git/git.git
19 months agoMerge branch 'sg/travis-gcc-4.8'
Junio C Hamano [Mon, 29 Jul 2019 19:39:13 +0000 (12:39 -0700)]
Merge branch 'sg/travis-gcc-4.8'

Add a job to build with a tad older GCC to make sure we are still
buildable.

* sg/travis-gcc-4.8:
  travis-ci: build with GCC 4.8 as well

19 months agoMerge branch 'bb/grep-pcre2-bug-message-fix'
Junio C Hamano [Mon, 29 Jul 2019 19:39:13 +0000 (12:39 -0700)]
Merge branch 'bb/grep-pcre2-bug-message-fix'

BUG() message fix.

The codepath may want to just simply be removed, though.

* bb/grep-pcre2-bug-message-fix:
  grep: print the pcre2_jit_on value

19 months agoSync with maint
Junio C Hamano [Thu, 25 Jul 2019 21:34:17 +0000 (14:34 -0700)]
Sync with maint

Signed-off-by: Junio C Hamano <gitster@pobox.com>
19 months agoFlush fixes up to the third batch post 2.22.0
Junio C Hamano [Thu, 25 Jul 2019 21:32:36 +0000 (14:32 -0700)]
Flush fixes up to the third batch post 2.22.0

Signed-off-by: Junio C Hamano <gitster@pobox.com>
19 months agoMerge branch 'ab/hash-object-doc' into maint
Junio C Hamano [Thu, 25 Jul 2019 21:27:16 +0000 (14:27 -0700)]
Merge branch 'ab/hash-object-doc' into maint

Doc update.

* ab/hash-object-doc:
  hash-object doc: stop mentioning git-cvsimport

19 months agoMerge branch 'cm/send-email-document-req-modules' into maint
Junio C Hamano [Thu, 25 Jul 2019 21:27:15 +0000 (14:27 -0700)]
Merge branch 'cm/send-email-document-req-modules' into maint

A doc update.

* cm/send-email-document-req-modules:
  send-email: update documentation of required Perl modules

19 months agoMerge branch 'sw/git-p4-unshelve-branched-files' into maint
Junio C Hamano [Thu, 25 Jul 2019 21:27:15 +0000 (14:27 -0700)]
Merge branch 'sw/git-p4-unshelve-branched-files' into maint

"git p4" update.

* sw/git-p4-unshelve-branched-files:
  git-p4: allow unshelving of branched files

19 months agoMerge branch 'js/bisect-helper-check-get-oid-return-value' into maint
Junio C Hamano [Thu, 25 Jul 2019 21:27:14 +0000 (14:27 -0700)]
Merge branch 'js/bisect-helper-check-get-oid-return-value' into maint

Code cleanup.

* js/bisect-helper-check-get-oid-return-value:
  bisect--helper: verify HEAD could be parsed before continuing

19 months agoMerge branch 'es/git-debugger-doc' into maint
Junio C Hamano [Thu, 25 Jul 2019 21:27:14 +0000 (14:27 -0700)]
Merge branch 'es/git-debugger-doc' into maint

Doc update.

* es/git-debugger-doc:
  doc: hint about GIT_DEBUGGER in CodingGuidelines

19 months agoMerge branch 'mo/clang-format-for-each-update' into maint
Junio C Hamano [Thu, 25 Jul 2019 21:27:14 +0000 (14:27 -0700)]
Merge branch 'mo/clang-format-for-each-update' into maint

The list of for-each like macros used by clang-format has been
updated.

* mo/clang-format-for-each-update:
  clang-format: use git grep to generate the ForEachMacros list

19 months agoMerge branch 'md/url-parse-harden' into maint
Junio C Hamano [Thu, 25 Jul 2019 21:27:13 +0000 (14:27 -0700)]
Merge branch 'md/url-parse-harden' into maint

The URL decoding code has been updated to avoid going past the end
of the string while parsing %-<hex>-<hex> sequence.

* md/url-parse-harden:
  url: do not allow %00 to represent NUL in URLs
  url: do not read past end of buffer

19 months agoMerge branch 'an/ignore-doc-update' into maint
Junio C Hamano [Thu, 25 Jul 2019 21:27:13 +0000 (14:27 -0700)]
Merge branch 'an/ignore-doc-update' into maint

The description about slashes in gitignore patterns (used to
indicate things like "anchored to this level only" and "only
matches directories") has been revamped.

* an/ignore-doc-update:
  gitignore.txt: make slash-rules more readable

19 months agoMerge branch 'md/list-objects-filter-memfix' into maint
Junio C Hamano [Thu, 25 Jul 2019 21:27:12 +0000 (14:27 -0700)]
Merge branch 'md/list-objects-filter-memfix' into maint

The filter_data used in the list-objects-filter (which manages a
lazily sparse clone repository) did not use the dynamic array API
correctly---'nr' is supposed to point at one past the last element
of the array in use.  This has been corrected.

* md/list-objects-filter-memfix:
  list-objects-filter: correct usage of ALLOC_GROW

19 months agoMerge branch 'jt/partial-clone-missing-ref-delta-base' into maint
Junio C Hamano [Thu, 25 Jul 2019 21:27:12 +0000 (14:27 -0700)]
Merge branch 'jt/partial-clone-missing-ref-delta-base' into maint

"git fetch" into a lazy clone forgot to fetch base objects that are
necessary to complete delta in a thin packfile, which has been
corrected.

* jt/partial-clone-missing-ref-delta-base:
  t5616: cover case of client having delta base
  t5616: use correct flag to check object is missing
  index-pack: prefetch missing REF_DELTA bases
  t5616: refactor packfile replacement

19 months agoMerge branch 'xl/record-partial-clone-origin' into maint
Junio C Hamano [Thu, 25 Jul 2019 21:27:12 +0000 (14:27 -0700)]
Merge branch 'xl/record-partial-clone-origin' into maint

When creating a partial clone, the object filtering criteria is
recorded for the origin of the clone, but this incorrectly used a
hardcoded name "origin" to name that remote; it has been corrected
to honor the "--origin <name>" option.

* xl/record-partial-clone-origin:
  clone: respect user supplied origin name when setting up partial clone

19 months agoMerge branch 'pb/request-pull-verify-remote-ref' into maint
Junio C Hamano [Thu, 25 Jul 2019 21:27:11 +0000 (14:27 -0700)]
Merge branch 'pb/request-pull-verify-remote-ref' into maint

"git request-pull" learned to warn when the ref we ask them to pull
from in the local repository and in the published repository are
different.

* pb/request-pull-verify-remote-ref:
  request-pull: warn if the remote object is not the same as the local one
  request-pull: quote regex metacharacters in local ref

19 months agoMerge branch 'mm/p4-unshelve-windows-fix' into maint
Junio C Hamano [Thu, 25 Jul 2019 21:27:11 +0000 (14:27 -0700)]
Merge branch 'mm/p4-unshelve-windows-fix' into maint

The command line to invoke a "git cat-file" command from inside
"git p4" was not properly quoted to protect a caret and running a
broken command on Windows, which has been corrected.

* mm/p4-unshelve-windows-fix:
  p4 unshelve: fix "Not a valid object name HEAD0" on Windows

19 months agoMerge branch 'bb/unicode-12.1-reiwa' into maint
Junio C Hamano [Thu, 25 Jul 2019 21:27:10 +0000 (14:27 -0700)]
Merge branch 'bb/unicode-12.1-reiwa' into maint

Update to Unicode 12.1 width table.

* bb/unicode-12.1-reiwa:
  unicode: update the width tables to Unicode 12.1

19 months agoMerge branch 'js/fsmonitor-unflake' into maint
Junio C Hamano [Thu, 25 Jul 2019 21:27:10 +0000 (14:27 -0700)]
Merge branch 'js/fsmonitor-unflake' into maint

The data collected by fsmonitor was not properly written back to
the on-disk index file, breaking t7519 tests occasionally, which
has been corrected.

* js/fsmonitor-unflake:
  mark_fsmonitor_valid(): mark the index as changed if needed
  fill_stat_cache_info(): prepare for an fsmonitor fix

19 months agoMerge branch 'vv/merge-squash-with-explicit-commit' into maint
Junio C Hamano [Thu, 25 Jul 2019 21:27:10 +0000 (14:27 -0700)]
Merge branch 'vv/merge-squash-with-explicit-commit' into maint

"git merge --squash" is designed to update the working tree and the
index without creating the commit, and this cannot be countermanded
by adding the "--commit" option; the command now refuses to work
when both options are given.

* vv/merge-squash-with-explicit-commit:
  merge: refuse --commit with --squash

19 months agoMerge branch 'js/bundle-verify-require-object-store' into maint
Junio C Hamano [Thu, 25 Jul 2019 21:27:09 +0000 (14:27 -0700)]
Merge branch 'js/bundle-verify-require-object-store' into maint

"git bundle verify" needs to see if prerequisite objects exist in
the receiving repository, but the command did not check if we are
in a repository upfront, which has been corrected.

* js/bundle-verify-require-object-store:
  bundle verify: error out if called without an object database

19 months agoMerge branch 'jk/am-i-resolved-fix' into maint
Junio C Hamano [Thu, 25 Jul 2019 21:27:09 +0000 (14:27 -0700)]
Merge branch 'jk/am-i-resolved-fix' into maint

"git am -i --resolved" segfaulted after trying to see a commit as
if it were a tree, which has been corrected.

* jk/am-i-resolved-fix:
  am: fix --interactive HEAD tree resolution
  am: drop tty requirement for --interactive
  am: read interactive input from stdin
  am: simplify prompt response handling

19 months agoMerge branch 'jk/HEAD-symref-in-xfer-namespaces' into maint
Junio C Hamano [Thu, 25 Jul 2019 21:27:09 +0000 (14:27 -0700)]
Merge branch 'jk/HEAD-symref-in-xfer-namespaces' into maint

The server side support for "git fetch" used to show incorrect
value for the HEAD symbolic ref when the namespace feature is in
use, which has been corrected.

* jk/HEAD-symref-in-xfer-namespaces:
  upload-pack: strip namespace from symref data

19 months agoMerge branch 'ew/server-info-remove-crufts' into maint
Junio C Hamano [Thu, 25 Jul 2019 21:27:08 +0000 (14:27 -0700)]
Merge branch 'ew/server-info-remove-crufts' into maint

"git update-server-info" used to leave stale packfiles in its
output, which has been corrected.

* ew/server-info-remove-crufts:
  server-info: do not list unlinked packs

19 months agoMerge branch 'es/grep-require-name-when-needed' into maint
Junio C Hamano [Thu, 25 Jul 2019 21:27:08 +0000 (14:27 -0700)]
Merge branch 'es/grep-require-name-when-needed' into maint

More parameter validation.

* es/grep-require-name-when-needed:
  grep: fail if call could output and name is null

19 months agoMerge branch 'ds/object-info-for-prefetch-fix' into maint
Junio C Hamano [Thu, 25 Jul 2019 21:27:08 +0000 (14:27 -0700)]
Merge branch 'ds/object-info-for-prefetch-fix' into maint

Code cleanup and futureproof.

* ds/object-info-for-prefetch-fix:
  sha1-file: split OBJECT_INFO_FOR_PREFETCH

19 months agoMerge branch 'mh/import-transport-fd-fix' into maint
Junio C Hamano [Thu, 25 Jul 2019 21:27:07 +0000 (14:27 -0700)]
Merge branch 'mh/import-transport-fd-fix' into maint

The ownership rule for the file descriptor to fast-import remote
backend was mixed up, leading to unrelated file descriptor getting
closed, which has been fixed.

* mh/import-transport-fd-fix:
  Use xmmap_gently instead of xmmap in use_pack
  dup() the input fd for fast-import used for remote helpers

19 months agoMerge branch 'nd/corrupt-worktrees' into maint
Junio C Hamano [Thu, 25 Jul 2019 21:27:07 +0000 (14:27 -0700)]
Merge branch 'nd/corrupt-worktrees' into maint

"git worktree add" used to fail when another worktree connected to
the same repository was corrupt, which has been corrected.

* nd/corrupt-worktrees:
  worktree add: be tolerant of corrupt worktrees

19 months agoMerge branch 'nd/init-relative-template-fix' into maint
Junio C Hamano [Thu, 25 Jul 2019 21:27:06 +0000 (14:27 -0700)]
Merge branch 'nd/init-relative-template-fix' into maint

A relative pathname given to "git init --template=<path> <repo>"
ought to be relative to the directory "git init" gets invoked in,
but it instead was made relative to the repository, which has been
corrected.

* nd/init-relative-template-fix:
  init: make --template path relative to $CWD

19 months agoThe seventh batch
Junio C Hamano [Thu, 25 Jul 2019 20:56:20 +0000 (13:56 -0700)]
The seventh batch

Signed-off-by: Junio C Hamano <gitster@pobox.com>
19 months agoMerge branch 'jk/test-commit-bulk'
Junio C Hamano [Thu, 25 Jul 2019 20:59:24 +0000 (13:59 -0700)]
Merge branch 'jk/test-commit-bulk'

A test helper has been introduced to optimize preparation of test
repositories with many simple commits, and a handful of test
scripts have been updated to use it.

* jk/test-commit-bulk:
  t6200: use test_commit_bulk
  t5703: use test_commit_bulk
  t5702: use test_commit_bulk
  t3311: use test_commit_bulk
  t5310: increase the number of bitmapped commits
  test-lib: introduce test_commit_bulk

19 months agoMerge branch 'di/readme-markup-fix'
Junio C Hamano [Thu, 25 Jul 2019 20:59:24 +0000 (13:59 -0700)]
Merge branch 'di/readme-markup-fix'

Docfix.

* di/readme-markup-fix:
  README: fix rendering of text in angle brackets

19 months agoMerge branch 'jc/denoise-rm-to-resolve'
Junio C Hamano [Thu, 25 Jul 2019 20:59:24 +0000 (13:59 -0700)]
Merge branch 'jc/denoise-rm-to-resolve'

"git rm" to resolve a conflicted path leaked an internal message
"needs merge" before actually removing the path, which was
confusing.  This has been corrected.

* jc/denoise-rm-to-resolve:
  rm: resolving by removal is not a warning-worthy event

19 months agoMerge branch 'js/clean-report-too-long-a-path'
Junio C Hamano [Thu, 25 Jul 2019 20:59:24 +0000 (13:59 -0700)]
Merge branch 'js/clean-report-too-long-a-path'

"git clean" silently skipped a path when it cannot lstat() it; now
it gives a warning.

* js/clean-report-too-long-a-path:
  clean: show an error message when the path is too long

19 months agoMerge branch 'tg/stash-keep-index-with-removed-paths'
Junio C Hamano [Thu, 25 Jul 2019 20:59:23 +0000 (13:59 -0700)]
Merge branch 'tg/stash-keep-index-with-removed-paths'

"git stash --keep-index" did not work correctly on paths that have
been removed, which has been fixed.

* tg/stash-keep-index-with-removed-paths:
  stash: fix handling removed files with --keep-index

19 months agoMerge branch 'sr/gpg-interface-stop-at-the-end'
Junio C Hamano [Thu, 25 Jul 2019 20:59:23 +0000 (13:59 -0700)]
Merge branch 'sr/gpg-interface-stop-at-the-end'

A codepath that reads from GPG for signed object verification read
past the end of allocated buffer, which has been fixed.

* sr/gpg-interface-stop-at-the-end:
  gpg-interface: do not scan past the end of buffer

19 months agoMerge branch 'js/mingw-spawn-with-spaces-in-path'
Junio C Hamano [Thu, 25 Jul 2019 20:59:23 +0000 (13:59 -0700)]
Merge branch 'js/mingw-spawn-with-spaces-in-path'

Window 7 update ;-)

* js/mingw-spawn-with-spaces-in-path:
  mingw: support spawning programs containing spaces in their names

19 months agoMerge branch 'jc/post-c89-rules-doc'
Junio C Hamano [Thu, 25 Jul 2019 20:59:22 +0000 (13:59 -0700)]
Merge branch 'jc/post-c89-rules-doc'

We have been trying out a few language features outside c89; the
coding guidelines document did not talk about them and instead had
a blanket ban against them.

* jc/post-c89-rules-doc:
  CodingGuidelines: spell out post-C89 rules

19 months agoMerge branch 'mt/dir-iterator-updates'
Junio C Hamano [Thu, 25 Jul 2019 20:59:22 +0000 (13:59 -0700)]
Merge branch 'mt/dir-iterator-updates'

Adjust the dir-iterator API and apply it to the local clone
optimization codepath.

* mt/dir-iterator-updates:
  clone: replace strcmp by fspathcmp
  clone: use dir-iterator to avoid explicit dir traversal
  clone: extract function from copy_or_link_directory
  clone: copy hidden paths at local clone
  dir-iterator: add flags parameter to dir_iterator_begin
  dir-iterator: refactor state machine model
  dir-iterator: use warning_errno when possible
  dir-iterator: add tests for dir-iterator API
  clone: better handle symlinked files at .git/objects/
  clone: test for our behavior on odd objects/* content

19 months agoMerge branch 'vn/xmmap-gently'
Junio C Hamano [Thu, 25 Jul 2019 20:59:22 +0000 (13:59 -0700)]
Merge branch 'vn/xmmap-gently'

Clean-up an error codepath.

* vn/xmmap-gently:
  read-cache.c: do not die if mmap fails

19 months agoMerge branch 'ac/log-use-mailmap-by-default-transition'
Junio C Hamano [Thu, 25 Jul 2019 20:59:22 +0000 (13:59 -0700)]
Merge branch 'ac/log-use-mailmap-by-default-transition'

The "git log" command learns to issue a warning when log.mailmap
configuration is not set and --[no-]mailmap option is not used, to
prepare users for future versions of Git that uses the mailmap by
default.

* ac/log-use-mailmap-by-default-transition:
  tests: defang pager tests by explicitly disabling the log.mailmap warning
  documentation: mention --no-use-mailmap and log.mailmap false setting
  log: add warning for unspecified log.mailmap setting

19 months agoMerge branch 'es/local-atomic-push-failure-with-http'
Junio C Hamano [Thu, 25 Jul 2019 20:59:22 +0000 (13:59 -0700)]
Merge branch 'es/local-atomic-push-failure-with-http'

"git push --atomic" that goes over the transport-helper (namely,
the smart http transport) failed to prevent refs to be pushed when
it can locally tell that one of the ref update will fail without
having to consult the other end, which has been corrected.

* es/local-atomic-push-failure-with-http:
  transport-helper: avoid var decl in for () loop control
  transport-helper: enforce atomic in push_refs_with_push

19 months agoMerge branch 'rm/gpg-program-doc-fix'
Junio C Hamano [Thu, 25 Jul 2019 20:59:21 +0000 (13:59 -0700)]
Merge branch 'rm/gpg-program-doc-fix'

Docfix.

* rm/gpg-program-doc-fix:
  gpg(docs): use correct --verify syntax

19 months agoMerge branch 'js/unmap-before-ext-diff'
Junio C Hamano [Thu, 25 Jul 2019 20:59:21 +0000 (13:59 -0700)]
Merge branch 'js/unmap-before-ext-diff'

Windows update.

* js/unmap-before-ext-diff:
  diff: munmap() file contents before running external diff

19 months agoMerge branch 'tg/range-diff-output-update'
Junio C Hamano [Thu, 25 Jul 2019 20:59:20 +0000 (13:59 -0700)]
Merge branch 'tg/range-diff-output-update'

"git range-diff" output has been tweaked for easier identification
of which part of what file the patch shown is about.

* tg/range-diff-output-update:
  range-diff: add headers to the outer hunk header
  range-diff: add filename to inner diff
  range-diff: add section header instead of diff header
  range-diff: suppress line count in outer diff
  range-diff: don't remove funcname from inner diff
  range-diff: split lines manually
  range-diff: fix function parameter indentation
  apply: make parse_git_diff_header public
  apply: only pass required data to gitdiff_* functions
  apply: only pass required data to find_name_*
  apply: only pass required data to check_header_line
  apply: only pass required data to git_header_name
  apply: only pass required data to skip_tree_prefix
  apply: replace marc.info link with public-inbox

19 months agoMerge branch 'ab/test-env'
Junio C Hamano [Thu, 25 Jul 2019 20:59:20 +0000 (13:59 -0700)]
Merge branch 'ab/test-env'

Many GIT_TEST_* environment variables control various aspects of
how our tests are run, but a few followed "non-empty is true, empty
or unset is false" while others followed the usual "there are a few
ways to spell true, like yes, on, etc., and also ways to spell
false, like no, off, etc." convention.

* ab/test-env:
  env--helper: mark a file-local symbol as static
  tests: make GIT_TEST_FAIL_PREREQS a boolean
  tests: replace test_tristate with "git env--helper"
  tests README: re-flow a previously changed paragraph
  tests: make GIT_TEST_GETTEXT_POISON a boolean
  t6040 test: stop using global "script" variable
  config.c: refactor die_bad_number() to not call gettext() early
  env--helper: new undocumented builtin wrapping git_env_*()
  config tests: simplify include cycle test

19 months agogrep: print the pcre2_jit_on value
Beat Bolli [Mon, 22 Jul 2019 18:19:23 +0000 (20:19 +0200)]
grep: print the pcre2_jit_on value

When pcre2_jit_on is neither 1 nor 0, the BUG() call printed the value
of pcre1_jit_on.

Print the value of pcre2_jit_on instead.

Signed-off-by: Beat Bolli <dev+git@drbeat.li>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
19 months agot6200: use test_commit_bulk
Jeff King [Fri, 19 Jul 2019 21:56:23 +0000 (17:56 -0400)]
t6200: use test_commit_bulk

There's a loop that creates 30 commits using test_commit. Using
test_commit_bulk speeds this up from:

  Benchmark #1: ./t6200-fmt-merge-msg.sh --root=/var/ram/git-tests
    Time (mean ± σ):      1.926 s ±  0.240 s    [User: 1.055 s, System: 0.963 s]
    Range (min … max):    1.431 s …  2.166 s    10 runs

to:

  Benchmark #1: ./t6200-fmt-merge-msg.sh --root=/var/ram/git-tests
    Time (mean ± σ):      1.343 s ±  0.179 s    [User: 766.5 ms, System: 662.9 ms]
    Range (min … max):    1.032 s …  1.664 s    10 runs

for an average savings of over 30%.

Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
19 months agotravis-ci: build with GCC 4.8 as well
SZEDER Gábor [Thu, 18 Jul 2019 15:22:34 +0000 (17:22 +0200)]
travis-ci: build with GCC 4.8 as well

C99 'for' loop initial declaration, i.e. 'for (int i = 0; i < n; i++)',
is not allowed in Git's codebase yet, to maintain compatibility with
some older compilers.

Our Travis CI builds used to catch 'for' loop initial declarations,
because the GETTEXT_POISON job has always built Git with the default
'cc', which in Travis CI's previous default Linux image (based on
Ubuntu 14.04 Trusty) is GCC 4.8, and that GCC version errors out on
this construct (not only with DEVELOPER=1, but with our default CFLAGS
as well).  Alas, that's not the case anymore, becase after 14.04's EOL
Travis CI's current default Linux image is based on Ubuntu 16.04
Xenial [1] and its default 'cc' is now GCC 5.4, which, just like all
later GCC and Clang versions, simply accepts this construct, even if
we don't explicitly specify '-std=c99'.

Ideally we would adjust our CFLAGS used with DEVELOPER=1 to catch this
undesired construct already when contributors build Git on their own
machines.  Unfortunately, however, there seems to be no compiler
option that would catch only this particular construct without choking
on many other things, e.g. while a later compiler with '-std=c90'
and/or '-ansi' does catch this construct, it can't build Git because
of several screenfulls of other errors.

Add the 'linux-gcc-4.8' job to Travis CI, in order to build Git with
GCC 4.8, and thus to timely catch any 'for' loop initial declarations.
To catch those it's sufficient to only build Git with GCC 4.8, so
don't run the test suite in this job, because 'make test' takes rather
long [2], and it's already run five times in other jobs, so we
wouldn't get our time's worth.

[1] The Azure Pipelines builds have been using Ubuntu 16.04 images
    from the start, so I belive they never caught 'for' loop initial
    declarations.

[2] On Travis CI 'make test' alone would take about 9 minutes in this
    new job (without running httpd, Subversion, and P4 tests).  For
    comparison, starting the job and building Git with GCC 4.8 takes
    only about 2 minutes.

Signed-off-by: SZEDER Gábor <szeder.dev@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
19 months agoThe sixth batch
Junio C Hamano [Fri, 19 Jul 2019 18:34:23 +0000 (11:34 -0700)]
The sixth batch

Signed-off-by: Junio C Hamano <gitster@pobox.com>
19 months agoMerge branch 'ma/ref-filter-leakfix'
Junio C Hamano [Fri, 19 Jul 2019 18:30:23 +0000 (11:30 -0700)]
Merge branch 'ma/ref-filter-leakfix'

Leakfix.

* ma/ref-filter-leakfix:
  ref-filter: fix memory leak in `free_array_item()`

19 months agoMerge branch 'js/trace2-signo-typofix'
Junio C Hamano [Fri, 19 Jul 2019 18:30:23 +0000 (11:30 -0700)]
Merge branch 'js/trace2-signo-typofix'

Documentation fix.

* js/trace2-signo-typofix:
  trace2: correct trace2 field name documentation

19 months agoMerge branch 'kb/mingw-set-home'
Junio C Hamano [Fri, 19 Jul 2019 18:30:23 +0000 (11:30 -0700)]
Merge branch 'kb/mingw-set-home'

Windows port update.

* kb/mingw-set-home:
  mingw: initialize HOME on startup

19 months agoMerge branch 'ea/merge-code-cleanup'
Junio C Hamano [Fri, 19 Jul 2019 18:30:23 +0000 (11:30 -0700)]
Merge branch 'ea/merge-code-cleanup'

A loop has been rewritten for conciseness and clarity.

* ea/merge-code-cleanup:
  builtin/merge.c - cleanup of code in for-cycle that tests strategies

19 months agoMerge branch 'jl/status-reduce-vertical-blank'
Junio C Hamano [Fri, 19 Jul 2019 18:30:23 +0000 (11:30 -0700)]
Merge branch 'jl/status-reduce-vertical-blank'

Extra blank lines in "git status" output have been reduced.

* jl/status-reduce-vertical-blank:
  status: remove the empty line after hints

19 months agoMerge branch 'pw/rebase-progress-test-cleanup'
Junio C Hamano [Fri, 19 Jul 2019 18:30:22 +0000 (11:30 -0700)]
Merge branch 'pw/rebase-progress-test-cleanup'

Test cleanup.

* pw/rebase-progress-test-cleanup:
  t3420: remove progress lines before comparing output

19 months agoMerge branch 'pw/prompt-cherry-pick-revert-fix'
Junio C Hamano [Fri, 19 Jul 2019 18:30:22 +0000 (11:30 -0700)]
Merge branch 'pw/prompt-cherry-pick-revert-fix'

When one step in multi step cherry-pick or revert is reset or
committed, the command line prompt script failed to notice the
current status, which has been improved.

* pw/prompt-cherry-pick-revert-fix:
  git-prompt: improve cherry-pick/revert detection

19 months agoMerge branch 'ew/repack-with-bitmaps-by-default'
Junio C Hamano [Fri, 19 Jul 2019 18:30:22 +0000 (11:30 -0700)]
Merge branch 'ew/repack-with-bitmaps-by-default'

Generation of pack bitmaps are now disabled when .keep files exist,
as these are mutually exclusive features.

* ew/repack-with-bitmaps-by-default:
  repack: disable bitmaps-by-default if .keep files exist

19 months agoMerge branch 'jk/check-connected-with-alternates'
Junio C Hamano [Fri, 19 Jul 2019 18:30:21 +0000 (11:30 -0700)]
Merge branch 'jk/check-connected-with-alternates'

The tips of refs from the alternate object store can be used as
starting point for reachability computation now.

* jk/check-connected-with-alternates:
  check_everything_connected: assume alternate ref tips are valid
  object-store.h: move for_each_alternate_ref() from transport.h

19 months agoMerge branch 'nd/tree-walk-with-repo'
Junio C Hamano [Fri, 19 Jul 2019 18:30:21 +0000 (11:30 -0700)]
Merge branch 'nd/tree-walk-with-repo'

The tree-walk API learned to pass an in-core repository
instance throughout more codepaths.

* nd/tree-walk-with-repo:
  t7814: do not generate same commits in different repos
  Use the right 'struct repository' instead of the_repository
  match-trees.c: remove the_repo from shift_tree*()
  tree-walk.c: remove the_repo from get_tree_entry_follow_symlinks()
  tree-walk.c: remove the_repo from get_tree_entry()
  tree-walk.c: remove the_repo from fill_tree_descriptor()
  sha1-file.c: remove the_repo from read_object_with_reference()

19 months agoMerge branch 'ra/cherry-pick-revert-skip'
Junio C Hamano [Fri, 19 Jul 2019 18:30:21 +0000 (11:30 -0700)]
Merge branch 'ra/cherry-pick-revert-skip'

"git cherry-pick/revert" learned a new "--skip" action.

* ra/cherry-pick-revert-skip:
  cherry-pick/revert: advise using --skip
  cherry-pick/revert: add --skip option
  sequencer: use argv_array in reset_merge
  sequencer: rename reset_for_rollback to reset_merge
  sequencer: add advice for revert

19 months agoMerge branch 'tb/ref-filter-multiple-patterns'
Junio C Hamano [Fri, 19 Jul 2019 18:30:21 +0000 (11:30 -0700)]
Merge branch 'tb/ref-filter-multiple-patterns'

"git for-each-ref" with multiple patterns have been optimized.

* tb/ref-filter-multiple-patterns:
  ref-filter.c: find disjoint pattern prefixes

19 months agoMerge branch 'pw/status-with-corrupt-sequencer-state'
Junio C Hamano [Fri, 19 Jul 2019 18:30:20 +0000 (11:30 -0700)]
Merge branch 'pw/status-with-corrupt-sequencer-state'

The code to read state files used by the sequencer machinery for
"git status" has been made more robust against a corrupt or stale
state files.

* pw/status-with-corrupt-sequencer-state:
  status: do not report errors in sequencer/todo
  sequencer: factor out todo command name parsing
  sequencer: always allow tab after command name

19 months agoMerge branch 'ds/commit-graph-incremental'
Junio C Hamano [Fri, 19 Jul 2019 18:30:20 +0000 (11:30 -0700)]
Merge branch 'ds/commit-graph-incremental'

The commits in a repository can be described by multiple
commit-graph files now, which allows the commit-graph files to be
updated incrementally.

* ds/commit-graph-incremental:
  commit-graph: test verify across alternates
  commit-graph: normalize commit-graph filenames
  commit-graph: test --split across alternate without --split
  commit-graph: test octopus merges with --split
  commit-graph: clean up chains after flattened write
  commit-graph: verify chains with --shallow mode
  commit-graph: create options for split files
  commit-graph: expire commit-graph files
  commit-graph: allow cross-alternate chains
  commit-graph: merge commit-graph chains
  commit-graph: add --split option to builtin
  commit-graph: write commit-graph chains
  commit-graph: rearrange chunk count logic
  commit-graph: add base graphs chunk
  commit-graph: load commit-graph chains
  commit-graph: rename commit_compare to oid_compare
  commit-graph: prepare for commit-graph chains
  commit-graph: document commit-graph chains

19 months agoMerge branch 'br/blame-ignore'
Junio C Hamano [Fri, 19 Jul 2019 18:30:20 +0000 (11:30 -0700)]
Merge branch 'br/blame-ignore'

"git blame" learned to "ignore" commits in the history, whose
effects (as well as their presence) get ignored.

* br/blame-ignore:
  t8014: remove unnecessary braces
  blame: drop some unused function parameters
  blame: add a test to cover blame_coalesce()
  blame: use the fingerprint heuristic to match ignored lines
  blame: add a fingerprint heuristic to match ignored lines
  blame: optionally track line fingerprints during fill_blame_origin()
  blame: add config options for the output of ignored or unblamable lines
  blame: add the ability to ignore commits and their changes
  blame: use a helper function in blame_chunk()
  Move oidset_parse_file() to oidset.c
  fsck: rename and touch up init_skiplist()

19 months agoMerge branch 'cc/test-oidmap'
Junio C Hamano [Fri, 19 Jul 2019 18:30:19 +0000 (11:30 -0700)]
Merge branch 'cc/test-oidmap'

Extend the test coverage a bit.

* cc/test-oidmap:
  t0016: add 'remove' subcommand test
  test-oidmap: remove 'add' subcommand
  test-hashmap: remove 'hash' command
  oidmap: use sha1hash() instead of static hash() function
  t: add t0016-oidmap.sh
  t/helper: add test-oidmap.c

19 months agoMerge branch 'ds/midx-expire-repack'
Junio C Hamano [Fri, 19 Jul 2019 18:30:19 +0000 (11:30 -0700)]
Merge branch 'ds/midx-expire-repack'

"git multi-pack-index" learned expire and repack subcommands.

* ds/midx-expire-repack:
  t5319: use 'test-tool path-utils' instead of 'ls -l'
  t5319-multi-pack-index.sh: test batch size zero
  midx: add test that 'expire' respects .keep files
  multi-pack-index: test expire while adding packs
  midx: implement midx_repack()
  multi-pack-index: prepare 'repack' subcommand
  multi-pack-index: implement 'expire' subcommand
  midx: refactor permutation logic and pack sorting
  midx: simplify computation of pack name lengths
  multi-pack-index: prepare for 'expire' subcommand
  Docs: rearrange subcommands for multi-pack-index
  repack: refactor pack deletion for future use

19 months agoclean: show an error message when the path is too long
Johannes Schindelin [Thu, 18 Jul 2019 09:30:33 +0000 (02:30 -0700)]
clean: show an error message when the path is too long

When `lstat()` failed, `git clean` would abort without an error
message, leaving the user quite puzzled.

In particular on Windows, where the default maximum path length is
quite small (yet there are ways to circumvent that limit in many
cases), it is very important that users be given an indication why
their command failed because of too long paths when it did.

This test case makes sure that a warning is issued that would have
helped the user who reported this issue:

https://github.com/git-for-windows/git/issues/521

Note that we temporarily set `core.longpaths = false` in the regression
test; this ensures forward-compatibility with the `core.longpaths`
feature that has not yet been upstreamed from Git for Windows.

Helped-by: René Scharfe <l.s.r@web.de>
Helped-by: SZEDER Gábor <szeder.dev@gmail.com>
Helped-by: Junio C Hamano <gitster@pobox.com>
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
19 months agoCodingGuidelines: spell out post-C89 rules
Junio C Hamano [Tue, 16 Jul 2019 17:21:20 +0000 (10:21 -0700)]
CodingGuidelines: spell out post-C89 rules

Even though we have been sticking to C89, there are a few handy
features we borrow from more recent C language in our codebase after
trying them in weather balloons and saw that nobody screamed.

Spell them out.

While at it, extend the existing variable declaration rule a bit to
read better with the newly spelled out rule for the for loop.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
19 months agoREADME: fix rendering of text in angle brackets
Doug Ilijev [Thu, 18 Jul 2019 19:08:45 +0000 (12:08 -0700)]
README: fix rendering of text in angle brackets

Markdown incorrectly interpreted `<commandname>` as an HTML tag;
use backticks to escape `Documentation/git-<commandname>.txt` to ensure
that it renders the text as intended.

Signed-off-by: Doug Ilijev <doug.ilijev@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
19 months agorm: resolving by removal is not a warning-worthy event
Junio C Hamano [Wed, 17 Jul 2019 20:28:24 +0000 (13:28 -0700)]
rm: resolving by removal is not a warning-worthy event

When resolving a conflict on a path in favor of removing it, using
"git rm" on it is the standard way to do so.  The user however is
greeted with a "needs merge" message during that operation:

$ git merge side-branch
$ edit conflicted-path-1
$ git add conflicted-path-1
$ git rm conflicted-path-2
conflicted-path-2: needs merge
rm 'conflicted-path-2'

The removal by "git rm" does get performed, but an uninitiated user
may find it confusing, "needs merge? so I need to resolve conflict
before being able to remove it???"

The message is coming from "update-index --refresh" that is called
internally to make sure "git rm" knows which paths are clean and
which paths are dirty, in order to prevent removal of paths modified
relative to the index without the "-f" option.  We somehow ended up
not squelching this message which seeped through to the UI surface.

Use the same mechanism used by "git commit", "git describe", etc. to
squelch the message.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
19 months agotransport-helper: avoid var decl in for () loop control
Junio C Hamano [Tue, 16 Jul 2019 20:28:21 +0000 (13:28 -0700)]
transport-helper: avoid var decl in for () loop control

We do allow a few selected C99 constructs in our codebase these
days, but this is not among them (yet).

Reported-by: Carlo Arenas <carenas@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
19 months agostash: fix handling removed files with --keep-index
Thomas Gummerer [Tue, 16 Jul 2019 14:23:22 +0000 (15:23 +0100)]
stash: fix handling removed files with --keep-index

git stash push --keep-index is supposed to keep all changes that have
been added to the index, both in the index and on disk.

Currently this doesn't behave correctly when a file is removed from
the index.  Instead of keeping it deleted on disk, --keep-index
currently restores the file.

Fix that behaviour by using 'git checkout' in no-overlay mode which
can faithfully restore the index and working tree.  This also
simplifies the code.

Note that this will overwrite untracked files if the untracked file
has the same name as a file that has been deleted in the index.

Signed-off-by: Thomas Gummerer <t.gummerer@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
19 months agomingw: support spawning programs containing spaces in their names
Johannes Schindelin [Tue, 16 Jul 2019 14:03:12 +0000 (07:03 -0700)]
mingw: support spawning programs containing spaces in their names

On some older Windows versions (e.g. Windows 7), the CreateProcessW()
function does not really support spaces in its first argument,
lpApplicationName. But it supports passing NULL as lpApplicationName,
which makes it figure out the application from the (possibly quoted)
first argument of lpCommandLine.

Let's use that trick (if we are certain that the first argument matches
the executable's path) to support launching programs whose path contains
spaces.

We will abuse the test-fake-ssh.exe helper to verify that this works and
does not regress.

This fixes https://github.com/git-for-windows/git/issues/692

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
19 months agogpg-interface: do not scan past the end of buffer
Steven Roberts [Tue, 16 Jul 2019 18:47:37 +0000 (11:47 -0700)]
gpg-interface: do not scan past the end of buffer

If the GPG output ends with trailing blank lines, after skipping
them over inside the loop to find the terminating NUL at the end,
the loop ends up looking for the next line, starting past the end.

Signed-off-by: Steven Roberts <sroberts@fenderq.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
19 months agotests: defang pager tests by explicitly disabling the log.mailmap warning
Ariadne Conill [Mon, 15 Jul 2019 12:41:06 +0000 (07:41 -0500)]
tests: defang pager tests by explicitly disabling the log.mailmap warning

In the previous patch, we added a deprecation warning for the current
log.mailmap setting. This warning only appears when git is attached to
a controlling terminal. Some tests however run under an emulated
terminal, so we need to disable the warning for those tests.

Thanks to Junio for suggesting that we do this in the setup function.

Signed-off-by: Ariadne Conill <ariadne@dereferenced.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
19 months agodocumentation: mention --no-use-mailmap and log.mailmap false setting
Ariadne Conill [Mon, 15 Jul 2019 12:41:05 +0000 (07:41 -0500)]
documentation: mention --no-use-mailmap and log.mailmap false setting

The log.mailmap setting may be explicitly set to false, which disables
the mailmap feature implicity. In practice, doing so is equivalent to
always using the previously undocumented --no-use-mailmap option on the
command line.

Accordingly, we document both the existence of --no-use-mailmap as
well as briefly discuss the equivalence of it to log.mailmap=False.

Signed-off-by: Ariadne Conill <ariadne@dereferenced.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
19 months agolog: add warning for unspecified log.mailmap setting
Ariadne Conill [Mon, 15 Jul 2019 12:41:04 +0000 (07:41 -0500)]
log: add warning for unspecified log.mailmap setting

Based on discussions around changing the log.mailmap default to being
enabled, it was decided that a transitional period is required.

Accordingly, we announce this transitional period with a warning
message.

Signed-off-by: Ariadne Conill <ariadne@dereferenced.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
19 months agoread-cache.c: do not die if mmap fails
Varun Naik [Sun, 14 Jul 2019 03:01:53 +0000 (20:01 -0700)]
read-cache.c: do not die if mmap fails

do_read_index() mmaps the index, or tries to die with an error message
on failure. It should call xmmap_gently(), which returns MAP_FAILED,
rather than xmmap(), which dies with its own error message.

An easy way to cause this mmap to fail is by setting $GIT_INDEX_FILE to
a path to a directory and then invoking any command that reads from the
index.

Signed-off-by: Varun Naik <vcnaik94@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
19 months agogpg(docs): use correct --verify syntax
Robert Morgan [Fri, 12 Jul 2019 15:33:57 +0000 (08:33 -0700)]
gpg(docs): use correct --verify syntax

The gpg --verify usage example within the 'gpg.program' variable
reference provides an incorrect example of the gpg --verify command
arguments.

The command argument order, when providing both a detached signature
and data, should be signature first and data second:
https://gnupg.org/documentation/manuals/gnupg/Operational-GPG-Commands.html

Signed-off-by: Robert Morgan <robert.thomas.morgan@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
19 months agotransport-helper: enforce atomic in push_refs_with_push
Emily Shaffer [Thu, 11 Jul 2019 21:19:19 +0000 (14:19 -0700)]
transport-helper: enforce atomic in push_refs_with_push

Teach transport-helper how to notice if skipping a ref during push would
violate atomicity on the client side. We notice that a ref would be
rejected, and choose not to send it, but don't notice that if the client
has asked for --atomic we are violating atomicity if all the other
pushes we are sending would succeed. Asking the server end to uphold
atomicity wouldn't work here as the server doesn't have any idea that we
tried to update a ref that's broken.

The added test-case is a succinct way to reproduce this issue that fails
today. The same steps work fine when we aren't using a transport-helper
to get to the upstream, i.e. when we've added a local repository as a
remote:

  git remote add ~/upstream upstream

Signed-off-by: Emily Shaffer <emilyshaffer@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
19 months agoThe fifth batch
Junio C Hamano [Thu, 11 Jul 2019 22:17:22 +0000 (15:17 -0700)]
The fifth batch

Signed-off-by: Junio C Hamano <gitster@pobox.com>
19 months agoMerge branch 'js/mingw-use-utf8'
Junio C Hamano [Thu, 11 Jul 2019 22:16:49 +0000 (15:16 -0700)]
Merge branch 'js/mingw-use-utf8'

Windows update.

* js/mingw-use-utf8:
  mingw: fix possible buffer overrun when calling `GetUserNameW()`
  mingw: use Unicode functions explicitly
  mingw: get pw_name in UTF-8 format

19 months agoMerge branch 'sg/ci-brew-gcc-workaround'
Junio C Hamano [Thu, 11 Jul 2019 22:16:49 +0000 (15:16 -0700)]
Merge branch 'sg/ci-brew-gcc-workaround'

Dev support update.

* sg/ci-brew-gcc-workaround:
  ci/lib.sh: update a comment about installed P4 and Git-LFS versions
  ci: disable Homebrew's auto cleanup
  ci: don't update Homebrew

19 months agoMerge branch 'kb/windows-force-utf8'
Junio C Hamano [Thu, 11 Jul 2019 22:16:49 +0000 (15:16 -0700)]
Merge branch 'kb/windows-force-utf8'

Windows update.

* kb/windows-force-utf8:
  gettext: always use UTF-8 on native Windows

19 months agoMerge branch 'dr/progress-i18n'
Junio C Hamano [Thu, 11 Jul 2019 22:16:49 +0000 (15:16 -0700)]
Merge branch 'dr/progress-i18n'

Progress messages have been made localizable.

* dr/progress-i18n:
  l10n: localizable upload progress messages

19 months agoMerge branch 'qn/clone-doc-use-long-form'
Junio C Hamano [Thu, 11 Jul 2019 22:16:48 +0000 (15:16 -0700)]
Merge branch 'qn/clone-doc-use-long-form'

The "git clone" documentation refers to command line options in its
description in the short form; they have been replaced with long
forms to make them more recognisable.

* qn/clone-doc-use-long-form:
  docs: git-clone: list short form of options first
  docs: git-clone: refer to long form of options

19 months agoMerge branch 'js/rebase-reschedule-applies-only-to-interactive'
Junio C Hamano [Thu, 11 Jul 2019 22:16:48 +0000 (15:16 -0700)]
Merge branch 'js/rebase-reschedule-applies-only-to-interactive'

The configuration variable rebase.rescheduleFailedExec should be
effective only while running an interactive rebase and should not
affect anything when running an non-interactive one, which was not
the case.  This has been corrected.

* js/rebase-reschedule-applies-only-to-interactive:
  rebase --am: ignore rebase.rescheduleFailedExec

19 months agoMerge branch 'sg/git-C-empty-doc'
Junio C Hamano [Thu, 11 Jul 2019 22:16:48 +0000 (15:16 -0700)]
Merge branch 'sg/git-C-empty-doc'

Doc update.

* sg/git-C-empty-doc:
  Document that 'git -C ""' works and doesn't change directory

19 months agoMerge branch 'jt/t5551-test-chunked'
Junio C Hamano [Thu, 11 Jul 2019 22:16:47 +0000 (15:16 -0700)]
Merge branch 'jt/t5551-test-chunked'

Update smart-http test.

* jt/t5551-test-chunked:
  t5551: test usage of chunked encoding explicitly

19 months agoMerge branch 'js/mingw-gcc-stack-protect'
Junio C Hamano [Thu, 11 Jul 2019 22:16:47 +0000 (15:16 -0700)]
Merge branch 'js/mingw-gcc-stack-protect'

Windows update.

* js/mingw-gcc-stack-protect:
  mingw: enable stack smashing protector

19 months agoMerge branch 'cb/windows-manifest'
Junio C Hamano [Thu, 11 Jul 2019 22:16:47 +0000 (15:16 -0700)]
Merge branch 'cb/windows-manifest'

Windows update.

* cb/windows-manifest:
  mingw: embed a manifest to trick UAC into Doing The Right Thing

19 months agoenv--helper: mark a file-local symbol as static
Ramsay Jones [Thu, 11 Jul 2019 17:06:13 +0000 (18:06 +0100)]
env--helper: mark a file-local symbol as static

Signed-off-by: Ramsay Jones <ramsay@ramsayjones.plus.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
19 months agorange-diff: add headers to the outer hunk header
Thomas Gummerer [Thu, 11 Jul 2019 16:08:51 +0000 (17:08 +0100)]
range-diff: add headers to the outer hunk header

Add the section headers/hunk headers we introduced in the previous
commits to the outer diff's hunk headers.  This makes it easier to
understand which change we are actually looking at.  For example an
outer hunk header might now look like:

    @@  Documentation/config/interactive.txt

while previously it would have only been

    @@

which doesn't give a lot of context for the change that follows.

For completeness also add section headers for the commit metadata and
the commit message, although they are arguably less important.

Signed-off-by: Thomas Gummerer <t.gummerer@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
19 months agorange-diff: add filename to inner diff
Thomas Gummerer [Thu, 11 Jul 2019 16:08:50 +0000 (17:08 +0100)]
range-diff: add filename to inner diff

In a range-diff it's not always clear which file a certain funcname of
the inner diff belongs to, because the diff header (or section header
as added in a previous commit) is not always visible in the
range-diff.

Add the filename to the inner diffs header, so it's always visible to
users.

This also allows us to add the filename + the funcname to the outer
diffs hunk headers using a custom userdiff pattern, which will be done
in the next commit.

Signed-off-by: Thomas Gummerer <t.gummerer@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
19 months agorange-diff: add section header instead of diff header
Thomas Gummerer [Thu, 11 Jul 2019 16:08:49 +0000 (17:08 +0100)]
range-diff: add section header instead of diff header

Currently range-diff keeps the diff header of the inner diff
intact (apart from stripping lines starting with index).  This diff
header is somewhat useful, especially when files get different
names in different ranges.

However there is no real need to keep the whole diff header for that.
The main reason we currently do that is probably because it is easy to
do.

Introduce a new range diff hunk header, that's enclosed by "##",
similar to how line numbers in diff hunks are enclosed by "@@", and
give human readable information of what exactly happened to the file,
including the file name.

This improves the readability of the range-diff by giving more concise
information to the users.  For example if a file was renamed in one
iteration, but not in another, the diff of the headers would be quite
noisy.  However the diff of a single line is concise and should be
easier to understand.

Additionally, this allows us to add these range diff section headers to
the outer diffs hunk headers using a custom userdiff pattern, which
should help making the range-diff more readable.

Signed-off-by: Thomas Gummerer <t.gummerer@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
19 months agorange-diff: suppress line count in outer diff
Thomas Gummerer [Thu, 11 Jul 2019 16:08:48 +0000 (17:08 +0100)]
range-diff: suppress line count in outer diff

The line count in the outer diff's hunk headers of a range diff is not
all that interesting.  It merely shows how far along the inner diff
are on both sides.  That number is of no use for human readers, and
range-diffs are not meant to be machine readable.

In a subsequent commit we're going to add some more contextual
information such as the filename corresponding to the diff to the hunk
headers.  Remove the unnecessary information, and just keep the "@@"
to indicate that a new hunk of the outer diff is starting.

Signed-off-by: Thomas Gummerer <t.gummerer@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
19 months agorange-diff: don't remove funcname from inner diff
Thomas Gummerer [Thu, 11 Jul 2019 16:08:47 +0000 (17:08 +0100)]
range-diff: don't remove funcname from inner diff

When postprocessing the inner diff in range-diff, we currently replace
the whole hunk header line with just "@@".  This matches how 'git
tbdiff' used to handle hunk headers as well.

Most likely this is being done because line numbers in the hunk header
are not relevant without other changes.  They can for example easily
change if a range is rebased, and lines are added/removed before a
change that we actually care about in our ranges.

However it can still be useful to have the function name that 'git
diff' extracts as additional context for the change.

Note that it is not guaranteed that the hunk header actually shows up
in the range-diff, and this change only aims to improve the case where
a hunk header would already be included in the final output.

Signed-off-by: Thomas Gummerer <t.gummerer@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
19 months agorange-diff: split lines manually
Thomas Gummerer [Thu, 11 Jul 2019 16:08:46 +0000 (17:08 +0100)]
range-diff: split lines manually

Currently range-diff uses the 'strbuf_getline()' function for doing
its line by line processing.  In a future patch we want to do parts of
that parsing using the 'parse_git_diff_header()' function.  That
function does its own line by line reading of the input, and doesn't
use strbufs.  This doesn't match with how we do the line-by-line
processing in range-diff currently.

Switch range-diff to do our own line by line parsing, so we can re-use
the 'parse_git_diff_header()' function later.

Signed-off-by: Thomas Gummerer <t.gummerer@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
19 months agorange-diff: fix function parameter indentation
Thomas Gummerer [Thu, 11 Jul 2019 16:08:45 +0000 (17:08 +0100)]
range-diff: fix function parameter indentation

Fix the indentation of the function parameters for a couple of
functions, to match the style in the rest of the file.

Signed-off-by: Thomas Gummerer <t.gummerer@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>