What's cooking (2016/09 #02)

Junio C Hamano committed Sep 8, 2016 at 15:20 UTC fa7f92e128af833b904d4345ed97e1abe36e2610
1 file changed +269 -22
whats-cooking.txt
+269 -22
@@ -1,10 +1,10 @@
1 To: git@vger.kernel.org
2 Bcc: lwn@lwn.net
3 -Subject: What's cooking in git.git (Sep 2016, #01; Fri, 2)
3 +Subject: What's cooking in git.git (Sep 2016, #02; Thu, 8)
4 X-master-at: 6ebdac1bab966b720d776aa43ca188fe378b1f4b
5 X-next-at: f2ff48496e8dc270b21b0290a2a3e4ef98e352ed
6
7 -What's cooking in git.git (Sep 2016, #01; Fri, 2)
7 +What's cooking in git.git (Sep 2016, #02; Thu, 8)
8 --------------------------------------------------
9
10 Here are the topics that have been cooking. Commits prefixed with
@@ -12,12 +12,11 @@ Here are the topics that have been cooking. Commits prefixed with
12 '+' are in 'next'. The ones marked with '.' do not appear in any of
13 the integration branches, but I am still holding onto them.
14
15 -Git 2.10 has been tagged. In a few days, many topics that have been
16 -cooking in 'next' would be merged to 'master'; as usual, I plan to
17 -keep the tip of 'next' closed for a week or so and encourage people
18 -to pay more attention to possible regressions than to show their
19 -shiny new toys, and then open the new cycle by rebuilding 'next'
20 -after that.
15 +There are a few more topics in flight that may be ready to be picked
16 +up but I haven't, and other topics in flight that may not be quite
17 +ready. I'll start merging topics that have been cooking in 'next'
18 +to 'master', rewind and rebuild 'next', and merge those that have
19 +been waiting in 'pu' to 'next' before picking them up.
20
21 You can find the changes described here in the integration branches
22 of the repositories listed at
@@ -27,21 +26,247 @@ of the repositories listed at
26 --------------------------------------------------
27 [New Topics]
28
30 -* jc/submodule-anchor-git-dir (2016-09-01) 1 commit
31 - - submodule: avoid auto-discovery in prepare_submodule_repo_env()
29 +* bc/object-id (2016-09-07) 20 commits
30 + - builtin/reset: convert to use struct object_id
31 + - builtin/commit-tree: convert to struct object_id
32 + - builtin/am: convert to struct object_id
33 + - refs: add an update_ref_oid function.
34 + - sha1_name: convert get_sha1_mb to struct object_id
35 + - builtin/update-index: convert file to struct object_id
36 + - notes: convert init_notes to use struct object_id
37 + - builtin/rm: convert to use struct object_id
38 + - builtin/blame: convert file to use struct object_id
39 + - Convert read_mmblob to take struct object_id.
40 + - notes-merge: convert struct notes_merge_pair to struct object_id
41 + - builtin/checkout: convert some static functions to struct object_id
42 + - streaming: make stream_blob_to_fd take struct object_id
43 + - builtin: convert textconv_object to use struct object_id
44 + - builtin/cat-file: convert some static functions to struct object_id
45 + - builtin/cat-file: convert struct expand_data to use struct object_id
46 + - builtin/log: convert some static functions to use struct object_id
47 + - builtin/blame: convert struct origin to use struct object_id
48 + - builtin/apply: convert static functions to struct object_id
49 + - cache: convert struct cache_entry to use struct object_id
50 +
51 + The "unsigned char sha1[20]" to "struct object_id" conversion
52 + continues. Notable changes in this round includes that ce->sha1,
53 + i.e. the object name recorded in the cache_entry, turns into an
54 + object_id.
55 +
56 + It had merge conflicts with a few topics in flight (Christian's
57 + "apply.c split", Dscho's "cat-file --filters" and Jeff Hostetler's
58 + "status --porcelain-v2"). Extra sets of eyes double-checking for
59 + mismerges are highly appreciated.
60 +
61 +
62 +* ep/use-git-trace-curl-in-tests (2016-09-07) 4 commits
63 + - t5551-http-fetch-smart.sh: use the GIT_TRACE_CURL environment var
64 + - t5550-http-fetch-dumb.sh: use the GIT_TRACE_CURL environment var
65 + - test-lib.sh: preserve GIT_TRACE_CURL from the environment
66 + - t5541-http-push-smart.sh: use the GIT_TRACE_CURL environment var
67 +
68 + Update a few tests that used to use GIT_CURL_VERBOSE to use the
69 + newer GIT_TRACE_CURL.
70
33 - Having a submodule whose ".git" repository is somehow corrupt
34 - caused a few commands that recurse into submodules loop forever.
71 + Will merge to 'next'.
72 +
73 +
74 +* jk/pack-tag-of-tag (2016-09-07) 5 commits
75 + - pack-objects: walk tag chains for --include-tag
76 + - t5305: simplify packname handling
77 + - t5305: use "git -C"
78 + - t5305: drop "dry-run" of unpack-objects
79 + - t5305: move cleanup into test block
80 +
81 + "git pack-objects --include-tag" was taught that when we know that
82 + we are sending an object C, we want a tag B that directly points at
83 + C but also a tag A that points at the tag B. We used to miss the
84 + intermediate tag B in some cases.
85 +
86 +
87 +* js/t6026-clean-up (2016-09-07) 1 commit
88 + - t6026-merge-attr: clean up background process at end of test case
89 +
90 + A test spawned a short-lived background process, which sometimes
91 + prevented the test directory from getting removed at the end of the
92 + script on some platforms.
93
94 Will merge to 'next'.
95
96
39 -* jc/forbid-symbolic-ref-d-HEAD (2016-09-02) 1 commit
40 - - symbolic-ref -d: do not allow removal of HEAD
97 +* js/t9903-chaining (2016-09-07) 1 commit
98 + - t9903: fix broken && chain
99 +
100 + Will merge to 'next'.
101 +
102 +
103 +* jt/accept-capability-advertisement-when-fetching-from-void (2016-09-07) 2 commits
104 + - connect: advertized capability is not a ref
105 + - tests: move test_lazy_prereq JGIT to test-lib.sh
106 +
107 + JGit can show a fake ref "capabilities^{}" to "git fetch" when it
108 + does not advertise any refs, but "git fetch" was not prepared to
109 + see such an advertisement.
110 +
111 + Waiting for a reroll.
112 + Rewording the log, and avoiding making it overly loose are needed.
113 + cf. <20160908013431.GC25016@google.com>
114 +
115 +
116 +* rs/compat-strdup (2016-09-07) 1 commit
117 + - compat: move strdup(3) replacement to its own file
118 +
119 + Will merge to 'next'.
120 +
121 +
122 +* rs/hex2chr (2016-09-07) 1 commit
123 + - introduce hex2chr() for converting two hexadecimal digits to a character
124 +
125 + Will merge to 'next'.
126 +
127 +
128 +* rt/rebase-i-broken-insn-advise (2016-09-07) 1 commit
129 + - rebase -i: improve advice on bad instruction lines
130 +
131 + When "git rebase -i" is given a broken instruction, it told the
132 + user to fix it with "--edit-todo", but didn't say what the step
133 + after that was (i.e. "--continue").
134 +
135 + Will hold.
136 + Dscho's "rebase -i" hopefully will become available in 'pu', by
137 + which time an equivalent of this fix would be ported to C. This is
138 + queued merely as a reminder.
139 +
140 +
141 +* sb/xdiff-remove-unused-static-decl (2016-09-07) 1 commit
142 + - xdiff: remove unneeded declarations
143 +
144 + Code cleanup.
145 +
146 + Will merge to 'next'.
147 +
148 +
149 +* sy/git-gui-i18n-ja (2016-09-07) 7 commits
150 + - Merge branch 'sy/i18n' of git-gui
151 + - git-gui: update Japanese information
152 + - git-gui: update Japanese translation
153 + - git-gui: add Japanese language code
154 + - git-gui: apply po template to Japanese translation
155 + - git-gui: consistently use the same word for "blame" in Japanese
156 + - git-gui: consistently use the same word for "remote" in Japanese
157 +
158 + Update Japanese translation for "git-gui".
159 +
160 +
161 +* ah/misc-message-fixes (2016-09-08) 5 commits
162 + - unpack-trees: do not capitalize "working"
163 + - git-merge-octopus: do not capitalize "octopus"
164 + - git-rebase--interactive: fix English grammar
165 + - cat-file: put spaces around pipes in usage string
166 + - am: put spaces around pipe in usage string
167 +
168 + Message cleanup.
169 +
170 + Will merge to 'next'.
171 +
172 +
173 +* jk/fix-remote-curl-url-wo-proto (2016-09-08) 1 commit
174 + - remote-curl: handle URLs without protocol
175 +
176 + "git fetch http::/site/path" did not die correctly and segfaulted
177 + instead.
178 +
179 + Will merge to 'next'.
180
42 - "git symbolic-ref -d HEAD" happily removes the symbolic ref, but
43 - the resulting repository becomes an invalid one. Teach the command
44 - to forbid removal of HEAD.
181 +
182 +* jn/fix-connect-unexpected-hangup-diag (2016-09-08) 1 commit
183 + - connect: tighten check for unexpected early hang up
184 +
185 + When the other side disconnects without giving any ref
186 + advertisement, we used to say "there may not be a repository at
187 + that URL", but we may have seen other advertisement like "shallow"
188 + and ".have" in which case we definitely know that a repository is
189 + there. The code to detect this case has been updated.
190 +
191 + Waiting for a reroll with test.
192 + cf. <20160908015040.GF25016@google.com>
193 +
194 +
195 +* jt/format-patch-base-info-above-sig (2016-09-08) 1 commit
196 + - format-patch: show base info before email signature
197 +
198 + "git format-patch --base=..." feature that was recently added
199 + showed the base commit information after "-- " e-mail signature
200 + line, which turned out to be inconvenient. The base information
201 + has been moved above the signature line.
202 +
203 + Needs rephrasing of the log message to describe an accidental bugfix.
204 + cf. <xmqqd1kef5k5.fsf@gitster.mtv.corp.google.com>
205 +
206 +
207 +* nd/checkout-disambiguation (2016-09-08) 3 commits
208 + - checkout: fix ambiguity check in subdir
209 + - checkout.txt: document a common case that ignores ambiguation rules
210 + - checkout: add some spaces between code and comment
211 +
212 + "git checkout <word>" does not follow the usual disambiguation
213 + rules when the <word> can be both a rev and a path, to allow
214 + checking out a branch 'foo' in a project that happens to have a
215 + file 'foo' in the working tree without having to disambiguate.
216 + This was poorly documented and the check was incorrect when the
217 + command was run from a subdirectory.
218 +
219 + Waiting for a reroll.
220 + cf. <xmqq7famgnwo.fsf@gitster.mtv.corp.google.com>
221 +
222 +
223 +* sb/diff-cleanup (2016-09-08) 3 commits
224 + - diff: remove dead code
225 + - diff: omit found pointer from emit_callback
226 + - diff.c: use diff_options directly
227 +
228 + Code cleanup.
229 +
230 + Will merge to 'next'.
231 +
232 +
233 +* sb/transport-report-missing-submodule-on-stderr (2016-09-08) 1 commit
234 + - transport: report missing submodule pushes consistently on stderr
235 +
236 + Message cleanup.
237 +
238 + Will merge to 'next'.
239 +
240 +
241 +* sg/fix-versioncmp-with-common-suffix (2016-09-08) 5 commits
242 + - versioncmp: cope with common leading parts in versionsort.prereleaseSuffix
243 + - versioncmp: pass full tagnames to swap_prereleases()
244 + - t7004-tag: add version sort tests to show prerelease reordering issues
245 + - t7004-tag: use test_config helper
246 + - t7004-tag: delete unnecessary tags with test_when_finished
247 +
248 + The prereleaseSuffix feature of version comparison that is used in
249 + "git tag -l" did not correctly when two or more prereleases for the
250 + same release were present (e.g. when 2.0, 2.0-beta1, and 2.0-beta2
251 + are there and the code needs to compare 2.0-beta1 and 2.0-beta2).
252 +
253 + Waiting for a reroll.
254 + cf. <20160908223727.Horde.jVOOJ278ssZ3qkyjkmyqZD-@webmail.informatik.kit.edu>
255 +
256 +
257 +* va/i18n (2016-09-08) 10 commits
258 + - i18n: update-index: mark warning for translation
259 + - i18n: show-branch: mark error messages for translation
260 + - i18n: receive-pack: mark messages for translation
261 + - notes: lowercase first word of error messages
262 + - i18n: notes: mark error messages for translation
263 + - i18n: merge-recursive: mark verbose message for translation
264 + - i18n: merge-recursive: mark error messages for translation
265 + - i18n: config: mark error message for translation
266 + - i18n: branch: mark option description for translation
267 + - i18n: blame: mark error messages for translation
268 +
269 + More i18n.
270
271 Will merge to 'next'.
272
@@ -298,6 +523,25 @@ of the repositories listed at
523 --------------------------------------------------
524 [Cooking]
525
526 +* jc/submodule-anchor-git-dir (2016-09-01) 1 commit
527 + - submodule: avoid auto-discovery in prepare_submodule_repo_env()
528 +
529 + Having a submodule whose ".git" repository is somehow corrupt
530 + caused a few commands that recurse into submodules loop forever.
531 +
532 + Will merge to 'next'.
533 +
534 +
535 +* jc/forbid-symbolic-ref-d-HEAD (2016-09-02) 1 commit
536 + - symbolic-ref -d: do not allow removal of HEAD
537 +
538 + "git symbolic-ref -d HEAD" happily removes the symbolic ref, but
539 + the resulting repository becomes an invalid one. Teach the command
540 + to forbid removal of HEAD.
541 +
542 + Will merge to 'next'.
543 +
544 +
545 * bh/diff-highlight-graph (2016-08-31) 6 commits
546 (merged to 'next' on 2016-08-31 at 523a15f)
547 + diff-highlight: avoid highlighting combined diffs
@@ -579,7 +823,10 @@ of the repositories listed at
823 Will merge to 'master'.
824
825
582 -* mh/diff-indent-heuristic (2016-08-23) 6 commits
826 +* mh/diff-indent-heuristic (2016-09-07) 9 commits
827 + - SQAUSH???
828 + - blame: honor the diff heuristic options and config
829 + - parse-options: add parse_opt_unknown_cb()
830 - diff: improve positioning of add/delete blocks in diffs
831 - xdl_change_compact(): introduce the concept of a change group
832 - recs_match(): take two xrecord_t pointers as arguments
@@ -593,11 +840,11 @@ of the repositories listed at
840 are the same. A command line option is added to help with the
841 experiment to find a good heuristics.
842
596 - Waiting for a reroll.
597 - cf. <a27aa17e-f602-fc49-92b3-2198e4772e47@ramsayjones.plus.com>
843 + Rerolled.
844 + Needs adjustment of test numbers. Otherwise looked OK.
845
846
600 -* cc/apply-am (2016-08-29) 41 commits
847 +* cc/apply-am (2016-09-07) 41 commits
848 - builtin/am: use apply API in run_apply()
849 - apply: learn to use a different index file
850 - apply: pass apply state to build_fake_ancestor()
@@ -643,7 +890,7 @@ of the repositories listed at
890 "git am" has been taught to make an internal call to "git apply"'s
891 innards without spawning the latter as a separate process.
892
646 - Waiting for a reroll.
893 + Will merge to 'next'.
894
895
896 * jk/pack-objects-optim-mru (2016-08-11) 4 commits