{"thread":{"id":"46898","subject":"What's cooking in git.git (Oct 2017, #01; Wed, 4)","startedAt":"2017-10-04T07:20:05Z","lastAt":"2017-10-05T05:21:13Z","messageCount":7,"participants":["Junio C Hamano","Jeff King","Taylor Blau"],"isPatch":false,"patchVersion":null,"patchTotal":null},"messages":[{"id":"329661","messageId":"xmqqy3orbcrr.fsf@gitster.mtv.corp.google.com","threadId":"46898","inReplyTo":null,"subject":"What's cooking in git.git (Oct 2017, #01; Wed, 4)","fromName":"Junio C Hamano","fromEmail":"gitster@pobox.com","sentAt":"2017-10-04T07:19:52Z","receivedAt":"2017-10-04T07:20:05Z","isPatch":false,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"Here are the topics that have been cooking.  Commits prefixed with\n'-' are only in 'pu' (proposed updates) while commits prefixed with\n'+' are in 'next'.  The ones marked with '.' do not appear in any of\nthe integration branches, but I am still holding onto them.\n\nI wanted to do v2.15-rc0 today but it seems that this has to slip,\nwith too many topics still in flight in 'next' among which many are\nfixes, one is even a fix for a recent regression.\n\nYou can find the changes described here in the integration branches\nof the repositories listed at\n\n    http://git-blame.blogspot.com/p/git-public-repositories.html\n\n--------------------------------------------------\n[Graduated to \"master\"]\n\n* ad/doc-markup-fix (2017-09-29) 1 commit\n  (merged to 'next' on 2017-10-02 at fdf600e8f3)\n + doc: correct command formatting\n\n Docfix.\n\n\n* bc/rev-parse-parseopt-fix (2017-09-25) 7 commits\n  (merged to 'next' on 2017-09-26 at f3e013eaa4)\n + parse-options: only insert newline in help text if needed\n + parse-options: write blank line to correct output stream\n + t0040,t1502: Demonstrate parse_options bugs\n  (merged to 'next' on 2017-09-24 at e479bce0ff)\n + git-rebase: don't ignore unexpected command line arguments\n + rev-parse parseopt: interpret any whitespace as start of help text\n + rev-parse parseopt: do not search help text for flag chars\n + t1502: demonstrate rev-parse --parseopt option mis-parsing\n\n Recent versions of \"git rev-parse --parseopt\" did not parse the\n option specification that does not have the optional flags (*=?!)\n correctly, which has been corrected.\n\n\n* bw/git-clang-format (2017-10-01) 1 commit\n  (merged to 'next' on 2017-10-02 at ee84c89b07)\n + clang-format: adjust line break penalties\n (this branch is used by sb/git-clang-format.)\n\n Adjust clang-format penalty parameters.\n\n\n* hn/path-ownership-comment (2017-09-27) 2 commits\n  (merged to 'next' on 2017-09-28 at 7cf8481c29)\n + read_gitfile_gently: clarify return value ownership.\n + real_path: clarify return value ownership\n\n Add comment to a few functions that use a short-lived buffer the\n caller can peek and copy out of.\n\n\n* hn/string-list-doc (2017-09-27) 1 commit\n  (merged to 'next' on 2017-09-28 at ffc8f65f4a)\n + string-list.h: move documentation from Documentation/api/ into header\n\n Doc reorg.\n\n\n* hn/submodule-comment (2017-09-26) 1 commit\n  (merged to 'next' on 2017-09-28 at a071ca3fde)\n + submodule.c: describe submodule_to_gitdir() in a new comment\n\n\n* jk/no-optional-locks (2017-09-27) 1 commit\n  (merged to 'next' on 2017-09-28 at eaffe9c638)\n + git: add --no-optional-locks option\n\n Some commands (most notably \"git status\") makes an opportunistic\n update when performing a read-only operation to help optimize later\n operations in the same repository.  The new \"--no-optional-locks\"\n option can be passed to Git to disable them.\n\n\n* jk/read-in-full (2017-09-27) 7 commits\n  (merged to 'next' on 2017-09-28 at 9d109b9233)\n + worktree: check the result of read_in_full()\n + worktree: use xsize_t to access file size\n + distinguish error versus short read from read_in_full()\n + avoid looking at errno for short read_in_full() returns\n + prefer \"!=\" when checking read_in_full() result\n + notes-merge: drop dead zero-write code\n + files-backend: prefer \"0\" for write_in_full() error check\n\n Code clean-up to prevent future mistakes by copying and pasting\n code that checks the result of read_in_full() function.\n\n\n* jk/validate-headref-fix (2017-09-27) 3 commits\n  (merged to 'next' on 2017-09-28 at dcea9d16f9)\n + validate_headref: use get_oid_hex for detached HEADs\n + validate_headref: use skip_prefix for symref parsing\n + validate_headref: NUL-terminate HEAD buffer\n\n Code clean-up.\n\n\n* js/rebase-i-final (2017-07-27) 10 commits\n  (merged to 'next' on 2017-09-26 at ea3f8f5e11)\n + rebase -i: rearrange fixup/squash lines using the rebase--helper\n + t3415: test fixup with wrapped oneline\n + rebase -i: skip unnecessary picks using the rebase--helper\n + rebase -i: check for missing commits in the rebase--helper\n + t3404: relax rebase.missingCommitsCheck tests\n + rebase -i: also expand/collapse the SHA-1s via the rebase--helper\n + rebase -i: do not invent onelines when expanding/collapsing SHA-1s\n + rebase -i: remove useless indentation\n + rebase -i: generate the script via rebase--helper\n + t3415: verify that an empty instructionFormat is handled as before\n\n The final batch to \"git rebase -i\" updates to move more code from\n the shell script to C.\n\n\n* mh/mmap-packed-refs (2017-09-25) 21 commits\n  (merged to 'next' on 2017-09-29 at 3639417666)\n + packed-backend.c: rename a bunch of things and update comments\n + mmapped_ref_iterator: inline into `packed_ref_iterator`\n + ref_cache: remove support for storing peeled values\n + packed_ref_store: get rid of the `ref_cache` entirely\n + ref_store: implement `refs_peel_ref()` generically\n + packed_read_raw_ref(): read the reference from the mmapped buffer\n + packed_ref_iterator_begin(): iterate using `mmapped_ref_iterator`\n + read_packed_refs(): ensure that references are ordered when read\n + packed_ref_cache: keep the `packed-refs` file mmapped if possible\n + packed-backend.c: reorder some definitions\n + mmapped_ref_iterator_advance(): no peeled value for broken refs\n + mmapped_ref_iterator: add iterator over a packed-refs file\n + packed_ref_cache: remember the file-wide peeling state\n + read_packed_refs(): read references with minimal copying\n + read_packed_refs(): make parsing of the header line more robust\n + read_packed_refs(): only check for a header at the top of the file\n + read_packed_refs(): use mmap to read the `packed-refs` file\n + die_unterminated_line(), die_invalid_line(): new functions\n + packed_ref_cache: add a backlink to the associated `packed_ref_store`\n + prefix_ref_iterator: break when we leave the prefix\n + ref_iterator: keep track of whether the iterator output is ordered\n\n Operations that do not touch (majority of) packed refs have been\n optimized by making accesses to packed-refs file lazy; we no longer\n pre-parse everything, and an access to a single ref in the\n packed-refs does not touch majority of irrelevant refs, either.\n\n\n* mr/doc-negative-pathspec (2017-09-25) 1 commit\n  (merged to 'next' on 2017-09-29 at 147bb7ef8b)\n + docs: improve discoverability of exclude pathspec\n\n Doc updates.\n\n\n* sb/diff-color-move (2017-09-28) 1 commit\n  (merged to 'next' on 2017-09-28 at cbdbc741be)\n + diff: correct newline in summary for renamed files\n\n The output from \"git diff --summary\" was broken in a recent topic\n that has been merged to 'master' and lost a LF after reporting of\n mode change.  This has been fixed.\n\n\n* sb/submodule-diff-header-fix (2017-09-28) 1 commit\n  (merged to 'next' on 2017-09-28 at 10e0f59303)\n + submodule: correct error message for missing commits\n\n Error message tweak.\n\n\n* sb/test-submodule-update-config (2017-09-27) 1 commit\n  (merged to 'next' on 2017-09-28 at 497214603c)\n + t7406: submodule.<name>.update command must not be run from .gitmodules\n\n\n* sd/branch-copy (2017-09-24) 4 commits\n  (merged to 'next' on 2017-09-28 at a6eceefa02)\n + branch: fix \"copy\" to never touch HEAD\n + branch: add a --copy (-c) option to go with --move (-m)\n + branch: add test for -m renaming multiple config sections\n + config: create a function to format section headers\n\n \"git branch\" learned \"-c/-C\" to create a new branch by copying an\n existing one.\n\n--------------------------------------------------\n[New Topics]\n\n* jc/merge-symlink-ours-theirs (2017-09-26) 1 commit\n - merge: teach -Xours/-Xtheirs to symbolic link merge\n\n \"git merge -Xours/-Xtheirs\" learned to use our/their version when\n resolving a conflicting updates to a symbolic link.\n\n Needs review.\n\n\n* jt/oidmap (2017-10-01) 1 commit\n - oidmap: map with OID as key\n\n Introduce a new \"oidmap\" API and rewrite oidset to use it.\n\n Will merge to 'next'.\n\n\n* jt/partial-clone-lazy-fetch (2017-10-02) 18 commits\n - fetch-pack: restore save_commit_buffer after use\n - unpack-trees: batch fetching of missing blobs\n - clone: configure blobmaxbytes in created repos\n - clone: support excluding large blobs\n - fetch: support excluding large blobs\n - fetch: refactor calculation of remote list\n - fetch-pack: support excluding large blobs\n - pack-objects: support --blob-max-bytes\n - pack-objects: rename want_.* to ignore_.*\n - gc: do not repack promisor packfiles\n - rev-list: support termination at promisor objects\n - sha1_file: support lazily fetching missing objects\n - introduce fetch-object: fetch one promisor object\n - index-pack: refactor writing of .keep files\n - fsck: support promisor objects as CLI argument\n - fsck: support referenced promisor objects\n - fsck: support refs pointing to promisor objects\n - fsck: introduce partialclone extension\n\n A journey for \"git clone\" and \"git fetch\" to become \"lazier\" by\n depending more on its remote repository---this is the beginning of\n it.\n\n Needs review.\n\n\n* ma/builtin-unleak (2017-10-02) 1 commit\n - builtin/: add UNLEAKs\n\n Many variables that points at a region of memory that will live\n throughout the life of the program have been marked with UNLEAK\n marker to help the leak checkers concentrate on real leaks..\n\n\n* ma/lockfile-fixes (2017-10-02) 11 commits\n - read-cache: roll back lock on error with `COMMIT_LOCK`\n - read-cache: don't leave dangling pointer in `do_write_index()`\n - read-cache: require flags for `write_locked_index()`\n - cache.h: document `write_locked_index()`\n - apply: remove `newfd` from `struct apply_state`\n - apply: move lockfile into `apply_state`\n - cache-tree: simplify locking logic\n - tempfile: fix documentation on `delete_tempfile()`\n - lockfile: fix documentation on `close_lock_file_gently()`\n - treewide: prefer lockfiles on the stack\n - sha1_file: do not leak `lock_file`\n\n An earlier update made it possible to use an on-stack in-core\n lockfile structure (as opposed to having to deliberately leak an\n on-heap one).  Many codepaths have been updated to take advantage\n of this new facility.\n\n\n* rs/cleanup-strbuf-users (2017-10-02) 3 commits\n  (merged to 'next' on 2017-10-03 at 322ec33ad8)\n + graph: use strbuf_addchars() to add spaces\n + use strbuf_addstr() for adding strings to strbufs\n + path: use strbuf_add_real_path()\n\n Code clean-up.\n\n Will merge to 'master'.\n\n\n* rs/cocci-de-paren-call-params (2017-10-02) 1 commit\n  (merged to 'next' on 2017-10-03 at ee97fe5c24)\n + coccinelle: remove parentheses that become unnecessary\n\n Code clean-up.\n\n Will merge to 'master'.\n\n\n* rs/run-command-use-alloc-array (2017-10-03) 1 commit\n  (merged to 'next' on 2017-10-03 at e3869b6c32)\n + run-command: use ALLOC_ARRAY\n\n Code clean-up.\n\n Will merge to 'master'.\n\n\n* rs/tag-null-pointer-arith-fix (2017-10-02) 1 commit\n  (merged to 'next' on 2017-10-03 at e05d92fec9)\n + tag: avoid NULL pointer arithmetic\n\n Code clean-up.\n\n Will merge to 'master'.\n\n\n* rs/use-free-and-null (2017-10-02) 1 commit\n  (merged to 'next' on 2017-10-03 at 6340234d67)\n + repository: use FREE_AND_NULL\n\n Code clean-up.\n\n Will merge to 'master'.\n\n\n* sb/git-clang-format (2017-10-02) 1 commit\n  (merged to 'next' on 2017-10-03 at b40d47b9d5)\n + clang-format: add a comment about the meaning/status of the\n\n Add comment to clarify that the style file is meant to be used with\n clang-5 and the rules are still work in progress.\n\n Will merge to 'master'.\n\n\n* rs/resolve-ref-optional-result (2017-10-01) 2 commits\n  (merged to 'next' on 2017-10-03 at f129d4b0e9)\n + refs: pass NULL to resolve_refdup() if hash is not needed\n + refs: pass NULL to refs_resolve_refdup() if hash is not needed\n\n Code clean-up.\n\n Will merge to 'master'.\n\n\n* ar/request-pull-phrasofix (2017-10-03) 1 commit\n  (merged to 'next' on 2017-10-03 at 07807bba90)\n + request-pull: capitalise \"Git\" to make it a proper noun\n\n Spell the name of our system as \"Git\" in the output from\n request-pull script.\n\n Will merge to 'master'.\n\n\n* tb/show-trailers-in-ref-filter (2017-10-02) 7 commits\n - ref-filter.c: parse trailers arguments with %(contents) atom\n - ref-filter.c: use trailer_opts to format trailers\n - t6300: refactor %(trailers) tests\n - doc: use \"`<literal>`\"-style quoting for literal strings\n - doc: 'trailers' is the preferred way to format trailers\n - t4205: unfold across multiple lines\n - pretty.c: delimit \"%(trailers)\" arguments with \",\"\n\n \"git for-each-ref --format=...\" learned a new format element,\n %(trailers), to show only the commit log trailer part of the log\n message.\n\n Will merge to 'next'.\n\n\n* jk/ui-color-always-to-auto (2017-10-04) 3 commits\n - Merge branch 'jk/ui-color-always-to-auto-maint' into jk/ui-color-always-to-auto\n - t7301: use test_terminal to check color\n - t4015: use --color with --color-moved\n (this branch uses jk/ui-color-always-to-auto-maint.)\n\n\n* jk/ui-color-always-to-auto-maint (2017-10-04) 10 commits\n - color: make \"always\" the same as \"auto\" in config\n - provide --color option for all ref-filter users\n - t3205: use --color instead of color.branch=always\n - t3203: drop \"always\" color test\n - t6006: drop \"always\" color config tests\n - t7502: use diff.noprefix for --verbose test\n - t7508: use test_terminal for color output\n - t3701: use test-terminal to collect color output\n - t4015: prefer --color to -c color.diff=always\n - test-terminal: set TERM=vt100\n (this branch is used by jk/ui-color-always-to-auto.)\n\n\n* jn/strbuf-doc-re-reuse (2017-10-04) 1 commit\n - strbuf doc: reuse after strbuf_release is fine\n\n Will merge to 'next'.\n\n\n* ks/branch-tweak-error-message-for-extra-args (2017-10-04) 1 commit\n - branch: change the error messages to be more meaningful\n\n Error message tweak.\n\n\n* ks/verify-filename-non-option-error-message-tweak (2017-10-04) 1 commit\n - setup: update error message to be more meaningful\n\n Error message tweak.\n\n\n* rs/qsort-s (2017-10-04) 1 commit\n - test-stringlist: avoid buffer underrun when sorting nothing\n\n Will merge to 'next'.\n\n\n* sb/branch-avoid-repeated-strbuf-release (2017-10-04) 1 commit\n - branch: reset instead of release a strbuf\n\n Will merge to 'next'.\n\n\n* tg/memfixes (2017-10-04) 3 commits\n - sub-process: use child_process.args instead of child_process.argv\n - http-push: fix construction of hex value from path\n - path.c: fix uninitialized memory access\n\n Fixes for a handful memory access issues identified by valgrind.\n\n Will merge to 'next'.\n\n--------------------------------------------------\n[Stalled]\n\n* mk/use-size-t-in-zlib (2017-08-10) 1 commit\n . zlib.c: use size_t for size\n\n The wrapper to call into zlib followed our long tradition to use\n \"unsigned long\" for sizes of regions in memory, which have been\n updated to use \"size_t\".\n\n Needs resurrecting by making sure the fix is good and still applies\n (or adjusted to today's codebase).\n\n\n* mg/status-in-progress-info (2017-05-10) 2 commits\n - status --short --inprogress: spell it as --in-progress\n - status: show in-progress info for short status\n\n \"git status\" learns an option to report various operations\n (e.g. \"merging\") that the user is in the middle of.\n\n cf. <xmqqmvakcdqw.fsf@gitster.mtv.corp.google.com>\n\n\n* nd/worktree-move (2017-04-20) 6 commits\n - worktree remove: new command\n - worktree move: refuse to move worktrees with submodules\n - worktree move: accept destination as directory\n - worktree move: new command\n - worktree.c: add update_worktree_location()\n - worktree.c: add validate_worktree()\n\n \"git worktree\" learned move and remove subcommands.\n\n Expecting a reroll.\n cf. <20170420101024.7593-1-pclouds@gmail.com>\n cf. <20170421145916.mknekgqzhxffu7di@sigill.intra.peff.net>\n cf. <d0e81b1e-5869-299e-f462-4d43dc997bd1@ramsayjones.plus.com>\n\n\n* sg/clone-refspec-from-command-line-config (2017-06-16) 2 commits\n - Documentation/clone: document ignored configuration variables\n - clone: respect additional configured fetch refspecs during initial fetch\n (this branch is used by sg/remote-no-string-refspecs.)\n\n \"git clone -c var=val\" is a way to set configuration variables in\n the resulting repository, but it is more useful to also make these\n variables take effect while the initial clone is happening,\n e.g. these configuration variables could be fetch refspecs.\n\n Waiting for a response.\n cf. <20170617112228.vugswym4o4owf6wj@sigill.intra.peff.net>\n cf. <xmqqmv8zhdap.fsf@gitster.mtv.corp.google.com>\n\n--------------------------------------------------\n[Cooking]\n\n* er/fast-import-dump-refs-on-checkpoint (2017-09-29) 1 commit\n  (merged to 'next' on 2017-10-03 at 4e7b0e7ec1)\n + fast-import: checkpoint: dump branches/tags/marks even if object_count==0\n\n The checkpoint command \"git fast-import\" did not flush updates to\n refs and marks unless at least one object was created since the\n last checkpoint, which has been corrected, as these things can\n happen without any new object getting created.\n\n Will merge to 'master'.\n\n\n* jr/hash-migration-plan-doc (2017-09-28) 1 commit\n - technical doc: add a design doc for hash function transition\n\n Lay out plans for weaning us off of SHA-1.\n\n Will merge to 'next'.\n\n\n* ot/mru-on-list (2017-10-01) 1 commit\n - mru: use double-linked list from list.h\n\n The first step to getting rid of mru API and using the\n doubly-linked list API directly instead.\n\n Needs review.\n Outreachy.\n\n\n* rb/compat-poll-fix (2017-09-29) 1 commit\n - poll.c: always set revents, even if to zero\n\n Backports a moral equivalent of 2015 fix to the poll emulation from\n the upstream gnulib to fix occasional breakages on HPE NonStop.\n\n Needs review.\n\n\n* jm/svn-pushmergeinfo-fix (2017-09-17) 1 commit\n - git-svn: fix svn.pushmergeinfo handling of svn+ssh usernames.\n\n \"git svn dcommit\" did not take into account the fact that a\n svn+ssh:// URL with a username@ (typically used for pushing) refers\n to the same SVN repository without the username@ and failed when\n svn.pushmergeinfo option is set.\n\n Needs a test, perhaps as a follow-up patch.\n\n\n* mg/merge-base-fork-point (2017-09-17) 3 commits\n - merge-base: find fork-point outside partial reflog\n - merge-base: return fork-point outside reflog\n - t6010: test actual test output\n\n \"merge-base --fork-point $branch $commit\" is used to guess on which\n commit among the commits that were once at the tip of the $branch the\n $commit was built on top of, and it learns these historical tips from\n the reflog of the $branch.  When the true fork-point is lost due to\n pruning of old reflog entries, the command does not give any output,\n because it has no way to guess correctly and does not want to mislead\n the user with a wrong guess.\n\n The command has been updated to give the best but not known to be\n correct guess, based on a hope that a merge-base between $commit and a\n virtual merge across all the reflog entries that still are available\n for $branch may still be a closer to the true fork-point than the\n merge-base between $commit and the current tip of the $branch.\n\n This may have to be offered by an additional option, to allow the\n users that are prepared to see a potentially incorrect guess to opt\n into the feature, without affecting the current callers that may not\n be prepared to accept a guess that is not known to be correct.\n\n What's the doneness of this one?\n\n\n* ds/find-unique-abbrev-optim (2017-10-03) 5 commits\n - sha1_name: minimize OID comparisons during disambiguation\n - sha1_name: parse less while finding common prefix\n - sha1_name: unroll len loop in find_unique_abbrev_r()\n - p0008-abbrev.sh: test find_unique_abbrev() perf\n - test-list-objects: list a subset of object ids\n\n\n* cc/perf-run-config (2017-09-24) 9 commits\n - perf: store subsection results in \"test-results/$GIT_PERF_SUBSECTION/\"\n - perf/run: show name of rev being built\n - perf/run: add run_subsection()\n - perf/run: update get_var_from_env_or_config() for subsections\n - perf/run: add get_subsections()\n - perf/run: add calls to get_var_from_env_or_config()\n - perf/run: add GIT_PERF_DIRS_OR_REVS\n - perf/run: add get_var_from_env_or_config()\n - perf/run: add '--config' option to the 'run' script\n\n Needs review.\n\n\n* mg/merge-pre-merge-hook (2017-09-24) 4 commits\n - t7503: add tests for pre-merge-hook\n - merge: --no-verify to bypass pre-merge hook\n - merge: do no-verify like commit\n - git-merge: honor pre-merge hook\n\n \"git merge\" learned a 'pre-merge' hook, that would be triggered\n before the merge commit is created by a clean auto-merge.\n\n Waiting for review to conclude.\n\n\n* bp/fsmonitor (2017-10-04) 13 commits\n - fsmonitor: MINGW support for watchman integration\n  (merged to 'next' on 2017-10-02 at cf0c67979c)\n + fsmonitor: add a performance test\n + fsmonitor: add a sample integration script for Watchman\n + fsmonitor: add test cases for fsmonitor extension\n + split-index: disable the fsmonitor extension when running the split index test\n + fsmonitor: add a test tool to dump the index extension\n + update-index: add fsmonitor support to update-index\n + ls-files: Add support in ls-files to display the fsmonitor valid bit\n + fsmonitor: add documentation for the fsmonitor extension.\n + fsmonitor: teach git to optionally utilize a file system monitor to speed up detecting new or changed files.\n + update-index: add a new --force-write-index option\n + preload-index: add override to enable testing preload-index\n + bswap: add 64 bit endianness helper get_be64\n\n We learned to talk to watchman to speed up \"git status\" and other\n operations that need to see which paths have been modified.\n\n\n* bw/protocol-v1 (2017-10-04) 10 commits\n - ssh: introduce a 'simple' ssh variant\n - i5700: add interop test for protocol transition\n - http: tell server that the client understands v1\n - connect: tell server that the client understands v1\n - connect: teach client to recognize v1 server response\n - upload-pack, receive-pack: introduce protocol version 1\n - daemon: recognize hidden request arguments\n - protocol: introduce protocol extention mechanisms\n - pkt-line: add packet_write function\n - connect: in ref advertisement, shallows are last\n\n A new mechanism to upgrade the wire protocol in place is proposed\n and demonstrated that it works with the older versions of Git\n without harming them.\n\n\n* pc/submodule-helper (2017-10-02) 3 commits\n - submodule: port submodule subcommand 'status' from shell to C\n - submodule--helper: introduce for_each_listed_submodule()\n - submodule--helper: introduce get_submodule_displaypath()\n\n GSoC.\n Is this ready for 'next'?\n\n\n* bc/hash-algo (2017-08-20) 5 commits\n - hash-algo: switch empty tree and blob lookups to use hash abstraction\n - hash-algo: integrate hash algorithm support with repo setup\n - hash-algo: add structure representing hash algorithm\n - setup: expose enumerated repo info\n - Merge branch 'bc/vcs-svn-cleanup' into bc/hash-algo\n\n RFC\n cf. <20170821000022.26729-1-sandals@crustytoothpaste.net>\n\n\n* jk/drop-ancient-curl (2017-08-09) 5 commits\n - http: #error on too-old curl\n - curl: remove ifdef'd code never used with curl >=7.19.4\n - http: drop support for curl < 7.19.4\n - http: drop support for curl < 7.16.0\n - http: drop support for curl < 7.11.1\n\n Some code in http.c that has bitrot is being removed.\n\n Will be rerolled after tc/curl-with-backports stabilizes.\n\n\n* sb/submodule-recursive-checkout-detach-head (2017-07-28) 2 commits\n - Documentation/checkout: clarify submodule HEADs to be detached\n - recursive submodules: detach HEAD from new state\n\n \"git checkout --recursive\" may overwrite and rewind the history of\n the branch that happens to be checked out in submodule\n repositories, which might not be desirable.  Detach the HEAD but\n still allow the recursive checkout to succeed in such a case.\n\n Undecided.\n This needs justification in a larger picture; it is unclear why\n this is better than rejecting recursive checkout, for example.\n\n\n* jk/check-ref-format-oor-fix (2017-07-14) 1 commit\n - check-ref-format: require a repository for --branch\n\n Discussion slowly continues...\n cf. <20170717172709.GL93855@aiede.mtv.corp.google.com>\n cf. <20170817102217.3yw7uxnkupdy3lh5@sigill.intra.peff.net>\n\n\n* rs/sha1-file-micro-optim (2017-07-09) 2 commits\n - SQUASH???\n - sha1_file: add slash once in for_each_file_in_obj_subdir()\n\n Code cleanup.\n\n Perhaps drop.\n cf. <f59c8256-716b-9305-2a4f-d4fe49f666ff@web.de>\n\n\n* jc/allow-lazy-cas (2017-07-06) 1 commit\n - push: disable lazy --force-with-lease by default\n\n Because \"git push --force-with-lease[=<ref>]\" that relies on the\n stability of remote-tracking branches is unsafe when something\n fetches into the repository behind user's back, it is now disabled\n by default.  A new configuration variable can be used to enable it\n by users who know what they are doing.  This would pave the way to\n possibly turn `--force` into `--force-with-lease`.\n\n Will discard.\n Independent from disabling the feature by default, this stirred up\n a discussion to replace the DWIM heuristics with a better one, which\n deserves to be its own topic.\n cf. <alpine.DEB.2.21.1.1707131435220.4193@virtualbox>\n\n\n* ex/deprecate-empty-pathspec-as-match-all (2017-06-23) 2 commits\n  (merged to 'next' on 2017-08-14 at 24db08a6e8)\n + pathspec: die on empty strings as pathspec\n + t0027: do not use an empty string as a pathspec element\n\n The final step to make an empty string as a pathspec element\n illegal.  We started this by first deprecating and warning a\n pathspec that has such an element in 2.11 (Nov 2016).\n\n Hopefully we can merge this down to the 'master' by the end of the\n year?  A deprecation warning period that is about 1 year does not\n sound too bad.\n\n Will cook in 'next'.\n As the draft RelNotes says, let's merge this to 'master' after\n this release cycle.\n\n\n--------------------------------------------------\n[Discarded]\n\n* jc/ref-filter-colors-fix (2017-10-03) 2 commits\n . colors: git_default_config() does not read color.ui\n . Revert \"color: check color.ui in git_default_config()\"\n\n Recent tip of 'master' and 'maint' (at least as of v2.14.2) has a\n regression in \"git add -p\" for those who set ui.color=always in\n their configuration, caused by us being overly aggressive in\n allowing plumbing commands to honor the configuration to fix a\n fallout from an earlier change that made everybody, even the\n plumbing commands, to automatically color its output by default.\n Revert it for now from the released version (and to be released\n one) until we come up with a different fix to the problem.\n\n\n"},{"id":"329663","messageId":"20171004080212.jatd4clbjlctazun@sigill.intra.peff.net","threadId":"46898","inReplyTo":"xmqqy3orbcrr.fsf@gitster.mtv.corp.google.com","subject":"Re: What's cooking in git.git (Oct 2017, #01; Wed, 4)","fromName":"Jeff King","fromEmail":"peff@peff.net","sentAt":"2017-10-04T08:02:13Z","receivedAt":"2017-10-04T08:02:21Z","isPatch":false,"sender":{"key":"peff@peff.net","avatar":"https://avatars.githubusercontent.com/u/45925?v=4"},"body":"On Wed, Oct 04, 2017 at 04:19:52PM +0900, Junio C Hamano wrote:\n\n> * jt/partial-clone-lazy-fetch (2017-10-02) 18 commits\n>  - fetch-pack: restore save_commit_buffer after use\n>  - unpack-trees: batch fetching of missing blobs\n>  - clone: configure blobmaxbytes in created repos\n>  - clone: support excluding large blobs\n>  - fetch: support excluding large blobs\n>  - fetch: refactor calculation of remote list\n>  - fetch-pack: support excluding large blobs\n>  - pack-objects: support --blob-max-bytes\n>  - pack-objects: rename want_.* to ignore_.*\n>  - gc: do not repack promisor packfiles\n>  - rev-list: support termination at promisor objects\n>  - sha1_file: support lazily fetching missing objects\n>  - introduce fetch-object: fetch one promisor object\n>  - index-pack: refactor writing of .keep files\n>  - fsck: support promisor objects as CLI argument\n>  - fsck: support referenced promisor objects\n>  - fsck: support refs pointing to promisor objects\n>  - fsck: introduce partialclone extension\n> \n>  A journey for \"git clone\" and \"git fetch\" to become \"lazier\" by\n>  depending more on its remote repository---this is the beginning of\n>  it.\n> \n>  Needs review.\n\nThe merge of this topic into jch (at 766f92478b0) causes the test suite\nto fail when compiled with ASan/UBSan. The simplest reproduction I could\ncome up with is:\n\n   $ make SANITIZE=address,undefined BLK_SHA1=1 &&\n     GIT_DIR=nope ./git shortlog </dev/null >/dev/null\n   repository.c:69:31: runtime error: index 1869098813 out of bounds for type 'git_hash_algo [1]'\n\nNote that the series is fine by itself, it's only the merge which fails.\nWhich implies to me it's some funny interaction with bc/hash-algo (which\nintroduces the hash_algo concept). But I didn't dig further.\n\n+cc brian and Jonathan.\n\n-Peff\n"},{"id":"329665","messageId":"20171004082433.tltvqsw3ov47f3lz@sigill.intra.peff.net","threadId":"46898","inReplyTo":"20171004080212.jatd4clbjlctazun@sigill.intra.peff.net","subject":"Re: What's cooking in git.git (Oct 2017, #01; Wed, 4)","fromName":"Jeff King","fromEmail":"peff@peff.net","sentAt":"2017-10-04T08:24:33Z","receivedAt":"2017-10-04T08:24:43Z","isPatch":false,"sender":{"key":"peff@peff.net","avatar":"https://avatars.githubusercontent.com/u/45925?v=4"},"body":"On Wed, Oct 04, 2017 at 04:02:12AM -0400, Jeff King wrote:\n\n> On Wed, Oct 04, 2017 at 04:19:52PM +0900, Junio C Hamano wrote:\n> \n> > * jt/partial-clone-lazy-fetch (2017-10-02) 18 commits\n> [...]\n> \n> The merge of this topic into jch (at 766f92478b0) causes the test suite\n> to fail when compiled with ASan/UBSan. The simplest reproduction I could\n> come up with is:\n> \n>    $ make SANITIZE=address,undefined BLK_SHA1=1 &&\n>      GIT_DIR=nope ./git shortlog </dev/null >/dev/null\n>    repository.c:69:31: runtime error: index 1869098813 out of bounds for type 'git_hash_algo [1]'\n> \n> Note that the series is fine by itself, it's only the merge which fails.\n> Which implies to me it's some funny interaction with bc/hash-algo (which\n> introduces the hash_algo concept). But I didn't dig further.\n> \n> +cc brian and Jonathan.\n\nActually, I take it back. Bisecting between \"master\" and \"pu\" turned up\nthat commit (and I verified that it fails at that commit but not at its\nfirst parent).\n\nBut if I go directly to the tip of bc/hash-algo, I can see the failure\nthere. And it bisects to 67a9dfcc00 (hash-algo: integrate hash algorithm\nsupport with repo setup, 2017-08-21).\n\nSo I think it's a read of uninitialized data, and it may come and go as\nvarious topics tickle the compiler differently. And that would make\njt/partial-clone-lazy-fetch a red herring.\n\nThis seems to make it go away (on top of 67a9dfcc00):\n\ndiff --git a/setup.c b/setup.c\nindex 289e24811c..310afe9736 100644\n--- a/setup.c\n+++ b/setup.c\n@@ -1126,7 +1126,9 @@ const char *setup_git_directory_gently(int *nongit_ok)\n \t\t\trepo_set_gitdir(the_repository, gitdir);\n \t\t\tsetup_git_env();\n \t\t}\n-\t\trepo_set_hash_algo(the_repository, repo_fmt.hash_algo);\n+\t\t/* This is only valid if we actually found a repository */\n+\t\tif (startup_info->have_repository)\n+\t\t\trepo_set_hash_algo(the_repository, repo_fmt.hash_algo);\n \t}\n \n \tstrbuf_release(&dir);\n\nThe problem is that we may not have a real repository at all, but we\nstill hit this code path, which tries to setup repository properties, if\nthere's a $GIT_DIR set.\n\nI actually suspect this repo_set_hash_algo() line should go into\ncheck_repository_format_gently(), where we set other globals (and which\nshould eventually take a pointer to struct repository rather than\ntouching the_repository).\n\nBut I'm sufficiently convinced now that this is just a bug in the\nRFC-marked bc/hash-algo topic, and in no danger of graduating to master\nsoon.\n\n-Peff\n"},{"id":"329666","messageId":"20171004083111.5y7su63scydg2yzk@sigill.intra.peff.net","threadId":"46898","inReplyTo":"xmqqy3orbcrr.fsf@gitster.mtv.corp.google.com","subject":"Re: What's cooking in git.git (Oct 2017, #01; Wed, 4)","fromName":"Jeff King","fromEmail":"peff@peff.net","sentAt":"2017-10-04T08:31:11Z","receivedAt":"2017-10-04T08:31:19Z","isPatch":false,"sender":{"key":"peff@peff.net","avatar":"https://avatars.githubusercontent.com/u/45925?v=4"},"body":"On Wed, Oct 04, 2017 at 04:19:52PM +0900, Junio C Hamano wrote:\n\n> I wanted to do v2.15-rc0 today but it seems that this has to slip,\n> with too many topics still in flight in 'next' among which many are\n> fixes, one is even a fix for a recent regression.\n\nYou probably already figured this out, but I want to be explicit just in\ncase:\n\n> * tb/show-trailers-in-ref-filter (2017-10-02) 7 commits\n>  - ref-filter.c: parse trailers arguments with %(contents) atom\n>  - ref-filter.c: use trailer_opts to format trailers\n>  - t6300: refactor %(trailers) tests\n>  - doc: use \"`<literal>`\"-style quoting for literal strings\n>  - doc: 'trailers' is the preferred way to format trailers\n>  - t4205: unfold across multiple lines\n>  - pretty.c: delimit \"%(trailers)\" arguments with \",\"\n> \n>  \"git for-each-ref --format=...\" learned a new format element,\n>  %(trailers), to show only the commit log trailer part of the log\n>  message.\n> \n>  Will merge to 'next'.\n\nI think we want the first patch of this series to graduate before v2.15,\neven if the rest doesn't make it. It tweaks a new syntax introduced\nearlier in this cycle by jk/trailers-parse. If we ship without the\ntweak, then we'll have to support the colon-delimiter to remain\nbackwards-compatible.\n\n-Peff\n"},{"id":"329668","messageId":"xmqqr2ujb8rm.fsf@gitster.mtv.corp.google.com","threadId":"46898","inReplyTo":"20171004083111.5y7su63scydg2yzk@sigill.intra.peff.net","subject":"Re: What's cooking in git.git (Oct 2017, #01; Wed, 4)","fromName":"Junio C Hamano","fromEmail":"gitster@pobox.com","sentAt":"2017-10-04T08:46:21Z","receivedAt":"2017-10-04T08:46:30Z","isPatch":false,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"Jeff King <peff@peff.net> writes:\n\n>>  - pretty.c: delimit \"%(trailers)\" arguments with \",\"\n>> \n>>  \"git for-each-ref --format=...\" learned a new format element,\n>>  %(trailers), to show only the commit log trailer part of the log\n>>  message.\n>> \n>>  Will merge to 'next'.\n>\n> I think we want the first patch of this series to graduate before v2.15,\n> even if the rest doesn't make it. It tweaks a new syntax introduced\n> earlier in this cycle by jk/trailers-parse. If we ship without the\n> tweak, then we'll have to support the colon-delimiter to remain\n> backwards-compatible.\n\nYeah, thanks for reminding me.  I actually was hoping that this will\nprove to be stable enough by the time -rc1 gets tagged, but yes, the\nbottom one looks innocuous/safe enough and should be fast-tracked to\n'master' soonish.\n\nThanks.\n"},{"id":"329701","messageId":"20171004180737.GA90224@D-173-250-188-46.dhcp4.washington.edu","threadId":"46898","inReplyTo":"xmqqr2ujb8rm.fsf@gitster.mtv.corp.google.com","subject":"Re: What's cooking in git.git (Oct 2017, #01; Wed, 4)","fromName":"Taylor Blau","fromEmail":"me@ttaylorr.com","sentAt":"2017-10-04T18:07:37Z","receivedAt":"2017-10-04T18:07:46Z","isPatch":false,"sender":{"key":"me@ttaylorr.com","avatar":"https://avatars.githubusercontent.com/u/301000140?v=4"},"body":"On Wed, Oct 04, 2017 at 05:46:21PM +0900, Junio C Hamano wrote:\n> Jeff King <peff@peff.net> writes:\n>\n> >>  - pretty.c: delimit \"%(trailers)\" arguments with \",\"\n> >>\n> >>  \"git for-each-ref --format=...\" learned a new format element,\n> >>  %(trailers), to show only the commit log trailer part of the log\n> >>  message.\n> >>\n> >>  Will merge to 'next'.\n> >\n> > I think we want the first patch of this series to graduate before v2.15,\n> > even if the rest doesn't make it. It tweaks a new syntax introduced\n> > earlier in this cycle by jk/trailers-parse. If we ship without the\n> > tweak, then we'll have to support the colon-delimiter to remain\n> > backwards-compatible.\n>\n> Yeah, thanks for reminding me.  I actually was hoping that this will\n> prove to be stable enough by the time -rc1 gets tagged, but yes, the\n> bottom one looks innocuous/safe enough and should be fast-tracked to\n> 'master' soonish.\n\nIt may make sense to send my other series to 'master' as well\n(\"ref-filter.c: pass empty-string as NULL to atom parsers\").\n\nThe series you're discussing here adds support for \"empty\" sub-agruments\n(via: --format=%(contents:trailers:)), but Peff points out that this is\nnot a consistent user experience:\n\n> Doh, that string_list behavior is what I was missing in my earlier\n> comments. I agree this is probably the best way of doing it. I'm tempted\n> to say that parse_ref_filter_atom() should do a similar thing. Right now\n> we've got:\n>\n>   $ git for-each-ref --format='%(refname)' | wc\n>      2206    2206   79929\n>   $ git for-each-ref --format='%(refname:short)' | wc\n>      2206    2206   53622\n>   $ git for-each-ref --format='%(refname:)' | wc\n>   fatal: unrecognized %(refname:) argument:\n>       0       0       0\n\n\"ref-filter.c: pass empty-string as NULL to atom parsers\" makes this\nbehavior of allowing empty sub-argument atom formats in\ngit-for-each-ref(1) consistently OK.\n\nTo avoid introducing a case where %(atom:) is sometimes allowed and\nsometimes not, I would recommend that both of these patches be applied\nto master at the same time.\n\n--\n- Taylor\n"},{"id":"329723","messageId":"xmqqd1628918.fsf@gitster.mtv.corp.google.com","threadId":"46898","inReplyTo":"20171004180737.GA90224@D-173-250-188-46.dhcp4.washington.edu","subject":"Re: What's cooking in git.git (Oct 2017, #01; Wed, 4)","fromName":"Junio C Hamano","fromEmail":"gitster@pobox.com","sentAt":"2017-10-05T05:21:07Z","receivedAt":"2017-10-05T05:21:13Z","isPatch":false,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"Taylor Blau <me@ttaylorr.com> writes:\n\n> It may make sense to send my other series to 'master' as well\n> (\"ref-filter.c: pass empty-string as NULL to atom parsers\").\n\nThanks for reminding.  I was involved in the review and remember\nthat everybody was happy with the direction.  The topic was left on\nthe list without getting picked up by a mere accident.\n\nNow it is queued.  Let's make sure nobody is depending on the\n%(atom) vs %(atom:) distinction by merging it to 'next' soonish.\n\nThose who don't raise stink while the change is in 'next' but only\ndo so after it hits a release won't be heard ;-)\n"}]}