1 To: git@vger.kernel.org
3 Subject: What's cooking in git.git (Aug 2013, #05; Mon, 19)
4 X-master-at: 4d06473928ee574910accbde05c19ef2263abdf6
5 X-next-at: fb33910d190a5cff4020f44651fc3a51f55c0a1f
7 What's cooking in git.git (Aug 2013, #05; Mon, 19)
8 --------------------------------------------------
10 Here are the topics that have been cooking. Commits prefixed with
11 '-' are only in 'pu' (proposed updates) while commits prefixed with
14 An extra release candidate -rc4 has been tagged and pushed out.
15 Hopefully this will be the last one before the final release of
16 1.8.4. As I expect we will have two more cycles of 1.8.x by the end
17 of the year and then 2.0 early next year, we may want to merge these
18 "for 2.0" topics to 'next' for real, starting the next cycle.
20 You can find the changes described here in the integration branches
21 of the repositories listed at
23 http://git-blame.blogspot.com/p/git-public-repositories.html
25 --------------------------------------------------
28 * es/rebase-i-respect-core-commentchar (2013-08-18) 1 commit
29 - rebase -i: fix cases ignoring core.commentchar
31 Will merge to and cook in 'next'.
34 * jx/branch-vv-always-compare-with-upstream (2013-08-18) 3 commits
35 - status: always show tracking branch even no change
36 - branch: mark missing tracking branch as gone
37 - branch: not report invalid tracking branch
40 * nd/fetch-into-shallow (2013-08-18) 6 commits
41 - list-objects: mark more commits as edges in mark_edges_uninteresting
42 - list-objects: reduce one argument in mark_edges_uninteresting
43 - upload-pack: delegate rev walking in shallow fetch to pack-objects
44 - shallow: add setup_temporary_shallow()
45 - shallow: only add shallow graft points to new shallow file
46 - move setup_alternate_shallow and write_shallow_commits to shallow.c
49 * sb/diff-delta-remove-needless-comparison (2013-08-18) 1 commit
50 - create_delta_index: simplify condition always evaluating to true
52 Will merge to and cook in 'next'.
55 * sg/bash-prompt-lf-in-cwd-test (2013-08-18) 1 commit
56 - bash prompt: test the prompt with newline in repository path
58 Will merge to and cook in 'next'.
61 * jl/some-submodule-config-are-not-boolean (2013-08-19) 1 commit
62 - avoid segfault on submodule.*.path set to an empty "true"
64 Will merge to and cook in 'next'.
66 --------------------------------------------------
69 * tf/gitweb-ss-tweak (2013-07-15) 4 commits
70 - gitweb: make search help link less ugly
71 - gitweb: omit the repository owner when it is unset
72 - gitweb: vertically centre contents of page footer
73 - gitweb: ensure OPML text fits inside its box
78 * rj/read-default-config-in-show-ref-pack-refs (2013-06-17) 3 commits
79 - ### DONTMERGE: needs better explanation on what config they need
80 - pack-refs.c: Add missing call to git_config()
81 - show-ref.c: Add missing call to git_config()
83 The changes themselves are probably good, but it is unclear what
84 basic setting needs to be read for which exact operation.
86 Waiting for clarification.
90 * jh/shorten-refname (2013-05-07) 4 commits
91 - t1514: refname shortening is done after dereferencing symbolic refs
92 - shorten_unambiguous_ref(): Fix shortening refs/remotes/origin/HEAD to origin
93 - t1514: Demonstrate failure to correctly shorten "refs/remotes/origin/HEAD"
94 - t1514: Add tests of shortening refnames in strict/loose mode
96 When remotes/origin/HEAD is not a symbolic ref, "rev-parse
97 --abbrev-ref remotes/origin/HEAD" ought to show "origin", not
98 "origin/HEAD", which is fixed with this series (if it is a symbolic
99 ref that points at remotes/origin/something, then it should show
100 "origin/something" and it already does).
102 Expecting a reroll, as an early part of a larger series.
106 * jk/list-objects-sans-blobs (2013-06-06) 4 commits
107 . archive: ignore blob objects when checking reachability
108 . list-objects: optimize "revs->blob_objects = 0" case
109 . upload-archive: restrict remote objects with reachability check
110 . clear parsed flag when we free tree buffers
112 Attempt to allow "archive --remote=$there $arbitrary_sha1" while
113 keeping the reachability safety.
115 Seems to break some tests in a trivial and obvious way.
118 * mg/more-textconv (2013-05-10) 7 commits
119 - grep: honor --textconv for the case rev:path
120 - grep: allow to use textconv filters
121 - t7008: demonstrate behavior of grep with textconv
122 - cat-file: do not die on --textconv without textconv filters
123 - show: honor --textconv for blobs
124 - diff_opt: track whether flags have been set explicitly
125 - t4030: demonstrate behavior of show with textconv
127 Make "git grep" and "git show" pay attention to --textconv when
128 dealing with blob objects.
130 I thought this was pretty well designed and executed, but it seems
131 there are some doubts on the list; kicked back to 'pu'.
134 * jc/format-patch (2013-04-22) 2 commits
135 - format-patch: --inline-single
136 - format-patch: rename "no_inline" field
138 A new option to send a single patch to the standard output to be
139 appended at the bottom of a message. I personally have no need for
140 this, but it was easy enough to cobble together. Tests, docs and
141 stripping out more MIMEy stuff are left as exercises to interested
144 Not ready for inclusion.
146 Will discard unless we hear from anybody who is interested in
147 tying its loose ends.
150 * jk/gitweb-utf8 (2013-04-08) 4 commits
151 - gitweb: Fix broken blob action parameters on blob/commitdiff pages
152 - gitweb: Don't append ';js=(0|1)' to external links
153 - gitweb: Make feed title valid utf8
154 - gitweb: Fix utf8 encoding for blob_plain, blobdiff_plain, commitdiff_plain, and patch
156 Various fixes to gitweb.
158 Drew Northup volunteered to take a look into this.
162 * jc/show-branch (2013-06-07) 5 commits
163 - show-branch: use commit slab to represent bitflags of arbitrary width
164 - show-branch.c: remove "all_mask"
165 - show-branch.c: abstract out "flags" operation
166 - show-branch.c: lift all_mask/all_revs to a global static
167 - show-branch.c: update comment style
169 Waiting for the final step to lift the hard-limit before sending it out.
171 --------------------------------------------------
174 * jc/ls-files-killed-optim (2013-08-15) 3 commits
175 - t3010: update to demonstrate "ls-files -k" optimization pitfalls
176 - ls-files -k: a directory only can be killed if the index has a non-directory
177 - dir.c: use the cache_* macro to access the current index
179 "git ls-files -k" needs to crawl only the part of the working tree
180 that may overlap the paths in the index to find killed files, but
181 shared code with the logic to find all the untracked files, which
182 made it unnecessarily inefficient.
184 Will merge to and cook in 'next'.
187 * es/blame-L-twice (2013-08-06) 16 commits
188 (merged to 'next' on 2013-08-08 at 9d6f821)
189 + line-range: reject -L line numbers less than 1
190 + t8001/t8002: blame: add tests of -L line numbers less than 1
191 + line-range: teach -L^:RE to search from start of file
192 + line-range: teach -L:RE to search from end of previous -L range
193 + line-range: teach -L^/RE/ to search from start of file
194 + line-range-format.txt: document -L/RE/ relative search
195 + log: teach -L/RE/ to search from end of previous -L range
196 + blame: teach -L/RE/ to search from end of previous -L range
197 + line-range: teach -L/RE/ to search relative to anchor point
198 + blame: document multiple -L support
199 + t8001/t8002: blame: add tests of multiple -L options
200 + blame: accept multiple -L ranges
201 + blame: inline one-line function into its lone caller
202 + range-set: publish API for re-use by git-blame -L
203 + line-range-format.txt: clarify -L:regex usage form
204 + git-log.txt: place each -L option variation on its own line
205 (this branch is used by es/contacts-blame-L-multi; uses es/blame-L-more.)
207 Teaches "git blame" to take more than one -L ranges.
212 * mm/no-shell-escape-in-die-message (2013-08-07) 1 commit
213 (merged to 'next' on 2013-08-08 at bddff86)
214 + die_with_status: use "printf '%s\n'", not "echo"
216 Fixes a minor bug in "git rebase -i" (there could be others, as the
217 root cause is pretty generic) where the code feeds a random, data
218 dependeant string to 'echo' and expects it to come out literally.
223 * tr/fd-gotcha-fixes (2013-08-06) 1 commit
224 (merged to 'next' on 2013-08-08 at f452c5b)
225 + t0070: test that git_mkstemps correctly checks return value of open()
227 Finishing touches to an earlier fix already in 'master'.
232 * ap/remote-hg-tilde-is-home-directory (2013-08-09) 1 commit
233 (merged to 'next' on 2013-08-14 at cd963e3)
234 + remote-hg: fix path when cloning with tilde expansion
239 * es/rebase-i-no-abbrev (2013-08-18) 4 commits
240 - rebase -i: fix core.commentchar regression
241 (merged to 'next' on 2013-08-14 at d449e0e)
242 + rebase: interactive: fix short SHA-1 collision
243 + t3404: rebase: interactive: demonstrate short SHA-1 collision
244 + t3404: restore specialized rebase-editor following commentchar test
249 * fc/remote-hg-shared-setup (2013-08-11) 2 commits
250 (merged to 'next' on 2013-08-14 at aae6858)
251 + remote-hg: add shared repo upgrade
252 + remote-hg: ensure shared repo is initialized
257 * jc/transport-do-not-use-connect-twice-in-fetch (2013-08-07) 5 commits
258 - fetch: work around "transport-take-over" hack
259 - fetch: refactor code that fetches leftover tags
260 - fetch: refactor code that prepares a transport
261 - fetch: rename file-scope global "transport" to "gtransport"
262 - t5802: add test for connect helper
264 The auto-tag-following code in "git fetch" tries to reuse the same
265 transport twice when the serving end does not cooperate and does
266 not give tags that point to commits that are asked for as part of
267 the primary transfer. Unfortunately, Git-aware transport helper
268 interface is not designed to be used more than once, hence this
269 does not work over smart-http transfer.
271 I'll try to see if this workaround can be done in a cleaner way
272 before merging this to 'next'.
275 * ks/p4-view-spec (2013-08-11) 3 commits
277 - git p4: implement view spec wildcards with "p4 where"
278 - git p4 test: sanitize P4CHARSET
283 * mm/war-on-whatchanged (2013-08-13) 2 commits
284 (merged to 'next' on 2013-08-14 at fe77c11)
285 + whatchanged: document its historical nature
286 + core-tutorial: trim the section on Inspecting Changes
291 * nd/gc-lock-against-each-other (2013-08-09) 1 commit
292 (merged to 'next' on 2013-08-14 at 6999651)
293 + gc: reject if another gc is running, unless --force is given
298 * rt/doc-merge-file-diff3 (2013-08-09) 1 commit
299 (merged to 'next' on 2013-08-14 at 1e5847b)
300 + Documentation/git-merge-file: document option "--diff3"
305 * sb/misc-cleanup (2013-08-09) 3 commits
306 (merged to 'next' on 2013-08-14 at 9e7ff9a)
307 + rm: remove unneeded null pointer check
308 + diff: fix a possible null pointer dereference
309 + diff: remove ternary operator evaluating always to true
314 * aj/p4-symlink-lose-nl (2013-08-12) 1 commit
315 (merged to 'next' on 2013-08-14 at a4959b7)
316 + git-p4: Fix occasional truncation of symlink contents.
321 * es/contacts-blame-L-multi (2013-08-13) 3 commits
322 (merged to 'next' on 2013-08-14 at 551ed3d)
323 + contacts: reduce git-blame invocations
324 + contacts: gather all blame sources prior to invoking git-blame
325 + contacts: validate hunk length earlier
326 (this branch uses es/blame-L-more and es/blame-L-twice.)
331 * fc/unpack-trees-leakfix (2013-08-13) 1 commit
332 (merged to 'next' on 2013-08-14 at 989f329)
333 + unpack-trees: plug a memory leak
338 * nd/push-no-thin (2013-08-13) 1 commit
339 (merged to 'next' on 2013-08-14 at b2d0fa9)
340 + push: respect --no-thin
345 * sh/pull-rebase-preserve (2013-08-13) 1 commit
346 (merged to 'next' on 2013-08-14 at 99a85dd)
347 + pull: Allow pull to preserve merges when rebasing.
352 * es/blame-L-more (2013-08-05) 11 commits
353 (merged to 'next' on 2013-08-06 at 2679f3e)
354 + blame: reject empty ranges -L,+0 and -L,-0
355 + t8001/t8002: blame: demonstrate acceptance of bogus -L,+0 and -L,-0
356 + blame: reject empty ranges -LX,+0 and -LX,-0
357 + t8001/t8002: blame: demonstrate acceptance of bogus -LX,+0 and -LX,-0
358 + log: fix -L bounds checking bug
359 + t4211: retire soon-to-be unimplementable tests
360 + t4211: log: demonstrate -L bounds checking bug
361 + blame: fix -L bounds checking bug
362 + t8001/t8002: blame: add empty file & partial-line tests
363 + t8001/t8002: blame: demonstrate -L bounds checking bug
364 + t8001/t8002: blame: decompose overly-large test
365 (this branch is used by es/blame-L-twice and es/contacts-blame-L-multi.)
367 More fixes to the code to parse the "-L" option in "log" and "blame".
372 * jk/cat-file-batch-optim (2013-08-05) 1 commit
373 (merged to 'next' on 2013-08-06 at cd48621)
374 + cat-file: only split on whitespace when %(rest) is used
376 Rework the reverted change to `cat-file --batch-check`.
381 * jn/post-receive-utf8 (2013-08-05) 3 commits
382 - hooks/post-receive-email: set declared encoding to utf-8
383 - hooks/post-receive-email: force log messages in UTF-8
384 - hooks/post-receive-email: use plumbing instead of git log/show
386 Update post-receive-email script to make sure the message contents
387 and pathnames are encoded consistently in UTF-8.
389 I have a feeling that it is a lost cause to solve the issue the
390 topic tries to address in general, because the patch text can have
391 payload in any encodings that are different from either the
392 pathnames or the log message. Patches that touch paths that use an
393 encoding that conflicts with the encoding of the payload and/or the
394 log message could be transferred with core.quotepath set and patch
395 generated as all binary, but that would be pretty much useless.
398 * sb/parseopt-boolean-removal (2013-08-07) 9 commits
399 (merged to 'next' on 2013-08-08 at b138a2d)
400 + revert: use the OPT_CMDMODE for parsing, reducing code
401 + checkout-index: fix negations of even numbers of -n
402 + config parsing options: allow one flag multiple times
403 + hash-object: replace stdin parsing OPT_BOOLEAN by OPT_COUNTUP
404 + branch, commit, name-rev: ease up boolean conditions
405 + checkout: remove superfluous local variable
406 + log, format-patch: parsing uses OPT__QUIET
407 + Replace deprecated OPT_BOOLEAN by OPT_BOOL
408 + Remove deprecated OPTION_BOOLEAN for parsing arguments
409 (this branch uses jc/parseopt-command-modes.)
411 Convert most uses of OPT_BOOLEAN/OPTION_BOOLEAN that can use
412 OPT_BOOL/OPTION_BOOLEAN which have much saner semantics, and turn
413 remaining ones into OPT_SET_INT, OPT_COUNTUP, etc. as necessary.
418 * mb/docs-favor-en-us (2013-08-01) 1 commit
419 (merged to 'next' on 2013-08-06 at 763d868)
420 + Provide some linguistic guidance for the documentation.
422 Declare that the official grammar & spelling of the source of this
423 project is en_US, but strongly discourage patches only to "fix"
424 existing en_UK strings to avoid unnecessary churns.
429 * jc/parseopt-command-modes (2013-07-30) 2 commits
430 (merged to 'next' on 2013-08-08 at 43abba9)
431 + tag: use OPT_CMDMODE
432 + parse-options: add OPT_CMDMODE()
433 (this branch is used by sb/parseopt-boolean-removal.)
435 Many commands use --dashed-option as a operation mode selector
436 (e.g. "git tag --delete") that the user can use at most one
437 (e.g. "git tag --delete --verify" is a nonsense) and you cannot
438 negate (e.g. "git tag --no-delete" is a nonsense). Make it easier
439 for users of parse_options() to enforce these restrictions.
444 * tr/log-full-diff-keep-true-parents (2013-08-05) 2 commits
445 (merged to 'next' on 2013-08-08 at 2fb633f)
446 + log: use true parents for diff when walking reflogs
447 + log: use true parents for diff even when rewriting
449 Output from "git log --full-diff -- <pathspec>" looked strange,
450 because comparison was done with the previous ancestor that touched
451 the specified <pathspec>, causing the patches for paths outside the
452 pathspec to show more than the single commit has changed.
454 Tweak "git reflog -p" for the same reason using the same mechanism.
459 * bc/unuse-packfile (2013-08-02) 2 commits
460 (merged to 'next' on 2013-08-06 at 01aa4fd)
461 + Don't close pack fd when free'ing pack windows
462 + sha1_file: introduce close_one_pack() to close packs on fd pressure
464 Handle memory pressure and file descriptor pressure separately when
465 deciding to release pack windows to honor resource limits.
470 * da/darwin (2013-08-05) 3 commits
471 (merged to 'next' on 2013-08-05 at 5c54cfa)
472 + OS X: Fix redeclaration of die warning
473 + Makefile: Fix APPLE_COMMON_CRYPTO with BLK_SHA1
474 (merged to 'next' on 2013-08-01 at d151a94)
475 + imap-send: use Apple's Security framework for base64 encoding
480 * nd/sq-quote-buf (2013-07-30) 3 commits
481 (merged to 'next' on 2013-08-01 at dc7934a)
482 + quote: remove sq_quote_print()
483 + tar-tree: remove dependency on sq_quote_print()
484 + for-each-ref, quote: convert *_quote_print -> *_quote_buf
486 Code simplification as a preparatory step to something larger.
491 * jc/url-match (2013-08-09) 7 commits
492 (merged to 'next' on 2013-08-12 at 381d358)
493 + builtin/config.c: compilation fix
494 (merged to 'next' on 2013-08-08 at c1ee470)
495 + config: "git config --get-urlmatch" parses section.<url>.key
496 + builtin/config: refactor collect_config()
497 + config: parse http.<url>.<variable> using urlmatch
498 + config: add generic callback wrapper to parse section.<url>.key
499 + config: add helper to normalize and match URLs
500 + http.c: fix parsing of http.sslCertPasswordProtected variable
502 Allow section.<urlpattern>.var configuration variables to be
503 treated as a "virtual" section.var given a URL, and use the
504 mechanism to enhance http.* configuration variables.
506 This is a reroll of Kyle J. McKay's work.
511 * jl/submodule-mv (2013-08-06) 5 commits
512 (merged to 'next' on 2013-08-08 at 1a57bdf)
513 + rm: delete .gitmodules entry of submodules removed from the work tree
514 + mv: update the path entry in .gitmodules for moved submodules
515 + submodule.c: add .gitmodules staging helper functions
516 + mv: move submodules using a gitfile
517 + mv: move submodules together with their work trees
518 (this branch uses nd/magic-pathspec.)
520 "git mv A B" when moving a submodule A does "the right thing",
521 inclusing relocating its working tree and adjusting the paths in
522 the .gitmodules file.
527 * nd/clone-connectivity-shortcut (2013-07-23) 1 commit
528 (merged to 'next' on 2013-07-30 at 8d16721)
529 + smart http: use the same connectivity check on cloning
534 * db/http-savecookies (2013-08-05) 2 commits
535 (merged to 'next' on 2013-08-05 at be6f3a3)
536 + t5551: Remove header from curl cookie file
537 (merged to 'next' on 2013-07-30 at d2ad6e5)
538 + http: add http.savecookies option to write out HTTP cookies
543 * sb/fsck-opt-bool (2013-07-29) 1 commit
544 (merged to 'next' on 2013-07-30 at 8a9964c)
545 + fsck: Replace deprecated OPT_BOOLEAN by OPT_BOOL
549 sb/parseopt-boolean-removal contains the moral equivalent of this
553 * rr/feed-real-path-to-editor (2013-07-29) 1 commit
554 (merged to 'next' on 2013-07-30 at 72c9bcb)
555 + editor: use canonicalized absolute path
560 * rj/doc-rev-parse (2013-07-22) 2 commits
561 (merged to 'next' on 2013-07-22 at 8188667)
562 + rev-parse(1): logically group options
563 + rev-parse: remove restrictions on some options
568 * jk/fast-import-empty-ls (2013-06-23) 4 commits
569 (merged to 'next' on 2013-07-22 at d26dab6)
570 + fast-import: allow moving the root tree
571 + fast-import: allow ls or filecopy of the root tree
572 + fast-import: set valid mode on root tree in "ls"
573 + t9300: document fast-import empty path issues
578 * jc/diff-filter-negation (2013-07-19) 6 commits
579 (merged to 'next' on 2013-07-22 at 80d167d)
580 + diff: deprecate -q option to diff-files
581 + diff: allow lowercase letter to specify what change class to exclude
582 + diff: reject unknown change class given to --diff-filter
583 + diff: preparse --diff-filter string argument
584 + diff: factor out match_filter()
585 + diff: pass the whole diff_options to diffcore_apply_filter()
586 (this branch is used by jc/hold-diff-remove-q-synonym-for-no-deletion.)
588 Teach "git diff --diff-filter" to express "I do not want to see
589 these classes of changes" more directly by listing only the
590 unwanted ones in lowercase (e.g. "--diff-filter=d" will show
591 everything but deletion) and deprecate "diff-files -q" which did
592 the same thing as "--diff-filter=d".
597 * nd/magic-pathspec (2013-07-23) 48 commits
598 (merged to 'next' on 2013-07-23 at 784c3a1)
599 + rm: do not set a variable twice without intermediate reading.
600 (merged to 'next' on 2013-07-22 at 39f7194)
601 + t6131 - skip tests if on case-insensitive file system
602 + parse_pathspec: accept :(icase)path syntax
603 + pathspec: support :(glob) syntax
604 + pathspec: make --literal-pathspecs disable pathspec magic
605 + pathspec: support :(literal) syntax for noglob pathspec
606 + kill limit_pathspec_to_literal() as it's only used by parse_pathspec()
607 + parse_pathspec: preserve prefix length via PATHSPEC_PREFIX_ORIGIN
608 + parse_pathspec: make sure the prefix part is wildcard-free
609 + rename field "raw" to "_raw" in struct pathspec
610 + tree-diff: remove the use of pathspec's raw[] in follow-rename codepath
611 + remove match_pathspec() in favor of match_pathspec_depth()
612 + remove init_pathspec() in favor of parse_pathspec()
613 + remove diff_tree_{setup,release}_paths
614 + convert common_prefix() to use struct pathspec
615 + convert add_files_to_cache to take struct pathspec
616 + convert {read,fill}_directory to take struct pathspec
617 + convert refresh_index to take struct pathspec
618 + convert report_path_error to take struct pathspec
619 + checkout: convert read_tree_some to take struct pathspec
620 + convert unmerge_cache to take struct pathspec
621 + convert run_add_interactive to use struct pathspec
622 + convert read_cache_preload() to take struct pathspec
623 + line-log: convert to use parse_pathspec
624 + reset: convert to use parse_pathspec
625 + add: convert to use parse_pathspec
626 + check-ignore: convert to use parse_pathspec
627 + archive: convert to use parse_pathspec
628 + ls-files: convert to use parse_pathspec
629 + rm: convert to use parse_pathspec
630 + checkout: convert to use parse_pathspec
631 + rerere: convert to use parse_pathspec
632 + status: convert to use parse_pathspec
633 + commit: convert to use parse_pathspec
634 + clean: convert to use parse_pathspec
635 + guard against new pathspec magic in pathspec matching code
636 + parse_pathspec: support prefixing original patterns
637 + parse_pathspec: support stripping/checking submodule paths
638 + parse_pathspec: support stripping submodule trailing slashes
639 + parse_pathspec: add special flag for max_depth feature
640 + convert some get_pathspec() calls to parse_pathspec()
641 + parse_pathspec: add PATHSPEC_PREFER_{CWD,FULL} flags
642 + parse_pathspec: save original pathspec for reporting
643 + add parse_pathspec() that converts cmdline args to struct pathspec
644 + pathspec: add copy_pathspec
645 + pathspec: i18n-ize error strings in pathspec parsing code
646 + move struct pathspec and related functions to pathspec.[ch]
647 + clean: remove unused variable "seen"
648 (this branch is used by jl/submodule-mv.)
650 Use "struct pathspec" interface in more places, instead of array of
651 characters, the latter of which cannot express magic pathspecs
652 (e.g. ":(icase)makefile" that matches both Makefile and makefile).
657 * jc/check-x-z (2013-07-11) 4 commits
658 (merged to 'next' on 2013-07-15 at 62751bc)
659 + check-attr -z: a single -z should apply to both input and output
660 + check-ignore -z: a single -z should apply to both input and output
661 + check-attr: the name of the character is NUL, not NULL
662 + check-ignore: the name of the character is NUL, not NULL
664 "git check-ignore -z" applied the NUL termination to both its input
665 (with --stdin) and its output, but "git check-attr -z" ignored the
666 option on the output side.
668 This is potentially a backward incompatible fix. I am tempted to
669 merge this to and keep it in 'next' for a while to see if anybody
670 screams before deciding if we want to do anything to help existing
671 users (there may be none).
676 * ms/fetch-prune-configuration (2013-07-18) 1 commit
677 (merged to 'next' on 2013-07-21 at 7ffb690)
678 + fetch: make --prune configurable
680 Allow fetch.prune and remote.*.prune configuration variables to be set,
681 and "git fetch" to behave as if "--prune" is given.
683 "git fetch" that honors remote.*.prune is fine, but I wonder if we
684 should somehow make "git push" aware of it as well. Perhaps
685 remote.*.prune should not be just a boolean, but a 4-way "none",
686 "push", "fetch", "both"?
691 * km/svn-1.8-serf-only (2013-07-18) 3 commits
692 (merged to 'next' on 2013-07-19 at ab8d034)
693 + Git.pm: revert _temp_cache use of temp_is_locked
694 (merged to 'next' on 2013-07-12 at 70e0b95)
695 + git-svn: allow git-svn fetching to work using serf
696 + Git.pm: add new temp_is_locked function
698 Subversion 1.8.0 that was recently released breaks older subversion
699 clients coming over http/https in various ways.
704 * tr/merge-recursive-index-only (2013-07-07) 3 commits
705 - merge-recursive: -Xindex-only to leave worktree unchanged
706 - merge-recursive: untangle double meaning of o->call_depth
707 - merge-recursive: remove dead conditional in update_stages()
709 Will hold, until we get any user.
712 * jc/push-cas (2013-08-02) 9 commits
713 (merged to 'next' on 2013-08-05 at 34f4cf3)
714 + push: teach --force-with-lease to smart-http transport
715 + send-pack: fix parsing of --force-with-lease option
716 (merged to 'next' on 2013-08-01 at a7604ee)
717 + t5540/5541: smart-http does not support "--force-with-lease"
718 (merged to 'next' on 2013-07-23 at 72371b0)
719 + t5533: test "push --force-with-lease"
720 + push --force-with-lease: tie it all together
721 + push --force-with-lease: implement logic to populate old_sha1_expect[]
722 + remote.c: add command line option parser for "--force-with-lease"
723 + builtin/push.c: use OPT_BOOL, not OPT_BOOLEAN
724 + cache.h: move remote/connect API out of it
726 Allow a safer "rewind of the remote tip" push than blind "--force",
727 by requiring that the overwritten remote ref to be unchanged since
728 the new history to replace it was prepared.
730 The machinery is more or less ready. The "--force" option is again
731 the big red button to override any safety, thanks to J6t's sanity
732 (the original round allowed --lockref to defeat --force).
734 The logic to choose the default implemented here is fragile
735 (e.g. "git fetch" after seeing a failure will update the
736 remote-tracking branch and will make the next "push" pass,
737 defeating the safety pretty easily). It is suitable only for the
738 simplest workflows, and it may hurt users more than it helps them.
743 * jc/pull-training-wheel (2013-07-19) 1 commit
744 (merged to 'next' on 2013-07-22 at 748180e)
745 + pull: require choice between rebase/merge on non-fast-forward pull
747 Make "git pull" (without arguments that say what branch to
748 integrate from where) refuse with "it does not fast forward; choose
749 between 'pull --merge' and 'pull --rebase'".
753 http://thread.gmane.org/gmane.comp.version-control.git/225146/focus=225326
754 http://thread.gmane.org/gmane.comp.version-control.git/225146/focus=229221
761 * jc/reflog-doc (2013-06-19) 1 commit
762 - setup_reflog_action: document the rules for using GIT_REFLOG_ACTION
764 Document rules to use GIT_REFLOG_ACTION variable in the scripted
765 Porcelain. git-rebase--interactive locally violates this, but it
766 is a leaf user that does not call out to or dot-sources other
767 scripts, so fixing it is not all that urgent.
770 * jn/add-2.0-u-A-sans-pathspec (2013-04-26) 1 commit
771 - git add: -u/-A now affects the entire working tree
773 Will cook in 'next' until Git 2.0.
776 * jc/core-checkstat-2.0 (2013-05-06) 1 commit
777 - core.statinfo: remove as promised in Git 2.0
779 Will cook in 'next' until Git 2.0.
782 * jc/push-2.0-default-to-simple (2013-06-18) 1 commit
783 - push: switch default from "matching" to "simple"
785 Will cook in 'next' until Git 2.0.
788 * jc/add-2.0-ignore-removal (2013-04-22) 1 commit
789 - git add <pathspec>... defaults to "-A"
791 Updated endgame for "git add <pathspec>" that defaults to "--all"
792 aka "--no-ignore-removal".
794 Will cook in 'next' until Git 2.0.
797 * jc/hold-diff-remove-q-synonym-for-no-deletion (2013-07-19) 1 commit
798 - diff: remove "diff-files -q" in a version of Git in a distant future
799 (this branch uses jc/diff-filter-negation.)
801 Will cook in 'next' until a distant future.
803 --------------------------------------------------
806 * jh/remote-hg-fetch-fix (2013-07-25) 2 commits
807 (merged to 'next' on 2013-07-25 at 33161ad)
808 + Revert "remotes-hg: bugfix for fetching non local remotes"
809 (merged to 'next' on 2013-07-24 at 9c96641)
810 + remotes-hg: bugfix for fetching non local remotes
814 Waiting for the final patch to replace, after discussion settles.
817 * km/http-curl-config-per-url (2013-07-31) 7 commits
818 (merged to 'next' on 2013-07-31 at e88aad5)
819 + Revert km/http-curl-config-per-url in its entirety
820 (merged to 'next' on 2013-07-25 at 531e8cb)
821 + docs: update http.<url>.* options documentation
822 (merged to 'next' on 2013-07-23 at f84689b)
823 + config: allow http.<url>.* any user matching
824 + tests: add new test for the url_normalize function
825 + config: improve support for http.<url>.* settings
826 + config: add support for http.<url>.* settings
827 + http.c: fix parsing of http.sslCertPasswordProtected variable
829 Reverted to be replaced with a reroll (jc/url-match topic).