commit
219
+ (merged to 'next' on 2016-04-06 at ac19e48)
220
+ + diffcore: fix iteration order of identical files during rename detection
221
222
+ "git diff -M" used to work better when two originally identical
223
+ files A and B got renamed to X/A and X/B by pairing A to X/A and B
224
+ to X/B, but this was broken in the 2.0 timeframe.
225
166
-* jc/send-email-skip-backup (2016-04-12) 1 commit
167
- - send-email: detect and offer to skip backup files
226
169
- A careless invocation of "git send-email directory/" after editing
170
- 0001-change.patch with an editor often ends up sending both
171
- 0001-change.patch and its backup file, 0001-change.patch~, causing
172
- embarrassment and a minor confusion. Detect such an input and
173
- offer to skip the backup files when sending the patches out.
227
+* sk/send-pack-all-fix (2016-03-31) 1 commit
228
+ (merged to 'next' on 2016-04-06 at 31e1e1b)
229
+ + git-send-pack: fix --all option when used with directory
230
175
- Needs review.
231
+ "git send-pack --all <there>" was broken when its command line
232
+ option parsing was written in the 2.6 timeframe.
233
234
178
-* jk/do-not-printf-NULL (2016-04-10) 3 commits
179
- - git_config_set_multivar_in_file: handle "unset" errors
180
- - git_config_set_multivar_in_file: all non-zero returns are errors
181
- - config: lower-case first word of error strings
235
+* tb/blame-force-read-cache-to-workaround-safe-crlf (2016-04-05) 1 commit
236
+ (merged to 'next' on 2016-04-06 at 263bba8)
237
+ + correct blame for files commited with CRLF
238
183
- "git config" had a codepath that tried to pass a NULL to
184
- printf("%s"), which nobody seems to have noticed.
239
+ When running "git blame $path" with unnormalized data in the index
240
+ for the path, the data in the working tree was blamed, even though
241
+ "git add" would not have changed what is already in the index, due
242
+ to "safe crlf" that disables the line-end conversion. It has been
243
+ corrected.
244
+
245
+--------------------------------------------------
246
+[New Topics]
247
+
248
+* en/merge-fixes (2016-04-12) 2 commits
249
+ - merge-recursive: do not check working copy when creating a virtual merge base
250
+ - merge-recursive: remove duplicate code
251
+
252
+ "merge-recursive" strategy incorrectly checked if a path that is
253
+ involved in its internal merge exists in the working tree.
254
255
Will merge to 'next'.
256
257
189
-* jk/use-write-script-more (2016-04-12) 3 commits
190
- - t3404: use write_script
191
- - t1020: do not overuse printf and use write_script
192
- - t5532: use write_script
258
+* en/merge-octopus-fix (2016-04-12) 2 commits
259
+ (merged to 'next' on 2016-04-13 at 600b479)
260
+ + merge-octopus: abort if index does not match HEAD
261
+ + t6044: new merge testcases for when index doesn't match HEAD
262
194
- Code clean-up.
263
+ "merge-octopus" strategy did not ensure that the index is clean
264
+ when merge begins.
265
196
- Will merge to 'next'.
266
+ Will merge to 'master'.
267
268
199
-* nf/mergetool-prompt (2016-04-12) 2 commits
200
- - SQUASH???
201
- - difftool/mergetool: make the form of yes/no questions consistent
269
+* en/merge-trivial-fix (2016-04-12) 2 commits
270
+ (merged to 'next' on 2016-04-13 at fb3ea86)
271
+ + builtin/merge.c: fix a bug with trivial merges
272
+ + t7605: add a testcase demonstrating a bug with trivial merges
273
203
- UI consistency improvements.
274
+ When "git merge" notices that the merge can be resolved purely at
275
+ the tree level (without having to merge blobs) and the resulting
276
+ tree happens to already exist in the object store, it forgot to
277
+ update the index, which lead to an inconsistent state for later
278
+ operations.
279
280
+ Will merge to 'master'.
281
206
-* va/i18n-misc-updates (2016-04-12) 4 commits
207
- - i18n: builtin/rm.c: remove a comma ',' from string
208
- - i18n: unpack-trees: mark strings for translation
209
- - i18n: builtin/branch.c: mark option for translation
210
- - i18n: index-pack: use plural string instead of normal one
282
212
- Mark several messages for translation.
283
+* jc/fsck-nul-in-commit (2016-04-14) 2 commits
284
+ - fsck: detect and warn a commit with embedded NUL
285
+ - fsck_commit_buffer(): do not special case the last validation
286
+
287
+ "git fsck" learned to catch NUL byte in a commit object as
288
+ potential error and warn.
289
+
290
+
291
+* jc/ll-merge-internal (2016-04-14) 2 commits
292
+ - ll-merge: use a longer conflict marker for internal merge
293
+ - ll-merge: fix typo in comment
294
+
295
+ RFC.
296
+
297
+
298
+* jk/diff-compact-heuristic (2016-04-18) 2 commits
299
+ - xdiff: implement empty line chunk heuristic
300
+ - xdiff: add recs_match helper function
301
+
302
+
303
+* nd/test-helpers (2016-04-15) 2 commits
304
+ - test helpers: move test-* to t/helper/ subdirectory
305
+ - Makefile: clean *.o files we create
306
+
307
+ Sources to many test helper binaries (and the generated helpers)
308
+ have been moved to t/helper/ subdirectory to reduce clutter at the
309
+ top level of the tree.
310
311
--------------------------------------------------
312
[Stalled]
@@ -249,16 +346,6 @@ of the repositories listed at
346
($gmane/290340)
347
348
252
-* sb/clone-shallow-passthru (2016-03-23) 3 commits
253
- - clone: add t5614 to test cloning submodules with shallowness involved
254
- - submodule clone: pass along `local` option
255
- - clone: add `--shallow-submodules` flag
256
-
257
- "git clone" learned "--shallow-submodules" option.
258
-
259
- Needs review.
260
-
261
-
349
* sg/completion-updates (2016-02-28) 21 commits
350
- completion: cache the path to the repository
351
- completion: extract repository discovery from __gitdir()
@@ -286,53 +373,6 @@ of the repositories listed at
373
($gmane/287839)
374
375
289
-* nd/shallow-deepen (2016-02-23) 25 commits
290
- - fetch, upload-pack: --deepen=N extends shallow boundary by N commits
291
- - upload-pack: add get_reachable_list()
292
- - upload-pack: split check_unreachable() in two, prep for get_reachable_list()
293
- - t5500, t5539: tests for shallow depth excluding a ref
294
- - clone: define shallow clone boundary with --shallow-exclude
295
- - fetch: define shallow boundary with --shallow-exclude
296
- - upload-pack: support define shallow boundary by excluding revisions
297
- - refs: add expand_ref()
298
- - t5500, t5539: tests for shallow depth since a specific date
299
- - clone: define shallow clone boundary based on time with --shallow-since
300
- - fetch: define shallow boundary with --shallow-since
301
- - upload-pack: add deepen-since to cut shallow repos based on time
302
- - shallow.c: implement a generic shallow boundary finder based on rev-list
303
- - fetch-pack: use a separate flag for fetch in deepening mode
304
- - fetch-pack: use a common function for verbose printing
305
- - fetch-pack: use skip_prefix() instead of starts_with()
306
- - upload-pack: move rev-list code out of check_non_tip()
307
- - upload-pack: tighten number parsing at "deepen" lines
308
- - upload-pack: use skip_prefix() instead of starts_with()
309
- - upload-pack: move "unshallow" sending code out of deepen()
310
- - upload-pack: remove unused variable "backup"
311
- - upload-pack: move "shallow" sending code out of deepen()
312
- - upload-pack: move shallow deepen code out of receive_needs()
313
- - transport-helper.c: refactor set_helper_option()
314
- - remote-curl.c: convert fetch_git() to use argv_array
315
-
316
- The existing "git fetch --depth=<n>" option was hard to use
317
- correctly when making the history of an existing shallow clone
318
- deeper. A new option, "--deepen=<n>", has been added to make this
319
- easier to use. "git clone" also learned "--shallow-since=<date>"
320
- and "--shallow-exclude=<tag>" options to make it easier to specify
321
- "I am interested only in the recent N months worth of history" and
322
- "Give me only the history since that version".
323
-
324
- Needs review.
325
-
326
-
327
-* sb/submodule-init (2016-03-15) 2 commits
328
- . submodule: port init from shell to C
329
- . submodule: port resolve_relative_url from shell to C
330
-
331
- Update of "git submodule" to move pieces of logic to C continues.
332
-
333
- Needs to be adjusted for the recent "prefix" fixes.
334
-
335
-
376
* az/p4-bare-no-rebase (2016-02-19) 1 commit
377
- git-p4.py: Don't try to rebase on submit from bare repository
378
@@ -444,31 +484,210 @@ of the repositories listed at
484
--------------------------------------------------
485
[Cooking]
486
447
-* jc/drop-git-spec-in (2016-04-06) 1 commit
448
- - Makefile: stop pretending to support rpmbuild
487
+* sb/submodule-init (2016-04-16) 4 commits
488
+ - submodule: port init from shell to C
489
+ - submodule: port resolve_relative_url from shell to C
490
+ - Merge branch 'sb/submodule-path-misc-bugs' into sb/submodule-init
491
+ - Merge branch 'sb/submodule-helper-clone-regression-fix' into sb/submodule-init
492
+ (this branch uses sb/submodule-helper-clone-regression-fix and sb/submodule-path-misc-bugs.)
493
450
- As nobody maintains our in-tree git.spec.in and distros use their
451
- own spec file, we stopped pretending that we support "make rpm".
494
+ Update of "git submodule" to move pieces of logic to C continues.
495
453
- Comments?
496
497
+* nd/shallow-deepen (2016-04-13) 26 commits
498
+ - fetch, upload-pack: --deepen=N extends shallow boundary by N commits
499
+ - upload-pack: add get_reachable_list()
500
+ - upload-pack: split check_unreachable() in two, prep for get_reachable_list()
501
+ - t5500, t5539: tests for shallow depth excluding a ref
502
+ - clone: define shallow clone boundary with --shallow-exclude
503
+ - fetch: define shallow boundary with --shallow-exclude
504
+ - upload-pack: support define shallow boundary by excluding revisions
505
+ - refs: add expand_ref()
506
+ - t5500, t5539: tests for shallow depth since a specific date
507
+ - clone: define shallow clone boundary based on time with --shallow-since
508
+ - fetch: define shallow boundary with --shallow-since
509
+ - upload-pack: add deepen-since to cut shallow repos based on time
510
+ - shallow.c: implement a generic shallow boundary finder based on rev-list
511
+ - fetch-pack: use a separate flag for fetch in deepening mode
512
+ - fetch-pack.c: mark strings for translating
513
+ - fetch-pack: use a common function for verbose printing
514
+ - fetch-pack: use skip_prefix() instead of starts_with()
515
+ - upload-pack: move rev-list code out of check_non_tip()
516
+ - upload-pack: tighten number parsing at "deepen" lines
517
+ - upload-pack: use skip_prefix() instead of starts_with()
518
+ - upload-pack: move "unshallow" sending code out of deepen()
519
+ - upload-pack: remove unused variable "backup"
520
+ - upload-pack: move "shallow" sending code out of deepen()
521
+ - upload-pack: move shallow deepen code out of receive_needs()
522
+ - transport-helper.c: refactor set_helper_option()
523
+ - remote-curl.c: convert fetch_git() to use argv_array
524
456
-* jc/makefile-redirection-stderr (2016-04-05) 1 commit
457
- (merged to 'next' on 2016-04-06 at e3f2ded)
458
- + Makefile: fix misdirected redirections
525
+ The existing "git fetch --depth=<n>" option was hard to use
526
+ correctly when making the history of an existing shallow clone
527
+ deeper. A new option, "--deepen=<n>", has been added to make this
528
+ easier to use. "git clone" also learned "--shallow-since=<date>"
529
+ and "--shallow-exclude=<tag>" options to make it easier to specify
530
+ "I am interested only in the recent N months worth of history" and
531
+ "Give me only the history since that version".
532
460
- A minor fix in the Makefile.
533
+ Needs review.
534
+
535
+
536
+* sb/clone-shallow-passthru (2016-04-13) 3 commits
537
+ - clone: add t5614 to test cloning submodules with shallowness involved
538
+ - clone: add `--shallow-submodules` flag
539
+ - submodule clone: pass along `local` option
540
+
541
+ "git clone" learned "--shallow-submodules" option.
542
+
543
+ Needs review.
544
+
545
+
546
+* ad/commit-have-m-option (2016-04-07) 2 commits
547
+ (merged to 'next' on 2016-04-13 at 74088c2)
548
+ + commit: do not ignore an empty message given by -m ''
549
+ + commit: --amend -m '' silently fails to wipe message
550
+
551
+ "git commit" misbehaved in a few minor ways when an empty message
552
+ is given via -m '', all of which has been corrected.
553
+
554
+ Will merge to 'master'.
555
+
556
+
557
+* jc/xstrfmt-null-with-prec-0 (2016-04-07) 1 commit
558
+ (merged to 'next' on 2016-04-13 at 2457462)
559
+ + setup.c: do not feed NULL to "%.*s" even with precision 0
560
+
561
+ Will merge to 'master'.
562
+
563
+
564
+* dt/pre-refs-backend (2016-04-10) 24 commits
565
+ (merged to 'next' on 2016-04-13 at 0a8f9dd)
566
+ + refs: on symref reflog expire, lock symref not referrent
567
+ + refs: move resolve_ref_unsafe into common code
568
+ + show_head_ref(): check the result of resolve_ref_namespace()
569
+ + check_aliased_update(): check that dst_name is non-NULL
570
+ + checkout_paths(): remove unneeded flag variable
571
+ + cmd_merge(): remove unneeded flag variable
572
+ + fsck_head_link(): remove unneeded flag variable
573
+ + read_raw_ref(): change flags parameter to unsigned int
574
+ + files-backend: inline resolve_ref_1() into resolve_ref_unsafe()
575
+ + read_raw_ref(): manage own scratch space
576
+ + files-backend: break out ref reading
577
+ + resolve_ref_1(): eliminate local variable "bad_name"
578
+ + resolve_ref_1(): reorder code
579
+ + resolve_ref_1(): eliminate local variable
580
+ + resolve_ref_unsafe(): ensure flags is always set
581
+ + resolve_ref_unsafe(): use for loop to count up to MAXDEPTH
582
+ + resolve_missing_loose_ref(): simplify semantics
583
+ + t1430: improve test coverage of deletion of badly-named refs
584
+ + t1430: test for-each-ref in the presence of badly-named refs
585
+ + t1430: don't rely on symbolic-ref for creating broken symrefs
586
+ + t1430: clean up broken refs/tags/shadow
587
+ + t1430: test the output and error of some commands more carefully
588
+ + refs: move for_each_*ref* functions into common code
589
+ + refs: move head_ref{,_submodule} to the common code
590
+
591
+ Code restructuring around the "refs" area to prepare for pluggable
592
+ refs backends.
593
+
594
+ Will merge to 'master'.
595
+
596
+
597
+* ky/imap-send (2016-04-13) 2 commits
598
+ (merged to 'next' on 2016-04-13 at 52cf493)
599
+ + imap-send: fix CRAM-MD5 response calculation
600
+ + imap-send: check for NOLOGIN capability only when using LOGIN command
601
+
602
+ Support for CRAM-MD5 authentication method in "git imap-send" did
603
+ not work well.
604
+
605
+ Will merge to 'master'.
606
+
607
+
608
+* ky/imap-send-openssl-1.1.0 (2016-04-08) 4 commits
609
+ (merged to 'next' on 2016-04-13 at 49d2643)
610
+ + configure: remove checking for HMAC_CTX_cleanup
611
+ + imap-send: avoid deprecated TLSv1_method()
612
+ + imap-send: check NULL return of SSL_CTX_new()
613
+ + imap-send: use HMAC() function provided by OpenSSL
614
+
615
+ Upcoming OpenSSL 1.1.0 will break compilation b updating a few APIs
616
+ we use in imap-send, which has been adjusted for the change.
617
+
618
+ Will merge to 'master'.
619
+
620
+
621
+* jc/http-socks5h (2016-04-10) 1 commit
622
+ (merged to 'next' on 2016-04-13 at eb27afc)
623
+ + http: differentiate socks5:// and socks5h://
624
+
625
+ The socks5:// proxy support added back in 2.6.4 days was not aware
626
+ that socks5h:// proxies behave differently.
627
+
628
+ Will merge to 'master'.
629
+
630
+
631
+* jc/send-email-skip-backup (2016-04-12) 1 commit
632
+ - send-email: detect and offer to skip backup files
633
+
634
+ A careless invocation of "git send-email directory/" after editing
635
+ 0001-change.patch with an editor often ends up sending both
636
+ 0001-change.patch and its backup file, 0001-change.patch~, causing
637
+ embarrassment and a minor confusion. Detect such an input and
638
+ offer to skip the backup files when sending the patches out.
639
+
640
+ Needs review.
641
+
642
+
643
+* jk/do-not-printf-NULL (2016-04-10) 3 commits
644
+ (merged to 'next' on 2016-04-13 at 60912e3)
645
+ + git_config_set_multivar_in_file: handle "unset" errors
646
+ + git_config_set_multivar_in_file: all non-zero returns are errors
647
+ + config: lower-case first word of error strings
648
+
649
+ "git config" had a codepath that tried to pass a NULL to
650
+ printf("%s"), which nobody seems to have noticed.
651
+
652
+ Will merge to 'master'.
653
+
654
+
655
+* jk/use-write-script-more (2016-04-12) 3 commits
656
+ (merged to 'next' on 2016-04-13 at d6718bf)
657
+ + t3404: use write_script
658
+ + t1020: do not overuse printf and use write_script
659
+ + t5532: use write_script
660
+
661
+ Code clean-up.
662
+
663
+ Will merge to 'master'.
664
+
665
+
666
+* nf/mergetool-prompt (2016-04-12) 2 commits
667
+ - SQUASH???
668
+ - difftool/mergetool: make the form of yes/no questions consistent
669
+
670
+ UI consistency improvements.
671
+
672
+
673
+* va/i18n-misc-updates (2016-04-13) 6 commits
674
+ - i18n: branch: move comment for translators
675
+ - i18n: branch: unmark string for translation
676
+ - i18n: builtin/rm.c: remove a comma ',' from string
677
+ - i18n: unpack-trees: mark strings for translation
678
+ - i18n: builtin/branch.c: mark option for translation
679
+ - i18n: index-pack: use plural string instead of normal one
680
462
- Will merge to 'master'.
681
+ Mark several messages for translation.
682
683
465
-* js/mingw-tests-2.8 (2016-04-04) 1 commit
466
- (merged to 'next' on 2016-04-06 at f85a013)
467
- + Windows: shorten code by re-using convert_slashes()
684
+* jc/drop-git-spec-in (2016-04-06) 1 commit
685
+ - Makefile: stop pretending to support rpmbuild
686
469
- Code clean-up.
687
+ As nobody maintains our in-tree git.spec.in and distros use their
688
+ own spec file, we stopped pretending that we support "make rpm".
689
471
- Will merge to 'master'.
690
+ Comments?
691
692
693
* ar/diff-args-osx-precompose (2016-04-05) 1 commit
@@ -483,105 +702,34 @@ of the repositories listed at
702
($gmane/290724)
703
704
486
-* ep/trace-doc-sample-fix (2016-04-05) 1 commit
487
- (merged to 'next' on 2016-04-06 at 0df7357)
488
- + api-trace.txt: fix typo
489
-
490
- Fix a typo in an example in the trace API documentation.
491
-
492
- Will merge to 'master'.
493
-
494
-
705
* ew/send-email-readable-message-id (2016-04-06) 1 commit
496
- - send-email: more meaningful Message-ID
706
+ (merged to 'next' on 2016-04-13 at 422959a)
707
+ + send-email: more meaningful Message-ID
708
709
"git send-email" now uses a more readable timestamps when
710
formulating a message ID.
711
501
- Will merge to 'next'.
502
-
503
-
504
-* mg/complete-cherry-mark-to-log (2016-04-05) 1 commit
505
- (merged to 'next' on 2016-04-06 at 3002be6)
506
- + completion: complete --cherry-mark for git log
507
-
508
- The completion scripts (in contrib/) did not include the
509
- "--cherry-mark" option when completing "git log <HT>".
510
-
511
- Will merge to 'master'.
512
-
513
-
514
-* tb/blame-force-read-cache-to-workaround-safe-crlf (2016-04-05) 1 commit
515
- (merged to 'next' on 2016-04-06 at 263bba8)
516
- + correct blame for files commited with CRLF
517
-
518
- When running "git blame $path" with unnormalized data in the index
519
- for the path, the data in the working tree was blamed, even though
520
- "git add" would not have changed what is already in the index, due
521
- to "safe crlf" that disables the line-end conversion. It has been
522
- corrected.
523
-
712
Will merge to 'master'.
713
714
527
-* st/verify-tag (2016-04-06) 3 commits
715
+* st/verify-tag (2016-04-18) 6 commits
716
+ - tag -v: verfy directly rather than exec-ing verify-tag
717
+ - verify-tag: move verification code to tag.c
718
+ - verify-tag: add sha1 argument to verify_tag()
719
- verify-tag: change variable name for readability
720
- t7030: test verifying multiple tags
721
- builtin/verify-tag.c: ignore SIGPIPE in gpg-interface
722
532
- Only the first three patches in a six-patch series.
533
- Needs further work.
723
+ Saw review comments on this v6.
724
+ Expecting a reroll.
725
726
727
* ew/send-email-drop-data-dumper (2016-04-06) 1 commit
537
- - send-email: do not load Data::Dumper
728
+ (merged to 'next' on 2016-04-13 at 180266c)
729
+ + send-email: do not load Data::Dumper
730
731
Code clean-up.
732
541
- Will merge to 'next'.
542
-
543
-
544
-* es/format-patch-doc-hide-no-patch (2016-04-04) 1 commit
545
- (merged to 'next' on 2016-04-06 at 25d79bb)
546
- + git-format-patch.txt: don't show -s as shorthand for multiple options
547
-
548
- "git format-patch --help" showed `-s` and `--no-patch` as if these
549
- are valid options to the command. We already hide `--patch` option
550
- from the documentation, because format-patch is about showing the
551
- diff, and the documentation now hides these options as well.
552
-
553
- Will merge to 'master'.
554
-
555
-
556
-* jk/branch-shortening-funny-symrefs (2016-04-04) 1 commit
557
- (merged to 'next' on 2016-04-06 at 1a3f8be)
558
- + branch: fix shortening of non-remote symrefs
559
-
560
- A change back in version 2.7 to "git branch" broke display of a
561
- symbolic ref in a non-standard place in the refs/ hierarchy (we
562
- expect symbolic refs to appear in refs/remotes/*/HEAD to point at
563
- the primary branch the remote has, and as .git/HEAD to point at the
564
- branch we locally checked out).
565
-
566
- Will merge to 'next' and later down to maint-2.7.
567
-
568
-
569
-* jn/mergetools-examdiff (2016-04-04) 2 commits
570
- (merged to 'next' on 2016-04-06 at 819e858)
571
- + mergetools: add support for ExamDiff
572
- + mergetools: create mergetool_find_win32_cmd() helper function for winmerge
573
-
574
- "git mergetools" learned to drive ExamDiff.
575
-
576
- Will merge to 'master'.
577
-
578
-
579
-* kn/for-each-tag-branch (2016-03-30) 1 commit
580
- (merged to 'next' on 2016-04-06 at 4595ad3)
581
- + for-each-ref: fix description of '--contains' in manpage
582
-
583
- A minor documentation update.
584
-
733
Will merge to 'master'.
734
735
@@ -610,77 +758,38 @@ of the repositories listed at
758
$gmane/291295 yet.
759
760
613
-* rz/worktree-no-checkout (2016-03-29) 1 commit
614
- (merged to 'next' on 2016-04-06 at e725216)
615
- + worktree: add: introduce --checkout option
616
-
617
- "git worktree add" can be given "--no-checkout" option to only
618
- create an empty worktree without checking out the files.
619
-
620
- Will merge to 'master'.
621
-
622
-
623
-* sb/misc-cleanups (2016-04-01) 4 commits
624
- (merged to 'next' on 2016-04-06 at 4e63691)
625
- + credential-cache, send_request: close fd when done
626
- + bundle: don't leak an fd in case of early return
627
- + abbrev_sha1_in_line: don't leak memory
628
- + notes: don't leak memory in git_config_get_notes_strategy
629
-
630
- Assorted minor clean-ups.
631
-
632
- Will merge to 'master'.
633
-
634
-
761
* sb/submodule-helper-clone-regression-fix (2016-04-01) 6 commits
636
- - submodule--helper, module_clone: catch fprintf failure
637
- - submodule--helper: do not borrow absolute_path() result for too long
638
- - submodule--helper, module_clone: always operate on absolute paths
639
- - submodule--helper clone: create the submodule path just once
640
- - submodule--helper: fix potential NULL-dereference
641
- - recursive submodules: test for relative paths
762
+ (merged to 'next' on 2016-04-13 at c6584bb)
763
+ + submodule--helper, module_clone: catch fprintf failure
764
+ + submodule--helper: do not borrow absolute_path() result for too long
765
+ + submodule--helper, module_clone: always operate on absolute paths
766
+ + submodule--helper clone: create the submodule path just once
767
+ + submodule--helper: fix potential NULL-dereference
768
+ + recursive submodules: test for relative paths
769
+ (this branch is used by sb/submodule-init.)
770
771
A partial rewrite of "git submodule" in the 2.7 timeframe changed
772
the way the gitdir: pointer in the submodules point at the real
773
repository location to use absolute paths by accident. This has
774
been corrected.
775
648
- Will merge to 'next'.
776
+ Will merge to 'master'.
777
778
779
* sb/submodule-path-misc-bugs (2016-03-30) 6 commits
652
- - t7407: make expectation as clear as possible
653
- - submodule update: test recursive path reporting from subdirectory
654
- - submodule update: align reporting path for custom command execution
655
- - submodule status: correct path handling in recursive submodules
656
- - submodule update --init: correct path handling in recursive submodules
657
- - submodule foreach: correct path display in recursive submodules
780
+ (merged to 'next' on 2016-04-18 at 9daa5ce)
781
+ + t7407: make expectation as clear as possible
782
+ + submodule update: test recursive path reporting from subdirectory
783
+ + submodule update: align reporting path for custom command execution
784
+ + submodule status: correct path handling in recursive submodules
785
+ + submodule update --init: correct path handling in recursive submodules
786
+ + submodule foreach: correct path display in recursive submodules
787
+ (this branch is used by sb/submodule-init.)
788
789
"git submodule" reports the paths of submodules the command
790
recurses into, but this was incorrect when the command was not run
791
from the root level of the superproject.
792
663
- Any further comments? Otherwise will merge to 'next'.
664
-
665
-
666
-* sg/diff-multiple-identical-renames (2016-03-30) 1 commit
667
- (merged to 'next' on 2016-04-06 at ac19e48)
668
- + diffcore: fix iteration order of identical files during rename detection
669
-
670
- "git diff -M" used to work better when two originally identical
671
- files A and B got renamed to X/A and X/B by pairing A to X/A and B
672
- to X/B, but this was broken in the 2.0 timeframe.
673
-
674
- Will merge to 'master'.
675
-
676
-
677
-* sk/send-pack-all-fix (2016-03-31) 1 commit
678
- (merged to 'next' on 2016-04-06 at 31e1e1b)
679
- + git-send-pack: fix --all option when used with directory
680
-
681
- "git send-pack --all <there>" was broken when its command line
682
- option parsing was written in the 2.6 timeframe.
683
-
793
Will merge to 'master'.
794
795
@@ -698,100 +807,6 @@ of the repositories listed at
807
($gmane/291198)
808
809
701
-* ak/use-hashmap-iter-first-in-submodule-config (2016-03-23) 1 commit
702
- (merged to 'next' on 2016-04-06 at 2aab890)
703
- + submodule-config: use hashmap_iter_first()
704
-
705
- Minor code cleanup.
706
-
707
- Will merge to 'master'.
708
-
709
-
710
-* ky/branch-d-worktree (2016-03-29) 1 commit
711
- (merged to 'next' on 2016-04-06 at 00f9bff)
712
- + branch -d: refuse deleting a branch which is currently checked out
713
-
714
- When "git worktree" feature is in use, "git branch -d" allowed
715
- deletion of a branch that is checked out in another worktree
716
-
717
- Will merge to 'master'.
718
-
719
-
720
-* ky/branch-m-worktree (2016-04-08) 3 commits
721
- (merged to 'next' on 2016-04-08 at b673b5e)
722
- + set_worktree_head_symref(): fix error message
723
- (merged to 'next' on 2016-04-06 at e7b285c)
724
- + branch -m: update all per-worktree HEADs
725
- + refs: add a new function set_worktree_head_symref
726
-
727
- When "git worktree" feature is in use, "git branch -m" renamed a
728
- branch that is checked out in another worktree without adjusting
729
- the HEAD symbolic ref for the worktree.
730
-
731
- Will merge to 'master'.
732
-
733
-
734
-* rt/completion-help (2016-03-24) 2 commits
735
- (merged to 'next' on 2016-04-06 at 8c3ee08)
736
- + completion: add 'revisions' and 'everyday' to 'git help'
737
- + completion: add option '--guides' to 'git help'
738
-
739
- Shell completion (in contrib/) updates.
740
-
741
- Will merge to 'master'.
742
-
743
-
744
-* rt/rebase-i-shorten-stop-report (2016-03-28) 1 commit
745
- (merged to 'next' on 2016-04-06 at 7a766b7)
746
- + rebase-i: print an abbreviated hash when stop for editing
747
-
748
- The commit object name reported when "rebase -i" stops has been
749
- shortened.
750
-
751
- Will merge to 'master'.
752
-
753
-
754
-* jk/check-repository-format (2016-03-11) 10 commits
755
- (merged to 'next' on 2016-04-06 at a0dada0)
756
- + verify_repository_format: mark messages for translation
757
- + setup: drop repository_format_version global
758
- + setup: unify repository version callbacks
759
- + init: use setup.c's repo version verification
760
- + setup: refactor repo format reading and verification
761
- + config: drop git_config_early
762
- + check_repository_format_gently: stop using git_config_early
763
- + lazily load core.sharedrepository
764
- + wrap shared_repository global in get/set accessors
765
- + setup: document check_repository_format()
766
- (this branch is used by dt/pre-refs-backend.)
767
-
768
- The repository set-up sequence has been streamlined (the biggest
769
- change is that there is no longer git_config_early()), so that we
770
- do not attempt to look into refs/* when we know we do not have a
771
- Git repository.
772
-
773
- Will merge to 'master'.
774
-
775
-
776
-* mj/pull-rebase-autostash (2016-04-04) 9 commits
777
- (merged to 'next' on 2016-04-06 at b4e4f31)
778
- + t5520: test --[no-]autostash with pull.rebase=true
779
- + t5520: reduce commom lines of code
780
- + t5520: factor out common "failing autostash" code
781
- + t5520: factor out common "successful autostash" code
782
- + t5520: use better test to check stderr output
783
- + t5520: ensure consistent test conditions
784
- + t5520: use consistent capitalization in test titles
785
- + pull --rebase: add --[no-]autostash flag
786
- + git-pull.c: introduce git_pull_config()
787
-
788
- "git pull --rebase" learned "--[no-]autostash" option, so that
789
- the rebase.autostash configuration variable set to true can be
790
- overridden from the command line.
791
-
792
- Will merge to 'master'.
793
-
794
-
810
* pb/commit-verbose-config (2016-04-12) 6 commits
811
. commit: add a commit.verbose config variable
812
. t7507-commit-verbose: improve test coverage by testing number of diffs
@@ -803,40 +818,10 @@ of the repositories listed at
818
"git commit" learned to pay attention to "commit.verbose"
819
configuration variable and act as if "--verbose" option was
820
given from the command line.
806
-
807
- Old one reverted, will wait for reroll.
808
- ($gmane/291303).
809
-
810
-
811
-* lt/pretty-expand-tabs (2016-04-04) 4 commits
812
- (merged to 'next' on 2016-04-06 at 186ac2a)
813
- + pretty: test --expand-tabs
814
- + pretty: allow tweaking tabwidth in --expand-tabs
815
- + pretty: enable --expand-tabs by default for selected pretty formats
816
- + pretty: expand tabs in indented logs to make things line up properly
817
-
818
- When "git log" shows the log message indented by 4-spaces, the
819
- remainder of a line after a HT does not align in the way the author
820
- originally intended. The command now expands tabs by default in
821
- such a case, and allows the users to override it with a new option,
822
- '--no-expand-tabs'.
823
-
824
- Will merge to 'master'.
825
-
826
-
827
-* cc/apply (2016-04-01) 4 commits
828
- (merged to 'next' on 2016-04-06 at 2e23c44)
829
- + builtin/apply: free patch when parse_chunk() fails
830
- + builtin/apply: handle parse_binary() failure
831
- + apply: remove unused call to free() in gitdiff_{old,new}name()
832
- + builtin/apply: get rid of useless 'name' variable
833
-
834
- Minor code clean-up.
835
-
836
- Will merge to 'master'.
821
+
822
823
839
-* dt/index-helper (2016-04-07) 16 commits
824
+* dt/index-helper (2016-04-14) 16 commits
825
- read-cache: config for waiting for index-helper
826
- index-helper: optionally automatically run
827
- index-helper: autorun mode
@@ -845,7 +830,7 @@ of the repositories listed at
830
- unpack-trees: preserve index extensions
831
- update-index: enable/disable watchman support
832
- index-helper: use watchman to avoid refreshing index with lstat()
848
- - add watchman support to reduce index refresh cost
833
+ - index-helper: add watchman support to reduce index refresh cost
834
- read-cache: add watchman 'WAMA' extension
835
- index-helper: add --detach
836
- daemonize(): set a flag before exiting the main process
@@ -878,23 +863,24 @@ of the repositories listed at
863
864
865
* jc/rerere-multi (2016-04-06) 11 commits
881
- - rerere: adjust 'forget' to multi-variant world order
882
- - rerere: split code to call ll_merge() further
883
- - rerere: move code related to "forget" together
884
- - rerere: gc and clear
885
- - rerere: do use multiple variants
886
- - t4200: rerere a merge with two identical conflicts
887
- - rerere: allow multiple variants to exist
888
- - rerere: delay the recording of preimage
889
- - rerere: handle leftover rr-cache/$ID directory and postimage files
890
- - rerere: scan $GIT_DIR/rr-cache/$ID when instantiating a rerere_id
891
- - rerere: split conflict ID further
866
+ (merged to 'next' on 2016-04-13 at 3db2753)
867
+ + rerere: adjust 'forget' to multi-variant world order
868
+ + rerere: split code to call ll_merge() further
869
+ + rerere: move code related to "forget" together
870
+ + rerere: gc and clear
871
+ + rerere: do use multiple variants
872
+ + t4200: rerere a merge with two identical conflicts
873
+ + rerere: allow multiple variants to exist
874
+ + rerere: delay the recording of preimage
875
+ + rerere: handle leftover rr-cache/$ID directory and postimage files
876
+ + rerere: scan $GIT_DIR/rr-cache/$ID when instantiating a rerere_id
877
+ + rerere: split conflict ID further
878
879
"git rerere" can encounter two or more files with the same conflict
880
signature that have to be resolved in different ways, but there was
881
no way to record these separate resolutions.
882
897
- Will merge to 'next'.
883
+ Will merge to 'master'.
884
885
886
* jc/merge-drop-old-syntax (2015-04-29) 1 commit