What's cooking (2019/06 #04)

Junio C Hamano committed Jun 14, 2019 at 13:46 UTC 6a57eddf289230fe867ee0d569d9de9fffb0c817
1 file changed +313 -305
whats-cooking.txt
+313 -305
@@ -1,10 +1,10 @@
1 To: git@vger.kernel.org
2 Bcc: lwn@lwn.net
3 -Subject: What's cooking in git.git (Jun 2019, #03; Wed, 12)
4 -X-master-at: b697d92f56511e804b8ba20ccbe7bdc85dc66810
5 -X-next-at: 95628af9bbb1414c85a80e992c77d3eb4efd24a4
3 +Subject: What's cooking in git.git (Jun 2019, #04; Fri, 14)
4 +X-master-at: 0aae918dd929862d3ce0ea2960897787bb269a3b
5 +X-next-at: d03e0c960ffbffc6714d0e076319c44163f03ac5
6
7 -What's cooking in git.git (Jun 2019, #03; Wed, 12)
7 +What's cooking in git.git (Jun 2019, #04; Fri, 14)
8 --------------------------------------------------
9
10 Here are the topics that have been cooking. Commits prefixed with
@@ -12,7 +12,9 @@ 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 2.22 final has been tagged.
15 +The first batch of topics post 2.22 are now in 'master'. Let's thin
16 +down what are left in 'next' before taking anything new, and try to
17 +keep this cycle short.
18
19 You can find the changes described here in the integration branches
20 of the repositories listed at
@@ -22,95 +24,222 @@ of the repositories listed at
24 --------------------------------------------------
25 [Graduated to "master"]
26
25 -* en/merge-directory-renames-fix (2019-06-05) 1 commit
26 - (merged to 'next' on 2019-06-06 at fd59bad9d7)
27 - + merge-recursive: restore accidentally dropped setting of path
27 +* ab/deprecate-R-for-dynpath (2019-05-19) 1 commit
28 + (merged to 'next' on 2019-05-19 at 944976e981)
29 + + Makefile: remove the NO_R_TO_GCC_LINKER flag
30
29 - Recent code restructuring of merge-recursive engine introduced a
30 - regression dealing with rename/add conflict.
31 + The way of specifying the path to find dynamic libraries at runtime
32 + has been simplified. The old default to pass -R/path/to/dir has been
33 + replaced with the new default to pass -Wl,-rpath,/path/to/dir,
34 + which is the more recent GCC uses. Those who need to build with an
35 + old GCC can still use "CC_LD_DYNPATH=-R"
36
32 ---------------------------------------------------
33 -[New Topics]
37
35 -* mo/hpux-dynpath (2019-06-07) 1 commit
36 - - configure: Detect linking style for HP aCC on HP-UX
38 +* ab/fail-prereqs-in-test (2019-05-14) 1 commit
39 + (merged to 'next' on 2019-05-16 at d1be55f485)
40 + + tests: add a special setup where prerequisites fail
41
38 - Auto-detect how to tell HP-UX aCC where to use dynamicallly linked
39 - libraries from at runtime.
42 + Developer support to emulate unsatisfied prerequisites in tests to
43 + ensure that the remainer of the tests still succeeds when tests
44 + with prerequisites are skipped.
45
41 - Will merge to 'next'.
46
47 +* ab/send-email-transferencoding-fix (2019-05-29) 7 commits
48 + (merged to 'next' on 2019-05-29 at c8a99d18c0)
49 + + send-email: fix regression in sendemail.identity parsing
50 + + send-email: document --no-[to|cc|bcc]
51 + + send-email: fix broken transferEncoding tests
52 + + send-email: remove cargo-culted multi-patch pattern in tests
53 + (merged to 'next' on 2019-05-13 at 38c6a1e7e0)
54 + + send-email: do defaults -> config -> getopt in that order
55 + + send-email: rename the @bcclist variable for consistency
56 + + send-email: move the read_config() function above getopts
57
44 -* js/mergetool-optim (2019-06-12) 4 commits
45 - - mergetool: use shell variable magic instead of `awk`
46 - - mergetool: dissect strings with shell variable magic instead of `expr`
47 - - t7610-mergetool: use test_cmp instead of test $(cat file) = $txt
48 - - t7610-mergetool: do not place pipelines headed by `yes` in subshells
58 + Since "git send-email" learned to take 'auto' as the value for the
59 + transfer-encoding, it by mistake stopped honoring the values given
60 + to the configuration variables sendemail.transferencoding and/or
61 + sendemail.<ident>.transferencoding. This has been corrected to
62 + (finally) redoing the order of setting the default, reading the
63 + configuration and command line options.
64
50 - "git mergetool" and its tests now spawn fewer subprocesses.
65
52 - Will merge to 'next'.
66 +* dl/format-patch-notes-config (2019-05-17) 2 commits
67 + (merged to 'next' on 2019-05-19 at d3f6f1872b)
68 + + format-patch: teach format.notes config option
69 + + git-format-patch.txt: document --no-notes option
70
71 + "git format-patch" learns a configuration to set the default for
72 + its --notes=<ref> option.
73
55 -* cc/test-oidmap (2019-06-11) 4 commits
56 - - oidmap: use sha1hash() instead of static hash() function
57 - - SQUASH??? sh style
58 - - t: add t0016-oidmap.sh
59 - - t/helper: add test-oidmap.c
74
75 +* en/fast-export-encoding (2019-05-14) 5 commits
76 + (merged to 'next' on 2019-05-16 at c88bd3edb5)
77 + + fast-export: do automatic reencoding of commit messages only if requested
78 + + fast-export: differentiate between explicitly UTF-8 and implicitly UTF-8
79 + + fast-export: avoid stripping encoding header if we cannot reencode
80 + + fast-import: support 'encoding' commit header
81 + + t9350: fix encoding test to actually test reencoding
82
62 -* ds/midx-expire-repack (2019-06-11) 11 commits
63 - - t5319-multi-pack-index.sh: test batch size zero
64 - - midx: add test that 'expire' respects .keep files
65 - - multi-pack-index: test expire while adding packs
66 - - midx: implement midx_repack()
67 - - multi-pack-index: prepare 'repack' subcommand
68 - - multi-pack-index: implement 'expire' subcommand
69 - - midx: refactor permutation logic and pack sorting
70 - - midx: simplify computation of pack name lengths
71 - - multi-pack-index: prepare for 'expire' subcommand
72 - - Docs: rearrange subcommands for multi-pack-index
73 - - repack: refactor pack deletion for future use
83 + The "git fast-export/import" pair has been taught to handle commits
84 + with log messages in encoding other than UTF-8 better.
85
75 - "git multi-pack-index" learned expire and repack subcommands.
86
77 - Will merge to 'next'.
87 +* ew/update-server-info (2019-05-15) 1 commit
88 + (merged to 'next' on 2019-05-19 at bf4f2871ab)
89 + + update-server-info: avoid needless overwrites
90
91 + "git update-server-info" learned not to rewrite the file with the
92 + same contents.
93
80 -* md/sort-detached-head-first (2019-06-11) 1 commit
81 - - ref-filter: sort detached HEAD lines firstly
94
83 - "git branch --list" learned to always output the detached HEAD as
84 - the first item (when the HEAD is detached, of course), instead of
85 - relying on the textual order of sorting
95 +* jk/help-unknown-ref-fix (2019-05-15) 2 commits
96 + (merged to 'next' on 2019-05-19 at e3e01160f7)
97 + + help_unknown_ref(): check for refname ambiguity
98 + + help_unknown_ref(): duplicate collected refnames
99
100 + Improve the code to show args with potential typo that cannot be
101 + interpreted as a commit-ish.
102
88 -* nd/completion-no-cache-failure (2019-06-12) 1 commit
89 - - completion: do not cache if --git-completion-helper fails
103
91 - An incorrect list of options was cached after command line
92 - completion failed (e.g. trying to complete a command that requires
93 - a repository outside one), which has been corrected.
104 +* jk/unused-params-final-batch (2019-05-13) 14 commits
105 + (merged to 'next' on 2019-05-15 at ef7435264c)
106 + + verify-commit: simplify parameters to run_gpg_verify()
107 + + show-branch: drop unused parameter from show_independent()
108 + + rev-list: drop unused void pointer from finish_commit()
109 + + remove_all_fetch_refspecs(): drop unused "remote" parameter
110 + + receive-pack: drop unused "commands" from prepare_shallow_update()
111 + + pack-objects: drop unused rev_info parameters
112 + + name-rev: drop unused parameters from is_better_name()
113 + + mktree: drop unused length parameter
114 + + wt-status: drop unused status parameter
115 + + read-cache: drop unused parameter from threaded load
116 + + clone: drop dest parameter from copy_alternates()
117 + + submodule: drop unused prefix parameter from some functions
118 + + builtin: consistently pass cmd_* prefix to parse_options
119 + + cmd_{read,write}_tree: rename "unused" variable that is used
120
95 - Will merge to 'next'.
121 + Remove many unused parameters throughout the codebase, with the
122 + ultimate aim to allow us compile with -Wunused-parameter cleanly.
123
124
98 -* sg/rebase-progress (2019-06-11) 4 commits
99 - - progress: use term_clear_line()
100 - - rebase: fix garbled progress display with '-x'
101 - - pager: add a helper function to clear the last line in the terminal
102 - - t3404-rebase-interactive: use the 'q_to_cr' helper
125 +* js/rebase-cleanup (2019-05-15) 5 commits
126 + (merged to 'next' on 2019-05-16 at ccfed8f263)
127 + + rebase: fold git-rebase--common into the -p backend
128 + + sequencer: the `am` and `rebase--interactive` scripts are gone
129 + + .gitignore: there is no longer a built-in `git-rebase--interactive`
130 + + t3400: stop referring to the scripted rebase
131 + + Drop unused git-rebase--am.sh
132
104 - Use "Erase in Line" CSI sequence that is already used in the editor
105 - support to clear cruft in the progress output.
133 + Update supporting parts of "git rebase" to remove code that should
134 + no longer be used.
135
136
108 -* sg/trace2-rename (2019-06-12) 1 commit
109 - - Revert "test-lib: whitelist GIT_TR2_* in the environment"
137 +* mh/import-transport-fd-fix (2019-05-16) 2 commits
138 + (merged to 'next' on 2019-05-19 at 5e86f92f7a)
139 + + Use xmmap_gently instead of xmmap in use_pack
140 + + dup() the input fd for fast-import used for remote helpers
141
111 - Dev support update to help tracing out tests.
142 + The ownership rule for the file descriptor to fast-import remote
143 + backend was mixed up, leading to unrelated file descriptor getting
144 + closed, which has been fixed.
145
113 - Will merge to 'next'.
146 +
147 +* nd/corrupt-worktrees (2019-05-15) 1 commit
148 + (merged to 'next' on 2019-05-16 at d92c25f800)
149 + + worktree add: be tolerant of corrupt worktrees
150 +
151 + "git worktree add" used to fail when another worktree connected to
152 + the same repository was corrupt, which has been corrected.
153 +
154 +
155 +* nd/init-relative-template-fix (2019-05-13) 1 commit
156 + (merged to 'next' on 2019-05-15 at 4d5b17f712)
157 + + init: make --template path relative to $CWD
158 +
159 + A relative pathname given to "git init --template=<path> <repo>"
160 + ought to be relative to the directory "git init" gets invoked in,
161 + but it instead was made relative to the repository, which has been
162 + corrected.
163 +
164 +
165 +* nd/merge-quit (2019-05-19) 2 commits
166 + (merged to 'next' on 2019-05-19 at 9880e7ee4e)
167 + + merge: add --quit
168 + + merge: remove drop_save() in favor of remove_merge_branch_state()
169 +
170 + "git merge" learned "--quit" option that cleans up the in-progress
171 + merge while leaving the working tree and the index still in a mess.
172 +
173 +
174 +* nd/worktree-name-sanitization (2019-05-15) 1 commit
175 + (merged to 'next' on 2019-05-16 at 9a2dd33122)
176 + + worktree add: sanitize worktree names
177 +
178 + In recent versions of Git, per-worktree refs are exposed in
179 + refs/worktrees/<wtname>/ hierarchy, which means that worktree names
180 + must be a valid refname component. The code now sanitizes the names
181 + given to worktrees, to make sure these refs are well-formed.
182 +
183 +
184 +* pw/rebase-edit-message-for-replayed-merge (2019-05-19) 1 commit
185 + (merged to 'next' on 2019-05-19 at dc3e30641c)
186 + + rebase -r: always reword merge -c
187 +
188 + A "merge -c" instruction during "git rebase --rebase-merges" should
189 + give the user a chance to edit the log message, even when there is
190 + otherwise no need to create a new merge and replace the existing
191 + one (i.e. fast-forward instead), but did not. Which has been
192 + corrected.
193 +
194 +
195 +* sb/format-patch-base-patch-id-fix (2019-05-08) 2 commits
196 + (merged to 'next' on 2019-05-15 at 1ab7d2b71c)
197 + + format-patch: make --base patch-id output stable
198 + + format-patch: inform user that patch-id generation is unstable
199 +
200 + The "--base" option of "format-patch" computed the patch-ids for
201 + prerequisite patches in an unstable way, which has been updated to
202 + compute in a way that is compatible with "git patch-id --stable".
203 +
204 +--------------------------------------------------
205 +[New Topics]
206 +
207 +* js/gcc-8-and-9 (2019-06-13) 4 commits
208 + - config: avoid calling `labs()` on too-large data type
209 + - winansi: simplify loading the GetCurrentConsoleFontEx() function
210 + - kwset: allow building with GCC 8
211 + - poll (mingw): allow compiling with GCC 8 and DEVELOPER=1
212 +
213 + Code clean-up for new compilers.
214 +
215 + The 'kwset' one may want to be discussed a bit longer. Perhaps
216 + merge the other three earlier to 'next' and then to 'master'
217 + separately?
218 +
219 +
220 +* pw/add-p-recount (2019-06-13) 1 commit
221 + - add -p: fix checkout -p with pathological context
222 +
223 + "git checkout -p" needs to selectively apply a patch in reverse,
224 + which did not work well.
225 +
226 + Will merge to 'next' and then to 'master'.
227 +
228 +
229 +* rs/avoid-overflow-in-midpoint-computation (2019-06-13) 1 commit
230 + - cleanup: fix possible overflow errors in binary search, part 2
231 +
232 + Code clean-up to avoid signed integer overlaps during binary search.
233 +
234 + Will merge to 'next' and then to 'master'.
235 +
236 +
237 +* js/t3404-typofix (2019-06-14) 1 commit
238 + - t3404: fix a typo
239 +
240 + Typofix.
241 +
242 + Will merge to 'next' and then to 'master'.
243
244 --------------------------------------------------
245 [Stalled]
@@ -151,41 +280,128 @@ of the repositories listed at
280 WIP for allowing a response to "git fetch" to instruct the bulk of
281 the pack contents to be instead taken from elsewhere (aka CDN).
282
154 - Stalled
283 + Stalled
284 +
285 +
286 +* js/protocol-advertise-multi (2018-12-28) 1 commit
287 + - protocol: advertise multiple supported versions
288 +
289 + The transport layer has been updated so that the protocol version
290 + used can be negotiated between the parties, by the initiator
291 + listing the protocol versions it is willing to talk, and the other
292 + side choosing from one of them.
293 +
294 + Expecting a reroll.
295 + cf. <CANq=j3u-zdb_FvNJGPCmygNMScseav63GhVvBX3NcVS4f7TejA@mail.gmail.com>
296 +
297 +
298 +* mk/use-size-t-in-zlib (2018-10-15) 1 commit
299 + - zlib.c: use size_t for size
300 +
301 + The wrapper to call into zlib followed our long tradition to use
302 + "unsigned long" for sizes of regions in memory, which have been
303 + updated to use "size_t".
304 +
305 +
306 +* dl/remote-save-to-push (2018-12-11) 1 commit
307 + - remote: add --save-to-push option to git remote set-url
308 +
309 + "git remote set-url" learned a new option that moves existing value
310 + of the URL field to pushURL field of the remote before replacing
311 + the URL field with a new value.
312 +
313 + Anybody who wants to champion this topic?
314 + I am personally not yet quite convinced if this is worth pursuing.
315 +
316 +--------------------------------------------------
317 +[Cooking]
318 +
319 +* mo/hpux-dynpath (2019-06-07) 1 commit
320 + - configure: Detect linking style for HP aCC on HP-UX
321 +
322 + Auto-detect how to tell HP-UX aCC where to use dynamicallly linked
323 + libraries from at runtime.
324 +
325 + Will merge to 'next' and then to 'master'.
326 +
327 +
328 +* js/mergetool-optim (2019-06-12) 4 commits
329 + - mergetool: use shell variable magic instead of `awk`
330 + - mergetool: dissect strings with shell variable magic instead of `expr`
331 + - t7610-mergetool: use test_cmp instead of test $(cat file) = $txt
332 + - t7610-mergetool: do not place pipelines headed by `yes` in subshells
333 +
334 + "git mergetool" and its tests now spawn fewer subprocesses.
335 +
336 + Will merge to 'next' and then to 'master'.
337 +
338 +
339 +* cc/test-oidmap (2019-06-13) 3 commits
340 + - oidmap: use sha1hash() instead of static hash() function
341 + - t: add t0016-oidmap.sh
342 + - t/helper: add test-oidmap.c
343 +
344 + Extend the test coverage a bit.
345 +
346 +
347 +* ds/midx-expire-repack (2019-06-11) 11 commits
348 + - t5319-multi-pack-index.sh: test batch size zero
349 + - midx: add test that 'expire' respects .keep files
350 + - multi-pack-index: test expire while adding packs
351 + - midx: implement midx_repack()
352 + - multi-pack-index: prepare 'repack' subcommand
353 + - multi-pack-index: implement 'expire' subcommand
354 + - midx: refactor permutation logic and pack sorting
355 + - midx: simplify computation of pack name lengths
356 + - multi-pack-index: prepare for 'expire' subcommand
357 + - Docs: rearrange subcommands for multi-pack-index
358 + - repack: refactor pack deletion for future use
359 +
360 + "git multi-pack-index" learned expire and repack subcommands.
361 +
362 + Will merge to 'next'.
363 +
364 +
365 +* md/sort-detached-head-first (2019-06-11) 1 commit
366 + - ref-filter: sort detached HEAD lines firstly
367 +
368 + "git branch --list" learned to always output the detached HEAD as
369 + the first item (when the HEAD is detached, of course), instead of
370 + relying on the textual order of sorting
371 +
372 + At least the test needs to be redone.
373
374
157 -* js/protocol-advertise-multi (2018-12-28) 1 commit
158 - - protocol: advertise multiple supported versions
375 +* nd/completion-no-cache-failure (2019-06-12) 1 commit
376 + - completion: do not cache if --git-completion-helper fails
377
160 - The transport layer has been updated so that the protocol version
161 - used can be negotiated between the parties, by the initiator
162 - listing the protocol versions it is willing to talk, and the other
163 - side choosing from one of them.
378 + An incorrect list of options was cached after command line
379 + completion failed (e.g. trying to complete a command that requires
380 + a repository outside one), which has been corrected.
381
165 - Expecting a reroll.
166 - cf. <CANq=j3u-zdb_FvNJGPCmygNMScseav63GhVvBX3NcVS4f7TejA@mail.gmail.com>
382 + Will merge to 'next' and then to 'master'.
383
384
169 -* mk/use-size-t-in-zlib (2018-10-15) 1 commit
170 - - zlib.c: use size_t for size
385 +* sg/rebase-progress (2019-06-11) 4 commits
386 + - progress: use term_clear_line()
387 + - rebase: fix garbled progress display with '-x'
388 + - pager: add a helper function to clear the last line in the terminal
389 + - t3404-rebase-interactive: use the 'q_to_cr' helper
390
172 - The wrapper to call into zlib followed our long tradition to use
173 - "unsigned long" for sizes of regions in memory, which have been
174 - updated to use "size_t".
391 + Use "Erase in Line" CSI sequence that is already used in the editor
392 + support to clear cruft in the progress output.
393
394 + Its test may have to be updated to work around glitch in test setup
395 + on Windows.
396
177 -* dl/remote-save-to-push (2018-12-11) 1 commit
178 - - remote: add --save-to-push option to git remote set-url
397
180 - "git remote set-url" learned a new option that moves existing value
181 - of the URL field to pushURL field of the remote before replacing
182 - the URL field with a new value.
398 +* sg/trace2-rename (2019-06-12) 1 commit
399 + - Revert "test-lib: whitelist GIT_TR2_* in the environment"
400
184 - Anybody who wants to champion this topic?
185 - I am personally not yet quite convinced if this is worth pursuing.
401 + Dev support update to help tracing out tests.
402 +
403 + Will merge to 'next' and then to 'master'.
404
187 ---------------------------------------------------
188 -[Cooking]
405
406 * fc/fetch-with-import-fix (2019-06-04) 5 commits
407 - fetch: fix regression with transport helpers
@@ -197,7 +413,7 @@ of the repositories listed at
413 Code restructuring during 2.20 period broke fetching tags via
414 "import" based transports.
415
200 - Will merge to 'next'.
416 + Will merge to 'next' and then to 'master'.
417
418
419 * jl/status-reduce-vertical-blank (2019-06-04) 1 commit
@@ -236,7 +452,7 @@ of the repositories listed at
452
453 Doc update.
454
239 - Will merge to 'next'.
455 + Will merge to 'next' and then to 'master'.
456
457
458 * dl/includeif-onbranch (2019-06-05) 1 commit
@@ -280,7 +496,7 @@ of the repositories listed at
496 A new tag.gpgSign configuration variable turns "git tag -a" into
497 "git tag -s".
498
283 - Will merge to 'next'.
499 + Will merge to 'next' and then to 'master'.
500
501
502 * md/list-objects-filter-memfix (2019-05-31) 1 commit
@@ -318,7 +534,7 @@ of the repositories listed at
534
535 Doc update.
536
321 - Will merge to 'next'.
537 + Will merge to 'next' and then to 'master'.
538
539
540 * bb/unicode-12.1-reiwa (2019-05-29) 1 commit
@@ -510,42 +726,6 @@ of the repositories listed at
726 Will merge to 'master'.
727
728
513 -* ab/fail-prereqs-in-test (2019-05-14) 1 commit
514 - (merged to 'next' on 2019-05-16 at d1be55f485)
515 - + tests: add a special setup where prerequisites fail
516 -
517 - Developer support to emulate unsatisfied prerequisites in tests to
518 - ensure that the remainer of the tests still succeeds when tests
519 - with prerequisites are skipped.
520 -
521 - Will merge to 'master'.
522 -
523 -
524 -* jk/help-unknown-ref-fix (2019-05-15) 2 commits
525 - (merged to 'next' on 2019-05-19 at e3e01160f7)
526 - + help_unknown_ref(): check for refname ambiguity
527 - + help_unknown_ref(): duplicate collected refnames
528 -
529 - Improve the code to show args with potential typo that cannot be
530 - interpreted as a commit-ish.
531 -
532 - Will merge to 'master'.
533 -
534 -
535 -* js/rebase-cleanup (2019-05-15) 5 commits
536 - (merged to 'next' on 2019-05-16 at ccfed8f263)
537 - + rebase: fold git-rebase--common into the -p backend
538 - + sequencer: the `am` and `rebase--interactive` scripts are gone
539 - + .gitignore: there is no longer a built-in `git-rebase--interactive`
540 - + t3400: stop referring to the scripted rebase
541 - + Drop unused git-rebase--am.sh
542 -
543 - Update supporting parts of "git rebase" to remove code that should
544 - no longer be used.
545 -
546 - Will merge to 'master'.
547 -
548 -
729 * jt/partial-clone-missing-ref-delta-base (2019-06-11) 4 commits
730 (merged to 'next' on 2019-06-12 at 95628af9bb)
731 + t5616: cover case of client having delta base
@@ -574,51 +754,6 @@ of the repositories listed at
754 Will merge to 'master'.
755
756
577 -* ew/update-server-info (2019-05-15) 1 commit
578 - (merged to 'next' on 2019-05-19 at bf4f2871ab)
579 - + update-server-info: avoid needless overwrites
580 -
581 - "git update-server-info" learned not to rewrite the file with the
582 - same contents.
583 -
584 - Will merge to 'master'.
585 -
586 -
587 -* nd/corrupt-worktrees (2019-05-15) 1 commit
588 - (merged to 'next' on 2019-05-16 at d92c25f800)
589 - + worktree add: be tolerant of corrupt worktrees
590 -
591 - "git worktree add" used to fail when another worktree connected to
592 - the same repository was corrupt, which has been corrected.
593 -
594 - Will merge to 'master'.
595 -
596 -
597 -* mh/import-transport-fd-fix (2019-05-16) 2 commits
598 - (merged to 'next' on 2019-05-19 at 5e86f92f7a)
599 - + Use xmmap_gently instead of xmmap in use_pack
600 - + dup() the input fd for fast-import used for remote helpers
601 -
602 - The ownership rule for the file descriptor to fast-import remote
603 - backend was mixed up, leading to unrelated file descriptor getting
604 - closed, which has been fixed.
605 -
606 - Will merge to 'master'.
607 -
608 -
609 -* ab/deprecate-R-for-dynpath (2019-05-19) 1 commit
610 - (merged to 'next' on 2019-05-19 at 944976e981)
611 - + Makefile: remove the NO_R_TO_GCC_LINKER flag
612 -
613 - The way of specifying the path to find dynamic libraries at runtime
614 - has been simplified. The old default to pass -R/path/to/dir has been
615 - replaced with the new default to pass -Wl,-rpath,/path/to/dir,
616 - which is the more recent GCC uses. Those who need to build with an
617 - old GCC can still use "CC_LD_DYNPATH=-R"
618 -
619 - Will merge to 'master'.
620 -
621 -
757 * ba/clone-remote-submodules (2019-05-28) 1 commit
758 (merged to 'next' on 2019-05-29 at 71972f94c2)
759 + clone: add `--remote-submodules` flag
@@ -644,7 +779,7 @@ of the repositories listed at
779 an existing commit-graph file now is closed before "gc" finalizes a
780 new instance to replace it.
781
647 - Will merge to 'next'.
782 + Will merge to 'next' and then to 'master'.
783
784
785 * ml/userdiff-rust (2019-05-30) 2 commits
@@ -659,98 +794,6 @@ of the repositories listed at
794 Will merge to 'master'.
795
796
662 -* pw/rebase-edit-message-for-replayed-merge (2019-05-19) 1 commit
663 - (merged to 'next' on 2019-05-19 at dc3e30641c)
664 - + rebase -r: always reword merge -c
665 -
666 - A "merge -c" instruction during "git rebase --rebase-merges" should
667 - give the user a chance to edit the log message, even when there is
668 - otherwise no need to create a new merge and replace the existing
669 - one (i.e. fast-forward instead), but did not. Which has been
670 - corrected.
671 -
672 - Will merge to 'master'.
673 -
674 -
675 -* sb/format-patch-base-patch-id-fix (2019-05-08) 2 commits
676 - (merged to 'next' on 2019-05-15 at 1ab7d2b71c)
677 - + format-patch: make --base patch-id output stable
678 - + format-patch: inform user that patch-id generation is unstable
679 -
680 - The "--base" option of "format-patch" computed the patch-ids for
681 - prerequisite patches in an unstable way, which has been updated to
682 - compute in a way that is compatible with "git patch-id --stable".
683 -
684 - Will merge to 'master'.
685 -
686 -
687 -* ab/send-email-transferencoding-fix (2019-05-29) 7 commits
688 - (merged to 'next' on 2019-05-29 at c8a99d18c0)
689 - + send-email: fix regression in sendemail.identity parsing
690 - + send-email: document --no-[to|cc|bcc]
691 - + send-email: fix broken transferEncoding tests
692 - + send-email: remove cargo-culted multi-patch pattern in tests
693 - (merged to 'next' on 2019-05-13 at 38c6a1e7e0)
694 - + send-email: do defaults -> config -> getopt in that order
695 - + send-email: rename the @bcclist variable for consistency
696 - + send-email: move the read_config() function above getopts
697 -
698 - Since "git send-email" learned to take 'auto' as the value for the
699 - transfer-encoding, it by mistake stopped honoring the values given
700 - to the configuration variables sendemail.transferencoding and/or
701 - sendemail.<ident>.transferencoding. This has been corrected to
702 - (finally) redoing the order of setting the default, reading the
703 - configuration and command line options.
704 -
705 - Will merge to 'master'.
706 -
707 -
708 -* dl/format-patch-notes-config (2019-05-17) 2 commits
709 - (merged to 'next' on 2019-05-19 at d3f6f1872b)
710 - + format-patch: teach format.notes config option
711 - + git-format-patch.txt: document --no-notes option
712 -
713 - "git format-patch" learns a configuration to set the default for
714 - its --notes=<ref> option.
715 -
716 - Will merge to 'master'.
717 -
718 -
719 -* jk/unused-params-final-batch (2019-05-13) 14 commits
720 - (merged to 'next' on 2019-05-15 at ef7435264c)
721 - + verify-commit: simplify parameters to run_gpg_verify()
722 - + show-branch: drop unused parameter from show_independent()
723 - + rev-list: drop unused void pointer from finish_commit()
724 - + remove_all_fetch_refspecs(): drop unused "remote" parameter
725 - + receive-pack: drop unused "commands" from prepare_shallow_update()
726 - + pack-objects: drop unused rev_info parameters
727 - + name-rev: drop unused parameters from is_better_name()
728 - + mktree: drop unused length parameter
729 - + wt-status: drop unused status parameter
730 - + read-cache: drop unused parameter from threaded load
731 - + clone: drop dest parameter from copy_alternates()
732 - + submodule: drop unused prefix parameter from some functions
733 - + builtin: consistently pass cmd_* prefix to parse_options
734 - + cmd_{read,write}_tree: rename "unused" variable that is used
735 -
736 - Remove many unused parameters throughout the codebase, with the
737 - ultimate aim to allow us compile with -Wunused-parameter cleanly.
738 -
739 - Will merge to 'master'.
740 -
741 -
742 -* nd/init-relative-template-fix (2019-05-13) 1 commit
743 - (merged to 'next' on 2019-05-15 at 4d5b17f712)
744 - + init: make --template path relative to $CWD
745 -
746 - A relative pathname given to "git init --template=<path> <repo>"
747 - ought to be relative to the directory "git init" gets invoked in,
748 - but it instead was made relative to the repository, which has been
749 - corrected.
750 -
751 - Will merge to 'master'.
752 -
753 -
797 * an/ignore-doc-update (2019-06-04) 1 commit
798 (merged to 'next' on 2019-06-06 at 8579d82c0d)
799 + gitignore.txt: make slash-rules more readable
@@ -762,31 +805,6 @@ of the repositories listed at
805 Will merge to 'master'.
806
807
765 -* en/fast-export-encoding (2019-05-14) 5 commits
766 - (merged to 'next' on 2019-05-16 at c88bd3edb5)
767 - + fast-export: do automatic reencoding of commit messages only if requested
768 - + fast-export: differentiate between explicitly UTF-8 and implicitly UTF-8
769 - + fast-export: avoid stripping encoding header if we cannot reencode
770 - + fast-import: support 'encoding' commit header
771 - + t9350: fix encoding test to actually test reencoding
772 -
773 - The "git fast-export/import" pair has been taught to handle commits
774 - with log messages in encoding other than UTF-8 better.
775 -
776 - Will merge to 'master'.
777 -
778 -
779 -* nd/merge-quit (2019-05-19) 2 commits
780 - (merged to 'next' on 2019-05-19 at 9880e7ee4e)
781 - + merge: add --quit
782 - + merge: remove drop_save() in favor of remove_merge_branch_state()
783 -
784 - "git merge" learned "--quit" option that cleans up the in-progress
785 - merge while leaving the working tree and the index still in a mess.
786 -
787 - Will merge to 'master'.
788 -
789 -
808 * pw/rebase-abort-clean-rewritten (2019-05-15) 4 commits
809 - rebase --abort/--quit: cleanup refs/rewritten
810 - sequencer: return errors from sequencer_remove_state()
@@ -810,7 +828,7 @@ of the repositories listed at
828 '+', similar to the way the currently checked out branch is shown
829 with '*' in front.
830
813 - Will merge to 'next'.
831 + Will merge to 'next' and then to 'master'.
832
833
834 * es/first-contrib-tutorial (2019-05-29) 3 commits
@@ -952,18 +970,6 @@ of the repositories listed at
970 cf. <xmqqa7fxionx.fsf@gitster-ct.c.googlers.com>
971
972
955 -* nd/worktree-name-sanitization (2019-05-15) 1 commit
956 - (merged to 'next' on 2019-05-16 at 9a2dd33122)
957 - + worktree add: sanitize worktree names
958 -
959 - In recent versions of Git, per-worktree refs are exposed in
960 - refs/worktrees/<wtname>/ hierarchy, which means that worktree names
961 - must be a valid refname component. The code now sanitizes the names
962 - given to worktrees, to make sure these refs are well-formed.
963 -
964 - Will merge to 'master'.
965 -
966 -
973 * ds/commit-graph-write-refactor (2019-06-12) 11 commits
974 - commit-graph: extract write_commit_graph_file()
975 - commit-graph: extract copy_oids_to_commits()
@@ -980,7 +986,7 @@ of the repositories listed at
986
987 Renamed from commit-graph-format-v2 and changed scope.
988
983 - Will merge to 'next'.
989 + Will merge to 'next' and then to 'master'.
990
991
992 * br/blame-ignore (2019-06-10) 9 commits
@@ -997,6 +1003,8 @@ of the repositories listed at
1003 "git blame" learned to "ignore" commits in the history, whose
1004 effects (as well as their presence) get ignored.
1005
1006 + Except for a minor nit in its tests, this seems to be more or less
1007 + ready for 'next'.
1008 cf. <20190610153014.42055-1-brho@google.com> (v8)
1009
1010 --------------------------------------------------