What's cooking (2011/05 #06)

Junio C Hamano committed May 13, 2011 at 12:53 UTC ceb2b86a81374153f7ace727a29fb7b27716eef0
1 file changed +135 -162
whats-cooking.txt
+135 -162
index f50055b056..9ee39acf74 100644 --- a/whats-cooking.txt +++ b/whats-cooking.txt @@ -1,115 +1,113 @@ To: git@vger.kernel.org -Subject: What's cooking in git.git (May 2011, #05; Wed, 11) -X-master-at: dbe9c5adf95e99c35dd728af2114fccfb9117c65 -X-next-at: 5aaf5cf1a3addf544db48634fb8abc1aec4e8174 +Subject: What's cooking in git.git (May 2011, #06; Fri, 13) +X-master-at: 2de58b398b5a1f81bf74a385e41cfc2fcd9cb427 +X-next-at: fcbf71a2ac7af454bec14832c8dfba57e793dda7 -What's cooking in git.git (May 2011, #05; Wed, 11) +What's cooking in git.git (May 2011, #06; Fri, 13) -------------------------------------------------- Here are the topics that have been cooking. Commits prefixed with '-' are only in 'pu' while commits prefixed with '+' are in 'next'. +I am somewhat frustrated that I so almost no reviews on gitweb +patches. Does this mean nobody is interested in viewing the projects +on gitweb? How can we improve this lack of review bandwidth situation? + The main part of this cycle is expected to run thru May, aiming for a feature freeze in early June. -We are in the middle of week #3. +We are about to finish the week #3 of this cycle. -------------------------------------------------- [Graduated to "master"] -* dm/http-cleanup (2011-05-04) 4 commits - (merged to 'next' on 2011-05-06 at 711ff78) - + t5541-http-push: add test for chunked - + http-push: refactor curl_easy_setup madness - + http-push: use const for strings in signatures - + http: make curl callbacks match contracts from curl header - -* jc/maint-add-p-overlapping-hunks (2011-05-08) 5 commits - (merged to 'next' on 2011-05-08 at bcf9228) - + t3701: add-p-fix makes the last test to pass - (merged to 'next' on 2011-05-02 at e57b66f) - + "add -p": work-around an old laziness that does not coalesce hunks - + add--interactive.perl: factor out repeated --recount option - + t3701: Editing a split hunk in an "add -p" session - + add -p: 'q' should really quit - -* jc/maint-branch-mergeoptions (2011-05-06) 1 commit - (merged to 'next' on 2011-05-08 at fa6b154) - + merge: make branch.<name>.mergeoptions correctly override merge.<option> - (this branch is used by mg/merge-ff-config.) - -* jn/ctags (2011-04-29) 3 commits - (merged to 'next' on 2011-05-06 at 5c574ff) - + gitweb: Mark matched 'ctag' / contents tag (?by_tag=foo) - + gitweb: Change the way "content tags" ('ctags') are handled - + gitweb: Restructure projects list generation - (this branch is used by jn/ctags-more.) - -* jn/gitweb-dependency (2011-05-08) 2 commits - (merged to 'next' on 2011-05-08 at 9101342) - + Remove gitweb/gitweb.cgi and other legacy targets from main Makefile - + git-instaweb: Simplify build dependency on gitweb - -* jn/maint-test-merge-verify-parents (2011-05-06) 2 commits - (merged to 'next' on 2011-05-08 at 35a7c66) - + tests: teach verify_parents to check for extra parents - + tests: eliminate unnecessary setup test assertions - -* sr/maint-fast-import-tighten-option-parsing (2011-05-05) 1 commit - (merged to 'next' on 2011-05-08 at 367b148) - + fast-import: fix option parser for no-arg options - -* vh/config-interactive-singlekey-doc (2011-05-05) 5 commits - (merged to 'next' on 2011-05-06 at 46b522c) - + git-reset.txt: better docs for '--patch' - + git-checkout.txt: better docs for '--patch' - + git-stash.txt: better docs for '--patch' - + git-add.txt: document 'interactive.singlekey' - + config.txt: 'interactive.singlekey; is used by... - (this branch is used by ci/commit--interactive-atomic.) +* bf/commit-template-no-cleanup (2011-05-11) 1 commit + (merged to 'next' on 2011-05-11 at 01fde0d) + + Do not strip empty lines / trailing spaces from a commit message template + +* cn/log-parse-opt (2011-04-14) 1 commit + (merged to 'next' on 2011-04-28 at 02f2eac) + + log: convert to parse-options + +* jc/fix-add-u-unmerged (2011-04-23) 1 commit + (merged to 'next' on 2011-04-28 at f7ed821) + + Fix "add -u" that sometimes fails to resolve unmerged paths + (this branch is used by jc/add-delete-default.) + +* jc/t1506-shell-param-expansion-gotcha (2011-05-09) 1 commit + (merged to 'next' on 2011-05-11 at 6c8619e) + + t1507: avoid "${parameter<op>'word'}" inside double-quotes + +* jh/dirstat-lines (2011-04-29) 8 commits + (merged to 'next' on 2011-04-29 at a302674) + + Mark dirstat error messages for translation + + Improve error handling when parsing dirstat parameters + + New --dirstat=lines mode, doing dirstat analysis based on diffstat + + Allow specifying --dirstat cut-off percentage as a floating point number + + Add config variable for specifying default --dirstat behavior + + Refactor --dirstat parsing; deprecate --cumulative and --dirstat-by-file + + Make --dirstat=0 output directories that contribute < 0.1% of changes + + Add several testcases for --dirstat and friends + +* jn/setup-revisions-glob-and-friends-passthru (2011-04-21) 2 commits + (merged to 'next' on 2011-04-28 at 6006cc4) + + revisions: allow --glob and friends in parse_options-enabled commands + + revisions: split out handle_revision_pseudo_opt function + +* js/maint-1.6.6-send-pack-stateless-rpc-deadlock-fix (2011-05-05) 2 commits + + send-pack: unbreak push over stateless rpc + + send-pack: avoid deadlock when pack-object dies early + (this branch is used by js/maint-send-pack-stateless-rpc-deadlock-fix.) + +* js/maint-send-pack-stateless-rpc-deadlock-fix (2011-05-05) 2 commits + (merged to 'next' on 2011-05-05 at 3f4ffb9) + + Merge branch 'js/maint-1.6.6-send-pack-stateless-rpc-deadlock-fix' into js/maint-send-pack-stateless-rpc-deadlock-fix + (merged to 'next' on 2011-04-28 at db7e04a) + + Merge branch 'js/maint-1.6.6-send-pack-stateless-rpc-deadlock-fix' into js/maint-send-pack-stateless-rpc-deadlock-fix + (this branch uses js/maint-1.6.6-send-pack-stateless-rpc-deadlock-fix.) + +* rr/rerere-libify-clear-gc (2011-05-08) 1 commit + (merged to 'next' on 2011-05-11 at b1d8c88) + + rerere: libify rerere_clear() and rerere_gc() + +This is pure code movement. Changing them to suit the need of other +callers can be built on it as separate patches. -------------------------------------------------- [New Topics] -* jc/bigfile (2011-05-09) 3 commits - - Bigfile: teach "git add" to send a large file straight to a pack - - index_fd(): split into two helper functions - - index_fd(): turn write_object and format_check arguments into one flag +* jc/maint-1.6.6-pathspec-stdin-and-cmdline (2011-05-11) 1 commit + + setup_revisions(): take pathspec from command line and --stdin correctly + (this branch is used by jc/magic-pathspec, jc/maint-1.7.4-pathspec-stdin-and-cmdline and jc/maint-pathspec-stdin-and-cmdline.) -This covers the entry point for a big file to the system. Other parts that -need to know about them are the exit point (i.e. write_entry()), packing -and repacking (as long as bigfilethreshold is sane this should work), -diffs and status (avoid slurping large binary in core only to do nothing, -which we already should), and transport (receive-pack/fetch-pack call -either unpack-objects or index-pack, which still want to hold the full -object in-core and need to be fixed). +* jc/maint-1.7.4-pathspec-stdin-and-cmdline (2011-05-11) 1 commit + - Merge branch 'jc/maint-1.6.6-pathspec-stdin-and-cmdline' + (this branch uses jc/maint-1.6.6-pathspec-stdin-and-cmdline; is tangled with jc/magic-pathspec and jc/maint-pathspec-stdin-and-cmdline.) -* jc/convert (2011-05-09) 4 commits - - convert: make it harder to screw up adding a conversion attribute - - convert: make it safer to add conversion attributes - - convert: give saner names to crlf/eol variables, types and functions - - convert: rename the "eol" global variable to "core_eol" - -Just a clean-up. -Will merge to "next" by the end of week #3. +* jc/maint-pathspec-stdin-and-cmdline (2011-05-11) 2 commits + - Merge branch 'jc/maint-1.7.4-pathspec-stdin-and-cmdline' + - Merge branch 'jc/maint-1.6.6-pathspec-stdin-and-cmdline' + (this branch uses jc/maint-1.6.6-pathspec-stdin-and-cmdline; is tangled with jc/magic-pathspec and jc/maint-1.7.4-pathspec-stdin-and-cmdline.) -* jk/blame-line-porcelain (2011-05-09) 3 commits - - blame: add --line-porcelain output format - - blame: refactor porcelain output - - add tests for various blame formats +These are to fix an ancient bug where feeding "git log --stdin path" with +an input that also has additional pathspec corrupted memory. -* ms/tagname-does-not-begin-with-dash (2011-05-10) 1 commit - - tag: disallow '-' as tag name +* jl/read-tree-m-dry-run (2011-05-11) 1 commit + - Teach read-tree the -n|--dry-run option -Will merge to "next" by the end of week #3. +Looked good. We _might_ want to make -n simply ignore -u, though. -* fc/completion-zsh (2011-05-10) 1 commit - + git-completion: fix regression in zsh support - (this branch is used by sg/completion-updates.) +* jc/streaming (2011-05-12) 5 commits + - write_entry(): use streaming API for writing working tree files + - stream: a streaming interface to read from the object store + - sha1_object_info_extended(): expose a bit more information from object-info + - write_entry(): separate two helper functions out + - packed_object_info_detail(): do not return a string + (this branch uses jc/convert.) -* nd/sparse-co-fix (2011-05-10) 2 commits - - sparse checkout: do not eagerly decide the fate for whole directory - - t1011: fix sparse-checkout initialization and add new file +The counterpart to jc/bigfile to cover the write-out codepath. This does +not yet have "read directly from pack or loose object without slurping the +whole thing in memory" yet, which is still work in progress. -------------------------------------------------- [Stalled] @@ -153,7 +151,6 @@ is a sane use case, though. May drop. * jc/add-delete-default (2011-04-27) 1 commit - git add: notice removal of tracked paths by default - (this branch uses jc/fix-add-u-unmerged.) * jk/maint-merge-rename-create (2011-03-25) 3 commits - merge: turn on rewrite detection @@ -197,6 +194,47 @@ Needs to tighten the advice to only govern the generic How-to part. -------------------------------------------------- [Cooking] +* jc/bigfile (2011-05-12) 3 commits + - Bigfile: teach "git add" to send a large file straight to a pack + - index_fd(): split into two helper functions + - index_fd(): turn write_object and format_check arguments into one flag + +This covers the entry point for a big file to the system. Other parts that +need to know about them are the exit point (i.e. write_entry()), packing +and repacking (as long as bigfilethreshold is sane this should work), +diffs and status (avoid slurping large binary in core only to do nothing, +which we already should), and transport (receive-pack/fetch-pack call +either unpack-objects or index-pack, which still want to hold the full +object in-core and need to be fixed). + +* jc/convert (2011-05-09) 4 commits + - convert: make it harder to screw up adding a conversion attribute + - convert: make it safer to add conversion attributes + - convert: give saner names to crlf/eol variables, types and functions + - convert: rename the "eol" global variable to "core_eol" + (this branch is used by jc/streaming.) + +Just a clean-up. +Will merge to "next" by the end of week #3. + +* jk/blame-line-porcelain (2011-05-09) 3 commits + - blame: add --line-porcelain output format + - blame: refactor porcelain output + - add tests for various blame formats + +* ms/tagname-does-not-begin-with-dash (2011-05-10) 1 commit + - tag: disallow '-' as tag name + +Will merge to "next" by the end of week #3. + +* fc/completion-zsh (2011-05-10) 1 commit + + git-completion: fix regression in zsh support + (this branch is used by sg/completion-updates.) + +* nd/sparse-co-fix (2011-05-10) 2 commits + - sparse checkout: do not eagerly decide the fate for whole directory + - t1011: fix sparse-checkout initialization and add new file + * kk/maint-prefix-in-config-mak (2011-05-09) 3 commits (merged to 'next' on 2011-05-09 at 267a3aa) + Honor $(prefix) set in config.mak* when defining ETC_GIT* @@ -209,12 +247,6 @@ reasonable. Will merge to "master" by the end of week #4. -* bf/commit-template-no-cleanup (2011-05-11) 1 commit - (merged to 'next' on 2011-05-11 at 01fde0d) - + Do not strip empty lines / trailing spaces from a commit message template - -Will merge to "master" by the end of week #3. - * ci/commit--interactive-atomic (2011-05-10) 6 commits (merged to 'next' on 2011-05-11 at 7f36fd8) + Test atomic git-commit --interactive @@ -226,15 +258,6 @@ Will merge to "master" by the end of week #3. Will merge to "master" by the end of week #4. -* rr/rerere-libify-clear-gc (2011-05-08) 1 commit - (merged to 'next' on 2011-05-11 at b1d8c88) - + rerere: libify rerere_clear() and rerere_gc() - -This is pure code movement. Changing them to suit the need of other -callers can be built on it as separate patches. - -Will merge to "master" by the end of week #3. - * ab/i18n-scripts (2011-05-08) 48 commits - i18n: git-bisect bisect_next_check "You need to" message - i18n: git-bisect [Y/n] messages @@ -293,13 +316,8 @@ Will merge to "master" by the end of week #3. - git-sh-i18n--envsubst: our own envsubst(1) for eval_gettext() (this branch is used by ab/i18n-scripts.) -Will merge to "next" by the end of week #4. - -* jc/t1506-shell-param-expansion-gotcha (2011-05-09) 1 commit - (merged to 'next' on 2011-05-11 at 6c8619e) - + t1507: avoid "${parameter<op>'word'}" inside double-quotes - -Will merge to "master" by the end of week #3. +Will be re-rolled to work around systems whose environment variable names +are case insensitive. * mg/merge-ff-config (2011-05-06) 3 commits (merged to 'next' on 2011-05-08 at 977b432) @@ -322,8 +340,6 @@ Will merge to "master" by the end of week #4. + contrib/completion: --line-number to git grep + Documentation: Add --line-number to git-grep synopsis -The tip is my "how about this" fix-up for the "Bail out" one. - Will merge to "next" by the end of week #3. * sg/completion-updates (2011-05-10) 4 commits @@ -350,56 +366,12 @@ Will merge to "master" by the end of week #4. The tip one will be re-rolled. -* jh/dirstat-lines (2011-04-29) 8 commits - (merged to 'next' on 2011-04-29 at a302674) - + Mark dirstat error messages for translation - + Improve error handling when parsing dirstat parameters - + New --dirstat=lines mode, doing dirstat analysis based on diffstat - + Allow specifying --dirstat cut-off percentage as a floating point number - + Add config variable for specifying default --dirstat behavior - + Refactor --dirstat parsing; deprecate --cumulative and --dirstat-by-file - + Make --dirstat=0 output directories that contribute < 0.1% of changes - + Add several testcases for --dirstat and friends - -Will merge to "master" by the end of week #3. - -* jn/setup-revisions-glob-and-friends-passthru (2011-04-21) 2 commits - (merged to 'next' on 2011-04-28 at 6006cc4) - + revisions: allow --glob and friends in parse_options-enabled commands - + revisions: split out handle_revision_pseudo_opt function - -Will merge to "master" by the end of week #3. - -* cn/log-parse-opt (2011-04-14) 1 commit - (merged to 'next' on 2011-04-28 at 02f2eac) - + log: convert to parse-options - -Will merge to "master" by the end of week #3. - -* jc/fix-add-u-unmerged (2011-04-23) 1 commit - (merged to 'next' on 2011-04-28 at f7ed821) - + Fix "add -u" that sometimes fails to resolve unmerged paths - (this branch is used by jc/add-delete-default.) - -Will merge to "master" by the end of week #3. - -* js/maint-1.6.6-send-pack-stateless-rpc-deadlock-fix (2011-05-05) 2 commits - + send-pack: unbreak push over stateless rpc - + send-pack: avoid deadlock when pack-object dies early - (this branch is used by js/maint-send-pack-stateless-rpc-deadlock-fix.) - -Will merge to "master" by the end of week #3. - -* js/maint-send-pack-stateless-rpc-deadlock-fix (2011-05-05) 2 commits - (merged to 'next' on 2011-05-05 at 3f4ffb9) - + Merge branch 'js/maint-1.6.6-send-pack-stateless-rpc-deadlock-fix' into js/maint-send-pack-stateless-rpc-deadlock-fix - (merged to 'next' on 2011-04-28 at db7e04a) - + Merge branch 'js/maint-1.6.6-send-pack-stateless-rpc-deadlock-fix' into js/maint-send-pack-stateless-rpc-deadlock-fix - (this branch uses js/maint-1.6.6-send-pack-stateless-rpc-deadlock-fix.) - -Will merge to "master" by the end of week #3. - -* jc/magic-pathspec (2011-05-10) 10 commits +* jc/magic-pathspec (2011-05-12) 13 commits + (merged to 'next' on 2011-05-12 at 656c8b5) + + t3703: Skip tests using directory name ":" on Windows + (merged to 'next' on 2011-05-11 at 86d1d6d) + + revision.c: leave a note for "a lone :" enhancement + + Merge branch 'jc/maint-1.6.6-pathspec-stdin-and-cmdline' (merged to 'next' on 2011-05-11 at aaad95e) + t3703, t4208: add test cases for magic pathspec + rev/path disambiguation: further restrict "misspelled index entry" diag @@ -412,5 +384,6 @@ Will merge to "master" by the end of week #3. + magic pathspec: add ":(icase)path" to match case insensitively + magic pathspec: futureproof shorthand form + magic pathspec: add tentative ":/path/from/top/level" pathspec support + (this branch uses jc/maint-1.6.6-pathspec-stdin-and-cmdline; is tangled with jc/maint-1.7.4-pathspec-stdin-and-cmdline and jc/maint-pathspec-stdin-and-cmdline.) This seems to need more time to mature than I expected.