Browse Source

What's cooking (2014/10 #05)

todo
Junio C Hamano 11 years ago
parent
commit
430fce517f
  1. 399
      whats-cooking.txt

399
whats-cooking.txt

@ -1,16 +1,26 @@ @@ -1,16 +1,26 @@
To: git@vger.kernel.org
Cc: Eric Wong <normalperson@yhbt.net>,
Paul Mackerras <paulus@samba.org>,
Pat Thoyts <patthoyts@users.sourceforge.net>
Bcc: lwn@lwn.net
Subject: What's cooking in git.git (Oct 2014, #04; Thu, 16)
X-master-at: 98349e5364da6164add4ede3e63ceba2cc39159b
X-next-at: 26a8a7df7daa247250ed7881245ac01138463bad
Subject: What's cooking in git.git (Oct 2014, #05; Mon, 20)
X-master-at: 13da0fc092b8cf082eda2f16971c75903aa5aefc
X-next-at: 72aedf1b35ff671435dd29c2b62e2b2250805689

What's cooking in git.git (Oct 2014, #04; Thu, 16)
What's cooking in git.git (Oct 2014, #05; Mon, 20)
--------------------------------------------------

Here are the topics that have been cooking. Commits prefixed with
'-' are only in 'pu' (proposed updates) while commits prefixed with
'+' are in 'next'.

According to tinyurl.com/gitCal, the feature freeze for this cycle
is scheduled for the end of this month. What's in the 'next' branch
as of today looks more or less complete picture, I would think. I'd
like to ask area maintainers of git-svn, gitk and git-gui to please
prepare to wrap up the accumulated changes for this cycle and throw
a pull request to me. Thanks.

You can find the changes described here in the integration branches
of the repositories listed at

@ -19,82 +29,113 @@ of the repositories listed at @@ -19,82 +29,113 @@ of the repositories listed at
--------------------------------------------------
[Graduated to "master"]

* bw/trace-no-inline-getnanotime (2014-09-29) 1 commit
(merged to 'next' on 2014-10-14 at 19facbb)
+ trace.c: do not mark getnanotime() as "inline"
* bc/asciidoc (2014-10-13) 2 commits
(merged to 'next' on 2014-10-14 at d9ff206)
+ Documentation: fix mismatched delimiters in git-imap-send
+ Documentation: adjust document title underlining
(this branch is used by bc/asciidoctor.)

No file-scope static variables in an inlined function, please.
Formatting nitpicks to help a (pickier) reimplementation of
AsciiDoc to grok our documentation.


* jc/completion-no-chdir (2014-10-09) 1 commit
(merged to 'next' on 2014-10-14 at 1cf12e1)
+ completion: use "git -C $there" instead of (cd $there && git ...)
* cc/interpret-trailers (2014-10-13) 11 commits
(merged to 'next' on 2014-10-17 at ee4519c)
+ Documentation: add documentation for 'git interpret-trailers'
+ trailer: add tests for commands in config file
+ trailer: execute command from 'trailer.<name>.command'
+ trailer: add tests for "git interpret-trailers"
+ trailer: add interpret-trailers command
+ trailer: put all the processing together and print
+ trailer: parse trailers from file or stdin
+ trailer: process command line trailer arguments
+ trailer: read and process config information
+ trailer: process trailers from input message and arguments
+ trailer: add data structures and basic functions

A new filter to programatically edit the tail end of the commit log
messages.

* po/everyday-doc (2014-10-10) 3 commits
(merged to 'next' on 2014-10-13 at daf1d03)
+ doc: add 'everyday' to 'git help'
+ doc: Makefile regularise OBSOLETE_HTML list building
+ doc: modernise everyday.txt wording and format in man page style

"git help everyday" to show the Everyday Git document.
* jk/test-shell-trace (2014-10-13) 3 commits
(merged to 'next' on 2014-10-14 at 9118506)
+ test-lib.sh: support -x option for shell-tracing
+ t5304: use helper to report failure of "test foo = bar"
+ t5304: use test_path_is_* instead of "test -f"
(this branch is used by jk/prune-mtime.)

--------------------------------------------------
[New Topics]
Test scripts were taught to notice "-x" option to show shell trace,
as if the tests were run under "sh -x".

* da/mergetool-temporary-directory (2014-10-16) 2 commits
- t7610-mergetool: add test cases for mergetool.writeToTemp
- mergetool: add an option for writing to a temporary directory
(this branch uses da/mergetool-temporary-filename and da/mergetool-tests; is tangled with da/mergetool-tool-help.)

Allow a temporary directory specified to be used while running "git
mergetool" backend.
* jn/parse-config-slot (2014-10-14) 2 commits
(merged to 'next' on 2014-10-14 at 7f0a3ad)
+ color_parse: do not mention variable name in error message
+ pass config slots as pointers instead of offsets

Will merge to 'next'.
Code cleanup.


* da/mergetool-tests (2014-10-16) 4 commits
- test-lib-functions: adjust style to match CodingGuidelines
- t7610-mergetool: use test_config to isolate tests
- t7610-mergetool: add missing && and remove commented-out code
- t7610-mergetool: use tabs instead of a mix of tabs and spaces
(this branch is used by da/mergetool-temporary-directory and da/mergetool-temporary-filename; is tangled with da/mergetool-tool-help.)
* rm/gitweb-start-form (2014-10-16) 1 commit
(merged to 'next' on 2014-10-17 at 7ca427f)
+ gitweb: use start_form, not startform that was removed in CGI.pm 4.04

The clean-up of this test script was long overdue and is a very
welcome change.

* rs/receive-pack-argv-leak-fix (2014-10-13) 1 commit
(merged to 'next' on 2014-10-14 at e64654d)
+ receive-pack: plug minor memory leak in unpack()


* ss/contrib-subtree-contacts (2014-10-15) 2 commits
(merged to 'next' on 2014-10-17 at 4b3a1d9)
+ contacts: add a Makefile to generate docs and install
+ subtree: add an install-html target


* ta/config-set (2014-10-13) 1 commit
(merged to 'next' on 2014-10-14 at ed660e4)
+ t1308: fix broken here document in test script

--------------------------------------------------
[New Topics]

* jk/pack-objects-no-bitmap-when-splitting (2014-10-19) 1 commit
- pack-objects: turn off bitmaps when we split packs

Splitting pack-objects output into multiple packs is incompatible
with the use of reachability bitmap.

Will merge to 'next'.


* bc/asciidoctor (2014-10-15) 2 commits
- Documentation: implement linkgit macro for Asciidoctor
- Documentation: move some AsciiDoc parameters into variables
(this branch uses bc/asciidoc.)
* po/doc-status-markup (2014-10-19) 1 commit
- doc: fix 'git status --help' character quoting

Add machinery to alternatively use AsciiDoctor to format our
documentation.
Update documentation mark-up.

Will merge to 'next'.


* da/mergetool-meld (2014-10-16) 1 commit
- mergetools/meld: make usage of `--output` configurable and more robust
* rs/run-command-env-array (2014-10-19) 2 commits
- use env_array member of struct child_process
- run-command: add env_array, an optional argv_array for env

Newer versions of 'meld' breaks the auto-detection we use to see if
they are new enough to support the `--output` option.
Add managed "env" array to child_process to clarify the lifetime
rules.

Will merge to 'next'.


* rm/gitweb-start-form (2014-10-16) 1 commit
- gitweb: use start_form, not startform that was removed in CGI.pm 4.04
* tb/core-filemode-doc (2014-10-19) 1 commit
- core.filemode may need manual action

Doc update.

Will merge to 'next'.


* ss/contrib-subtree-contacts (2014-10-15) 2 commits
- contacts: add a Makefile to generate docs and install
- subtree: add an install-html target
* wk/t1304-wo-USER (2014-10-19) 1 commit
- t1304: Set LOGNAME even if USER is unset or null

Will merge to 'next'.

@ -114,15 +155,6 @@ of the repositories listed at @@ -114,15 +155,6 @@ of the repositories listed at
Waiting for an Ack.


* eb/no-pthreads (2014-10-13) 2 commits
- pack-objects: set number of threads before checking and warning
- index-pack: fix compilation with NO_PTHREADS

Allow us build with NO_PTHREADS=NoThanks compilation option. The
last change (not queued) needs a bit more explanation in its log
message.


* tr/remerge-diff (2014-09-08) 8 commits
- log --remerge-diff: show what the conflict resolution changed
- name-hash: allow dir hashing even when !ignore_case
@ -279,123 +311,153 @@ of the repositories listed at @@ -279,123 +311,153 @@ of the repositories listed at
--------------------------------------------------
[Cooking]

* jk/prune-mtime (2014-10-16) 25 commits
- pack-objects: double-check options before discarding objects
- repack: pack objects mentioned by the index
- pack-objects: use argv_array
- reachable: use revision machinery's --index-objects code
- rev-list: add --index-objects option
- rev-list: document --reflog option
- traverse_commit_list: support pending blobs/trees with paths
- make add_object_array_with_context interface more sane
- write_sha1_file: freshen existing objects
- pack-objects: match prune logic for discarding objects
- pack-objects: refactor unpack-unreachable expiration check
- prune: keep objects reachable from recent objects
- sha1_file: add for_each iterators for loose and packed objects
- count-objects: use for_each_loose_file_in_objdir
- count-objects: do not use xsize_t when counting object size
- prune-packed: use for_each_loose_file_in_objdir
- reachable: mark index blobs as SEEN
- prune: factor out loose-object directory traversal
- reachable: reuse revision.c "add all reflogs" code
- reachable: use traverse_commit_list instead of custom walk
- clean up name allocation in prepare_revision_walk
- object_array: add a "clear" function
- object_array: factor out slopbuf-freeing logic
- isxdigit: cast input to unsigned char
- foreach_alt_odb: propagate return value from callback
(this branch uses jk/test-shell-trace.)

Looked mostly sensible modulo minor nits (and I wasn't looking for
typos very carefully). Object enumeration is broken.
* eb/no-pthreads (2014-10-19) 3 commits
- Handle atexit list internaly for unthreaded builds
- pack-objects: set number of threads before checking and warning
- index-pack: fix compilation with NO_PTHREADS

Allow us build with NO_PTHREADS=NoThanks compilation option. The
last change (not queued) needs a bit more explanation in its log
message.

* rs/ref-transaction (2014-10-15) 25 commits
- ref_transaction_commit: bail out on failure to remove a ref
- lockfile: remove unable_to_lock_error
- refs.c: do not permit err == NULL
- remote rm/prune: print a message when writing packed-refs fails
- for-each-ref: skip and warn about broken ref names
- refs.c: allow listing and deleting badly named refs
- test: put tests for handling of bad ref names in one place
- packed-ref cache: forbid dot-components in refnames
- branch -d: simplify by using RESOLVE_REF_READING
- branch -d: avoid repeated symref resolution
- reflog test: test interaction with detached HEAD
- refs.c: change resolve_ref_unsafe reading argument to be a flags field
- refs.c: make write_ref_sha1 static
- fetch.c: change s_update_ref to use a ref transaction
- refs.c: ref_transaction_commit: distinguish name conflicts from other errors
- refs.c: pass a list of names to skip to is_refname_available
- refs.c: call lock_ref_sha1_basic directly from commit
- refs.c: refuse to lock badly named refs in lock_ref_sha1_basic
- rename_ref: don't ask read_ref_full where the ref came from
- refs.c: pass the ref log message to _create/delete/update instead of _commit
- refs.c: add an err argument to delete_ref_loose
- wrapper.c: add a new function unlink_or_msg
- refs.c: lock_ref_sha1_basic is used for all refs
- wrapper.c: remove/unlink_or_warn: simplify, treat ENOENT as success
- mv test: recreate mod/ directory instead of relying on stale copy
Will merge to 'next'.

The API to update refs have been restructured to allow introducing
a true transactional updates later. We would even allow storing
refs in backends other than the traditional filesystem-based one.

Will merge to 'next'.
* da/mergetool-temporary-directory (2014-10-16) 2 commits
(merged to 'next' on 2014-10-17 at 261e74e)
+ t7610-mergetool: add test cases for mergetool.writeToTemp
+ mergetool: add an option for writing to a temporary directory
(this branch uses da/mergetool-temporary-filename and da/mergetool-tests; is tangled with da/mergetool-tool-help.)

Allow a temporary directory specified to be used while running "git
mergetool" backend.

* bc/asciidoc (2014-10-13) 2 commits
(merged to 'next' on 2014-10-14 at d9ff206)
+ Documentation: fix mismatched delimiters in git-imap-send
+ Documentation: adjust document title underlining
(this branch is used by bc/asciidoctor.)
Will merge to 'master'.

Formatting nitpicks to help a (pickier) reimplementation of
AsciiDoc to grok our documentation.

* da/mergetool-tests (2014-10-16) 4 commits
(merged to 'next' on 2014-10-17 at 239fefe)
+ test-lib-functions: adjust style to match CodingGuidelines
+ t7610-mergetool: use test_config to isolate tests
+ t7610-mergetool: add missing && and remove commented-out code
+ t7610-mergetool: use tabs instead of a mix of tabs and spaces
(this branch is used by da/mergetool-temporary-directory and da/mergetool-temporary-filename; is tangled with da/mergetool-tool-help.)

The clean-up of this test script was long overdue and is a very
welcome change.

Will merge to 'master'.


* da/mergetool-tool-help (2014-10-15) 8 commits
- difftool: don't assume that default sh is sane
- mergetool: don't require a work tree for --tool-help
- git-sh-setup: move GIT_DIR initialization into a function
- mergetool: use more conservative temporary filenames
- test-lib-functions: adjust style to match CodingGuidelines
- t7610-mergetool: prefer test_config over git config
- t7610-mergetool: add missing && and remove commented-out code
- t7610-mergetool: use tabs instead of a mix of tabs and spaces
(this branch is tangled with da/mergetool-temporary-directory, da/mergetool-temporary-filename and da/mergetool-tests.)
* bc/asciidoctor (2014-10-15) 2 commits
(merged to 'next' on 2014-10-17 at 12ce2e2)
+ Documentation: implement linkgit macro for Asciidoctor
+ Documentation: move some AsciiDoc parameters into variables

Allow "git mergetool --help" to run outside a Git repository.
Add machinery to alternatively use AsciiDoctor to format our
documentation.

Will merge to 'next'.
Brian seems to have a bit more tweaks; will wait for them before
moving this topic forward.


* jk/test-shell-trace (2014-10-13) 3 commits
(merged to 'next' on 2014-10-14 at 9118506)
+ test-lib.sh: support -x option for shell-tracing
+ t5304: use helper to report failure of "test foo = bar"
+ t5304: use test_path_is_* instead of "test -f"
(this branch is used by jk/prune-mtime.)
* da/mergetool-meld (2014-10-16) 1 commit
(merged to 'next' on 2014-10-17 at 00ee553)
+ mergetools/meld: make usage of `--output` configurable and more robust

Test scripts were taught to notice "-x" option to show shell trace,
as if the tests were run under "sh -x".
Newer versions of 'meld' breaks the auto-detection we use to see if
they are new enough to support the `--output` option.

Will merge to 'master'.


* rs/receive-pack-argv-leak-fix (2014-10-13) 1 commit
(merged to 'next' on 2014-10-14 at e64654d)
+ receive-pack: plug minor memory leak in unpack()
* jk/prune-mtime (2014-10-19) 28 commits
(merged to 'next' on 2014-10-20 at fdd7ff6)
+ drop add_object_array_with_mode
+ revision: remove definition of unused 'add_object' function
+ pack-objects: double-check options before discarding objects
+ repack: pack objects mentioned by the index
+ pack-objects: use argv_array
+ reachable: use revision machinery's --indexed-objects code
+ rev-list: add --indexed-objects option
+ rev-list: document --reflog option
+ t5516: test pushing a tag of an otherwise unreferenced blob
+ traverse_commit_list: support pending blobs/trees with paths
+ make add_object_array_with_context interface more sane
+ write_sha1_file: freshen existing objects
+ pack-objects: match prune logic for discarding objects
+ pack-objects: refactor unpack-unreachable expiration check
+ prune: keep objects reachable from recent objects
+ sha1_file: add for_each iterators for loose and packed objects
+ count-objects: use for_each_loose_file_in_objdir
+ count-objects: do not use xsize_t when counting object size
+ prune-packed: use for_each_loose_file_in_objdir
+ reachable: mark index blobs as SEEN
+ prune: factor out loose-object directory traversal
+ reachable: reuse revision.c "add all reflogs" code
+ reachable: use traverse_commit_list instead of custom walk
+ clean up name allocation in prepare_revision_walk
+ object_array: add a "clear" function
+ object_array: factor out slopbuf-freeing logic
+ isxdigit: cast input to unsigned char
+ foreach_alt_odb: propagate return value from callback

Tighten the logic to decide that an unreachable cruft is
sufficiently old by covering corner cases such as an ancient object
becoming reachable and then going unreachable again, in which case
its retention period should be prolonged.

Will merge to 'master'.


* ta/config-set (2014-10-13) 1 commit
(merged to 'next' on 2014-10-14 at ed660e4)
+ t1308: fix broken here document in test script
* rs/ref-transaction (2014-10-15) 25 commits
(merged to 'next' on 2014-10-17 at 1390624)
+ ref_transaction_commit: bail out on failure to remove a ref
+ lockfile: remove unable_to_lock_error
+ refs.c: do not permit err == NULL
+ remote rm/prune: print a message when writing packed-refs fails
+ for-each-ref: skip and warn about broken ref names
+ refs.c: allow listing and deleting badly named refs
+ test: put tests for handling of bad ref names in one place
+ packed-ref cache: forbid dot-components in refnames
+ branch -d: simplify by using RESOLVE_REF_READING
+ branch -d: avoid repeated symref resolution
+ reflog test: test interaction with detached HEAD
+ refs.c: change resolve_ref_unsafe reading argument to be a flags field
+ refs.c: make write_ref_sha1 static
+ fetch.c: change s_update_ref to use a ref transaction
+ refs.c: ref_transaction_commit: distinguish name conflicts from other errors
+ refs.c: pass a list of names to skip to is_refname_available
+ refs.c: call lock_ref_sha1_basic directly from commit
+ refs.c: refuse to lock badly named refs in lock_ref_sha1_basic
+ rename_ref: don't ask read_ref_full where the ref came from
+ refs.c: pass the ref log message to _create/delete/update instead of _commit
+ refs.c: add an err argument to delete_ref_loose
+ wrapper.c: add a new function unlink_or_msg
+ refs.c: lock_ref_sha1_basic is used for all refs
+ wrapper.c: remove/unlink_or_warn: simplify, treat ENOENT as success
+ mv test: recreate mod/ directory instead of relying on stale copy

The API to update refs have been restructured to allow introducing
a true transactional updates later. We would even allow storing
refs in backends other than the traditional filesystem-based one.

Will merge to 'master'.


* da/mergetool-tool-help (2014-10-15) 8 commits
(merged to 'next' on 2014-10-17 at 854654c)
+ difftool: don't assume that default sh is sane
+ mergetool: don't require a work tree for --tool-help
+ git-sh-setup: move GIT_DIR initialization into a function
+ mergetool: use more conservative temporary filenames
+ test-lib-functions: adjust style to match CodingGuidelines
+ t7610-mergetool: prefer test_config over git config
+ t7610-mergetool: add missing && and remove commented-out code
+ t7610-mergetool: use tabs instead of a mix of tabs and spaces
(this branch is tangled with da/mergetool-temporary-directory, da/mergetool-temporary-filename and da/mergetool-tests.)

Allow "git mergetool --help" to run outside a Git repository.

Will merge to 'master'.

@ -407,6 +469,8 @@ of the repositories listed at @@ -407,6 +469,8 @@ of the repositories listed at
local branch by setting its branch.*.merge to branch name (not a
full refname) incorrectly said that the upstream is gone.

Will merge to 'next'.


* jc/clone-borrow (2014-10-15) 1 commit
- clone: --dissociate option to mark that reference is only temporary
@ -415,25 +479,19 @@ of the repositories listed at @@ -415,25 +479,19 @@ of the repositories listed at


* js/completion-hide-not-a-repo (2014-10-14) 1 commit
- completion: silence "fatal: Not a git repository" error
(merged to 'next' on 2014-10-17 at 466de3c)
+ completion: silence "fatal: Not a git repository" error

Some internal error messages leaked out of the bash completion when
typing "git cmd <TAB>" and the machinery tried to complete
refnames.


* jn/parse-config-slot (2014-10-14) 2 commits
(merged to 'next' on 2014-10-14 at 7f0a3ad)
+ color_parse: do not mention variable name in error message
+ pass config slots as pointers instead of offsets

Code cleanup.

Will merge to 'master'.


* da/mergetool-temporary-filename (2014-10-16) 1 commit
- mergetool: use more conservative temporary filenames
(merged to 'next' on 2014-10-17 at 3fb95b5)
+ mergetool: use more conservative temporary filenames
(this branch is used by da/mergetool-temporary-directory; uses da/mergetool-tests; is tangled with da/mergetool-tool-help.)

Tweak the names of the three throw-away files "git mergetool" comes
@ -441,7 +499,7 @@ of the repositories listed at @@ -441,7 +499,7 @@ of the repositories listed at
single dot in its name in the original (e.g. "hello.c") will have
only one dot in these variants (e.g. "hello_BASE_4321.c").

Will merge to 'next'.
Will merge to 'master'.


* jc/push-cert-hmac-optim (2014-09-25) 2 commits
@ -497,23 +555,6 @@ of the repositories listed at @@ -497,23 +555,6 @@ of the repositories listed at
reroll.


* cc/interpret-trailers (2014-10-13) 11 commits
- Documentation: add documentation for 'git interpret-trailers'
- trailer: add tests for commands in config file
- trailer: execute command from 'trailer.<name>.command'
- trailer: add tests for "git interpret-trailers"
- trailer: add interpret-trailers command
- trailer: put all the processing together and print
- trailer: parse trailers from file or stdin
- trailer: process command line trailer arguments
- trailer: read and process config information
- trailer: process trailers from input message and arguments
- trailer: add data structures and basic functions

A new filter to programatically edit the tail end of the commit log
messages.


* mt/patch-id-stable (2014-06-10) 1 commit
- patch-id: change default to stable


Loading…
Cancel
Save