summaryrefslogtreecommitdiff
path: root/sequencer.c
AgeCommit message (Collapse)AuthorFilesLines
2019-01-04Merge branch 'nd/the-index'Libravatar Junio C Hamano1-203/+250
More codepaths become aware of working with in-core repository instance other than the default "the_repository". * nd/the-index: (22 commits) rebase-interactive.c: remove the_repository references rerere.c: remove the_repository references pack-*.c: remove the_repository references pack-check.c: remove the_repository references notes-cache.c: remove the_repository references line-log.c: remove the_repository reference diff-lib.c: remove the_repository references delta-islands.c: remove the_repository references cache-tree.c: remove the_repository references bundle.c: remove the_repository references branch.c: remove the_repository reference bisect.c: remove the_repository reference blame.c: remove implicit dependency the_repository sequencer.c: remove implicit dependency on the_repository sequencer.c: remove implicit dependency on the_index transport.c: remove implicit dependency on the_index notes-merge.c: remove implicit dependency the_repository notes-merge.c: remove implicit dependency on the_index list-objects.c: reduce the_repository references list-objects-filter.c: remove implicit dependency on the_index ...
2018-11-18Merge branch 'js/rebase-r-and-merge-head'Libravatar Junio C Hamano1-4/+6
Bugfix for the recently graduated "git rebase --rebase-merges". * js/rebase-r-and-merge-head: status: rebase and merge can be in progress at the same time built-in rebase --skip/--abort: clean up stale .git/<name> files rebase -i: include MERGE_HEAD into files to clean up rebase -r: do not write MERGE_HEAD unless needed rebase -r: demonstrate bug with conflicting merges
2018-11-13Merge branch 'bp/refresh-index-using-preload'Libravatar Junio C Hamano1-1/+1
The helper function to refresh the cached stat information in the in-core index has learned to perform the lstat() part of the operation in parallel on multi-core platforms. * bp/refresh-index-using-preload: refresh_index: remove unnecessary calls to preload_index() speed up refresh_index() by utilizing preload_index()
2018-11-13Merge branch 'ag/rebase-i-in-c'Libravatar Junio C Hamano1-1/+1
Code clean-up for a topic already in 'master'. * ag/rebase-i-in-c: sequencer.c: remove a stray semicolon
2018-11-13Merge branch 'pw/am-rebase-read-author-script'Libravatar Junio C Hamano1-71/+121
Unify code to read the author-script used in "git am" and the commands that use the sequencer machinery, e.g. "git rebase -i". * pw/am-rebase-read-author-script: sequencer: use read_author_script() add read_author_script() to libgit am: rename read_author_script() am: improve author-script error reporting am: don't die in read_author_script()
2018-11-13Merge branch 'ma/sequencer-do-reset-saner-loop-termination'Libravatar Junio C Hamano1-2/+5
Code readability fix. * ma/sequencer-do-reset-saner-loop-termination: sequencer: break out of loop explicitly
2018-11-13rebase -i: include MERGE_HEAD into files to clean upLibravatar Johannes Schindelin1-0/+2
Every once in a while, the interactive rebase makes sure that no stale files are lying around. These days, we need to include MERGE_HEAD into that set of files, as the `merge` command will generate them. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-11-13rebase -r: do not write MERGE_HEAD unless neededLibravatar Johannes Schindelin1-4/+4
When we detect that a `merge` can be skipped because the merged commit is already an ancestor of HEAD, we do not need to commit, therefore writing the MERGE_HEAD file is useless. It is actually worse than useless: a subsequent `git commit` will pick it up and think that we want to merge that commit, still. To avoid that, move the code that writes the MERGE_HEAD file to a location where we already know that the `merge` cannot be skipped. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-11-12diff-lib.c: remove the_repository referencesLibravatar Nguyễn Thái Ngọc Duy1-2/+2
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-11-12cache-tree.c: remove the_repository referencesLibravatar Nguyễn Thái Ngọc Duy1-1/+1
This case is more interesting than other boring "remove the_repo" commits because while we need access to the object database, we cannot simply use r->index because unpack-trees.c can operate on a temporary index, not $GIT_DIR/index. Ideally we should be able to pass an object database to lookup_tree() but that ship has sailed. Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-11-12sequencer.c: remove implicit dependency on the_repositoryLibravatar Nguyễn Thái Ngọc Duy1-63/+71
Note that the_hash_algo stays, even if we can easily replace it with repo->hash_algo. My reason is I still believe tying hash_algo to a struct repository is a wrong move. But if I'm wrong, we can always go for another round of conversion. Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-11-12sequencer.c: remove implicit dependency on the_indexLibravatar Nguyễn Thái Ngọc Duy1-141/+180
Since we're going to pass 'struct repository *' around most of the time instead of 'struct index_state *' because most sequencer.c operations need more than just the index, the_repository is replaced as well in the functions that now take 'struct repository *'. the_repository is still present in this file, but total clean up will be done later. It's not the main focus of this patch. Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-11-12wt-status.c: remove implicit dependency on the_indexLibravatar Nguyễn Thái Ngọc Duy1-4/+4
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-11-06refresh_index: remove unnecessary calls to preload_index()Libravatar Ben Peart1-1/+1
With refresh_index() learning to utilize preload_index() to speed up its operation there is no longer any benefit to having the caller preload the index first. Remove those unneeded calls by calling read_index() instead of the preload variant. There is no measurable performance impact of this patch - the 2nd call to preload_index() bails out quickly but there is no reason to call it twice. Signed-off-by: Ben Peart <benpeart@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-11-05sequencer.c: remove a stray semicolonLibravatar Nguyễn Thái Ngọc Duy1-1/+1
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-11-02Merge branch 'js/rebase-i-shortopt'Libravatar Junio C Hamano1-1/+2
"git rebase -i" learned to take 'b' as the short form of 'break' option in the todo list. * js/rebase-i-shortopt: rebase -i: recognize short commands without arguments
2018-11-02Merge branch 'js/rebase-i-break'Libravatar Junio C Hamano1-1/+24
"git rebase -i" learned a new insn, 'break', that the user can insert in the to-do list. Upon hitting it, the command returns control back to the user. * js/rebase-i-break: rebase -i: introduce the 'break' command rebase -i: clarify what happens on a failed `exec`
2018-11-02Merge branch 'cb/printf-empty-format'Libravatar Junio C Hamano1-1/+1
Build fix for a topic in flight. * cb/printf-empty-format: sequencer: cleanup for gcc warning in non developer mode
2018-11-02Merge branch 'ag/rebase-i-in-c'Libravatar Junio C Hamano1-59/+261
Rewrite of the remaining "rebase -i" machinery in C. * ag/rebase-i-in-c: rebase -i: move rebase--helper modes to rebase--interactive rebase -i: remove git-rebase--interactive.sh rebase--interactive2: rewrite the submodes of interactive rebase in C rebase -i: implement the main part of interactive rebase as a builtin rebase -i: rewrite init_basic_state() in C rebase -i: rewrite write_basic_state() in C rebase -i: rewrite the rest of init_revisions_and_shortrevisions() in C rebase -i: implement the logic to initialize $revisions in C rebase -i: remove unused modes and functions rebase -i: rewrite complete_action() in C t3404: todo list with commented-out commands only aborts sequencer: change the way skip_unnecessary_picks() returns its result sequencer: refactor append_todo_help() to write its message to a buffer rebase -i: rewrite checkout_onto() in C rebase -i: rewrite setup_reflog_action() in C sequencer: add a new function to silence a command, except if it fails rebase -i: rewrite the edit-todo functionality in C editor: add a function to launch the sequence editor rebase -i: rewrite append_todo_help() in C sequencer: make three functions and an enum from sequencer.c public
2018-11-01sequencer: use read_author_script()Libravatar Phillip Wood1-76/+21
Use the new function added in the last commit to read the author script, updating read_env_script() and read_author_ident(). We now have a single code path that reads the author script for am and all flavors of rebase. This changes the behavior of read_env_script() as previously it would set any environment variables that were in the author-script file. Now it is an error if the file contains other variables or any of GIT_AUTHOR_NAME, GIT_AUTHOR_EMAIL and GIT_AUTHOR_DATE are missing. This is what am and the non interactive version of rebase have been doing for several years so hopefully it will not cause a problem for interactive rebase users. The advantage is that we are reusing existing code from am which uses sq_dequote() to properly dequote variables. This fixes potential problems with user edited scripts as read_env_script() which did not track quotes properly. This commit also removes the fallback code for checking for a broken author script after git is upgraded when a rebase is stopped. Now that the parsing uses sq_dequote() it will reliably return an error if the quoting is broken and the user will have to abort the rebase and restart. This isn't ideal but it's a corner case and the detection of the broken quoting could be confused by user edited author scripts. Signed-off-by: Phillip Wood <phillip.wood@dunelm.org.uk> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-11-01add read_author_script() to libgitLibravatar Phillip Wood1-0/+105
Add read_author_script() to sequencer.c based on the implementation in builtin/am.c and update read_am_author_script() to use read_author_script(). The sequencer code that reads the author script will be updated in the next commit. Signed-off-by: Phillip Wood <phillip.wood@dunelm.org.uk> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-10-31sequencer: break out of loop explicitlyLibravatar Martin Ågren1-2/+5
It came up in review [1, 2] that this non-idiomatic loop is a bit tricky. When we find a space, we set `len = i`, which gives us the answer we are looking for, but which also breaks out of the loop. It turns out that this loop can confuse compilers as well. My copy of gcc 7.3.0 realizes that we are essentially evaluating `(len + 1) < len` and warns that the behavior is undefined if `len` is `INT_MAX`. (Because the assignment `len = i` is guaranteed to decrease `len`, such undefined behavior is not actually possible.) Rewrite the loop to a more idiomatic variant which doesn't muck with `len` in the loop body. That should help compilers and human readers figure out what is going on here. But do note that we need to update `len` since it is not only used just after this loop (where we could have used `i` directly), but also later in this function. While at it, reduce the scope of `i`. [1] https://public-inbox.org/git/CAPig+cQbG2s-LrAo9+7C7=dXifbWFJ3SzuNa-QePHDk7egK=jg@mail.gmail.com/ [2] https://public-inbox.org/git/CAPig+cRjU6niXpT2FrDWZ0x1HmGf1ojVZj3uk2qXEGe-S7i_HQ@mail.gmail.com/ Helped-by: Eric Sunshine <sunshine@sunshineco.com> Signed-off-by: Martin Ågren <martin.agren@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-10-27sequencer: cleanup for gcc warning in non developer modeLibravatar Carlo Marcelo Arenas Belón1-1/+1
as shown by: sequencer.c: In function ‘write_basic_state’: sequencer.c:2392:37: warning: zero-length gnu_printf format string [-Wformat-zero-length] write_file(rebase_path_verbose(), ""); where write_file will create an empty file if told to write an empty string as can be inferred by the previous call the somehow more convoluted syntax works around the issue by providing a non empty format string and is already being used for the abort safety file since 1e41229d96 ("sequencer: make sequencer abort safer", 2016-12-07) Signed-off-by: Carlo Marcelo Arenas Belón <carenas@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-10-26rebase -i: recognize short commands without argumentsLibravatar Johannes Sixt1-1/+2
The sequencer instruction 'b', short for 'break', is rejected: error: invalid line 2: b The reason is that the parser expects all short commands to have an argument. Permit short commands without arguments. Signed-off-by: Johannes Sixt <j6t@kdbg.org> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-10-19Merge branch 'nd/status-refresh-progress'Libravatar Junio C Hamano1-1/+1
"git status" learns to show progress bar when refreshing the index takes a long time. * nd/status-refresh-progress: status: show progress bar if refreshing the index takes too long
2018-10-19Merge branch 'nd/the-index'Libravatar Junio C Hamano1-9/+9
Various codepaths in the core-ish part learn to work on an arbitrary in-core index structure, not necessarily the default instance "the_index". * nd/the-index: (23 commits) revision.c: reduce implicit dependency the_repository revision.c: remove implicit dependency on the_index ws.c: remove implicit dependency on the_index tree-diff.c: remove implicit dependency on the_index submodule.c: remove implicit dependency on the_index line-range.c: remove implicit dependency on the_index userdiff.c: remove implicit dependency on the_index rerere.c: remove implicit dependency on the_index sha1-file.c: remove implicit dependency on the_index patch-ids.c: remove implicit dependency on the_index merge.c: remove implicit dependency on the_index merge-blobs.c: remove implicit dependency on the_index ll-merge.c: remove implicit dependency on the_index diff-lib.c: remove implicit dependency on the_index read-cache.c: remove implicit dependency on the_index diff.c: remove implicit dependency on the_index grep.c: remove implicit dependency on the_index diff.c: remove the_index dependency in textconv() functions blame.c: rename "repo" argument to "r" combine-diff.c: remove implicit dependency on the_index ...
2018-10-16Merge branch 'rs/sequencer-oidset-insert-avoids-dups'Libravatar Junio C Hamano1-3/+1
Code clean-up. * rs/sequencer-oidset-insert-avoids-dups: sequencer: use return value of oidset_insert()
2018-10-12rebase -i: introduce the 'break' commandLibravatar Johannes Schindelin1-1/+23
The 'edit' command can be used to cherry-pick a commit and then immediately drop out of the interactive rebase, with exit code 0, to let the user amend the commit, or test it, or look around. Sometimes this functionality would come in handy *without* cherry-picking a commit, e.g. to interrupt the interactive rebase even before cherry-picking a commit, or immediately after an 'exec' or a 'merge'. This commit introduces that functionality, as the spanking new 'break' command. Suggested-by: Stefan Beller <sbeller@google.com> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-10-03sequencer: use return value of oidset_insert()Libravatar René Scharfe1-3/+1
oidset_insert() returns 1 if the object ID is already in the set and doesn't add it again, or 0 if it hadn't been present. Make use of that fact instead of checking with an extra oidset_contains() call. Signed-off-by: Rene Scharfe <l.s.r@web.de> Acked-by: Johannes Schindelin <johannes.schindelin@gmx.de> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-09-24Merge branch 'en/sequencer-empty-edit-result-aborts'Libravatar Junio C Hamano1-2/+2
"git rebase" etc. in Git 2.19 fails to abort when given an empty commit log message as result of editing, which has been corrected. * en/sequencer-empty-edit-result-aborts: sequencer: fix --allow-empty-message behavior, make it smarter
2018-09-24Merge branch 'js/rebase-i-autosquash-fix'Libravatar Junio C Hamano1-3/+14
"git rebase -i" did not clear the state files correctly when a run of "squash/fixup" is aborted and then the user manually amended the commit instead, which has been corrected. * js/rebase-i-autosquash-fix: rebase -i: be careful to wrap up fixup/squash chains rebase -i --autosquash: demonstrate a problem skipping the last squash
2018-09-21revision.c: remove implicit dependency on the_indexLibravatar Nguyễn Thái Ngọc Duy1-4/+4
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-09-21rerere.c: remove implicit dependency on the_indexLibravatar Nguyễn Thái Ngọc Duy1-2/+2
The reason rerere(), rerere_forget() and rerere_remaining() take a struct repository instead of struct index_state is not obvious from the patch: Deep in update_paths() and find_conflict(), hold_locked_index() and read_index() are called. These functions assumes the index path at $GIT_DIR/index which is not always true when you take an arbitrary index state. Taking a repository will allow us to point to the right index path later when we replace them with repo_ versions. Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-09-21merge.c: remove implicit dependency on the_indexLibravatar Nguyễn Thái Ngọc Duy1-3/+3
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-09-17Merge branch 'jk/cocci'Libravatar Junio C Hamano1-20/+20
spatch transformation to replace boolean uses of !hashcmp() to newly introduced oideq() is added, and applied, to regain performance lost due to support of multiple hash algorithms. * jk/cocci: show_dirstat: simplify same-content check read-cache: use oideq() in ce_compare functions convert hashmap comparison functions to oideq() convert "hashcmp() != 0" to "!hasheq()" convert "oidcmp() != 0" to "!oideq()" convert "hashcmp() == 0" to hasheq() convert "oidcmp() == 0" to oideq() introduce hasheq() and oideq() coccinelle: use <...> for function exclusion
2018-09-17Merge branch 'jk/trailer-fixes'Libravatar Junio C Hamano1-4/+7
"git interpret-trailers" and its underlying machinery had a buggy code that attempted to ignore patch text after commit log message, which triggered in various codepaths that will always get the log message alone and never get such an input. * jk/trailer-fixes: append_signoff: use size_t for string offsets sequencer: ignore "---" divider when parsing trailers pretty, ref-filter: format %(trailers) with no_divider option interpret-trailers: allow suppressing "---" divider interpret-trailers: tighten check for "---" patch boundary trailer: pass process_trailer_opts to trailer_info_get() trailer: use size_t for iterating trailer list trailer: use size_t for string offsets
2018-09-17Merge branch 'ds/reachable'Libravatar Junio C Hamano1-0/+1
The code for computing history reachability has been shuffled, obtained a bunch of new tests to cover them, and then being improved. * ds/reachable: commit-reach: correct accidental #include of C file commit-reach: use can_all_from_reach commit-reach: make can_all_from_reach... linear commit-reach: replace ref_newer logic test-reach: test commit_contains test-reach: test can_all_from_reach_with_flags test-reach: test reduce_heads test-reach: test get_merge_bases_many test-reach: test is_descendant_of test-reach: test in_merge_bases test-reach: create new test tool for ref_newer commit-reach: move can_all_from_reach_with_flags upload-pack: generalize commit date cutoff upload-pack: refactor ok_to_give_up() upload-pack: make reachable() more generic commit-reach: move commit_contains from ref-filter commit-reach: move ref_newer from remote.c commit.h: remove method declarations commit-reach: move walk methods from commit.c
2018-09-17status: show progress bar if refreshing the index takes too longLibravatar Nguyễn Thái Ngọc Duy1-1/+1
Refreshing the index is usually very fast, but it can still take a long time sometimes. Cold cache is one. Or copying a repo to a new place (*). It's good to show something to let the user know "git status" is not hanging, it's just busy doing something. (*) In this case, all stat info in the index becomes invalid and git falls back to rehashing all file content to see if there's any difference between updating stat info in the index. This is quite expensive. Even with a repo as small as git.git, it takes 3 seconds. Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-09-13sequencer: fix --allow-empty-message behavior, make it smarterLibravatar Elijah Newren1-2/+2
In commit b00bf1c9a8dd ("git-rebase: make --allow-empty-message the default", 2018-06-27), several arguments were given for transplanting empty commits without halting and asking the user for confirmation on each commit. These arguments were incomplete because the logic clearly assumed the only cases under consideration were transplanting of commits with empty messages (see the comment about "There are two sources for commits with empty messages). It didn't discuss or even consider rewords, squashes, etc. where the user is explicitly asked for a new commit message and provides an empty one. (My bad, I totally should have thought about that at the time, but just didn't.) Rewords and squashes are significantly different, though, as described by SZEDER: Let's suppose you start an interactive rebase, choose a commit to squash, save the instruction sheet, rebase fires up your editor, and then you notice that you mistakenly chose the wrong commit to squash. What do you do, how do you abort? Before [that commit] you could clear the commit message, exit the editor, and then rebase would say "Aborting commit due to empty commit message.", and you get to run 'git rebase --abort', and start over. But [since that commit, ...] saving the commit message as is would let rebase continue and create a bunch of unnecessary objects, and then you would have to use the reflog to return to the pre-rebase state. Also, he states: The instructions in the commit message template, which is shown for 'reword' and 'squash', too, still say... # Please enter the commit message for your changes. Lines starting # with '#' will be ignored, and an empty message aborts the commit. These are sound arguments that when editing commit messages during a sequencer operation, that if the commit message is empty then the operation should halt and ask the user to correct. The arguments in commit b00bf1c9a8dd (referenced above) still apply when transplanting previously created commits with empty commit messages, so the sequencer should not halt for those. Furthermore, all rationale so far applies equally for cherry-pick as for rebase. Therefore, make the code default to --allow-empty-message when transplanting an existing commit, and to default to halting when the user is asked to edit a commit message and provides an empty one -- for both rebase and cherry-pick. Signed-off-by: Elijah Newren <newren@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-09-04Merge branch 'pw/rebase-i-author-script-fix'Libravatar Junio C Hamano1-9/+38
Recent "git rebase -i" update started to write bogusly formatted author-script, with a matching broken reading code. These are fixed. * pw/rebase-i-author-script-fix: sequencer: fix quoting in write_author_script sequencer: handle errors from read_author_ident()
2018-09-04rebase -i: be careful to wrap up fixup/squash chainsLibravatar Johannes Schindelin1-3/+14
When an interactive rebase was stopped at the end of a fixup/squash chain, the user might have edited the commit manually before continuing (with either `git rebase --skip` or `git rebase --continue`, it does not really matter which). We need to be very careful to wrap up the fixup/squash chain also in this scenario: otherwise the next fixup/squash chain would try to pick up where the previous one was left. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
2018-08-29rebase -i: rewrite write_basic_state() in CLibravatar Alban Gruin1-11/+66
This rewrites write_basic_state() from git-rebase.sh in C. This is the first step in the conversion of init_basic_state(), hence the mode in rebase--helper.c is called INIT_BASIC_STATE. init_basic_state() will be converted in the next commit. The part of read_strategy_opts() that parses the stategy options is moved to a new function to allow its use in rebase--helper.c. Finally, the call to write_basic_state() is removed from git-rebase--interactive.sh, replaced by a call to `--init-basic-state`. Signed-off-by: Alban Gruin <alban.gruin@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-08-29rebase -i: remove unused modes and functionsLibravatar Alban Gruin1-4/+4
This removes the modes `--skip-unnecessary-picks`, `--append-todo-help`, and `--checkout-onto` from rebase--helper.c, the functions of git-rebase--interactive.sh that were rendered useless by the rewrite of complete_action(), and append_todo_help_to_file() from rebase-interactive.c. skip_unnecessary_picks() and checkout_onto() becomes static, as they are only used inside of the sequencer. Signed-off-by: Alban Gruin <alban.gruin@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-08-29rebase -i: rewrite complete_action() in CLibravatar Alban Gruin1-0/+104
This rewrites complete_action() from shell to C. A new mode is added to rebase--helper (`--complete-action`), as well as a new flag (`--autosquash`). Finally, complete_action() is stripped from git-rebase--interactive.sh. The original complete_action() would return the code 2 when the todo list contained no actions. This was a special case for rebase -i and -p; git-rebase.sh would then apply the autostash, delete the state directory, and die with the message "Nothing to do". This cleanup is rewritten in C instead of returning 2. As rebase -i no longer returns 2, the comment describing this behaviour in git-rebase.sh is updated to reflect this change. The message "Nothing to do" is now printed with error(), and so becomes "error: nothing to do". Some tests in t3404 check this value, so they are updated to fit this change. The first check might seem useless as we write "noop" to the todo list if it is empty. Actually, the todo list might contain commented commands (ie. empty commits). In this case, complete_action() won’t write "noop", and will abort without starting the editor. Signed-off-by: Alban Gruin <alban.gruin@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-08-29convert "oidcmp() != 0" to "!oideq()"Libravatar Jeff King1-4/+4
This is the flip side of the previous two patches: checking for a non-zero oidcmp() can be more strictly expressed as inequality. Like those patches, we write "!= 0" in the coccinelle transformation, which covers by isomorphism the more common: if (oidcmp(E1, E2)) As with the previous two patches, this patch can be achieved almost entirely by running "make coccicheck"; the only differences are manual line-wrap fixes to match the original code. There is one thing to note for anybody replicating this, though: coccinelle 1.0.4 seems to miss the case in builtin/tag.c, even though it's basically the same as all the others. Running with 1.0.7 does catch this, so presumably it's just a coccinelle bug that was fixed in the interim. Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-08-29convert "oidcmp() == 0" to oideq()Libravatar Jeff King1-16/+16
Using the more restrictive oideq() should, in the long run, give the compiler more opportunities to optimize these callsites. For now, this conversion should be a complete noop with respect to the generated code. The result is also perhaps a little more readable, as it avoids the "zero is equal" idiom. Since it's so prevalent in C, I think seasoned programmers tend not to even notice it anymore, but it can sometimes make for awkward double negations (e.g., we can drop a few !!oidcmp() instances here). This patch was generated almost entirely by the included coccinelle patch. This mechanical conversion should be completely safe, because we check explicitly for cases where oidcmp() is compared to 0, which is what oideq() is doing under the hood. Note that we don't have to catch "!oidcmp()" separately; coccinelle's standard isomorphisms make sure the two are treated equivalently. I say "almost" because I did hand-edit the coccinelle output to fix up a few style violations (it mostly keeps the original formatting, but sometimes unwraps long lines). Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-08-29coccinelle: use <...> for function exclusionLibravatar Jeff King1-1/+1
Sometimes we want to suppress a coccinelle transformation inside a particular function. For example, in finding conversions of hashcmp() to oidcmp(), we should not convert the call in oidcmp() itself, since that would cause infinite recursion. We write that like this: @@ identifier f != oidcmp; expression E1, E2; @@ f(...) {... - hashcmp(E1->hash, E2->hash) + oidcmp(E1, E2) ...} to match the interior of any function _except_ oidcmp(). Unfortunately, this doesn't catch all cases (e.g., the one in sequencer.c that this patch fixes). The problem, as explained by one of the Coccinelle developers in [1], is: For transformation, A ... B requires that B occur on every execution path starting with A, unless that execution path ends up in error handling code. (eg, if (...) { ... return; }). Here your A is the start of the function. So you need a call to hashcmp on every path through the function, which fails when you add ifs. [...] Another issue with A ... B is that by default A and B should not appear in the matched region. So your original rule matches only the case where every execution path contains exactly one call to hashcmp, not more than one. One way to solve this is to put the pattern inside an angle-bracket pattern like "<... P ...>", which allows zero or more matches of P. That works (and is what this patch does), but it has one drawback: it matches more than we care about, and Coccinelle uses extra CPU. Here are timings for "make coccicheck" before and after this patch: [before] real 1m27.122s user 7m34.451s sys 0m37.330s [after] real 2m18.040s user 10m58.310s sys 0m41.549s That's not ideal, but it's more important for this to be correct than to be fast. And coccicheck is already fairly slow (and people don't run it for every single patch). So it's an acceptable tradeoff. There _is_ a better way to do it, which is to record the position at which we find hashcmp(), and then check it against the forbidden function list. Like: @@ position p : script:python() { p[0].current_element != "oidcmp" }; expression E1,E2; @@ - hashcmp@p(E1->hash, E2->hash) + oidcmp(E1, E2) This is only a little slower than the current code, and does the right thing in all cases. Unfortunately, not all builds of Coccinelle include python support (including the ones in Debian). Requiring it may mean that fewer people can easily run the tool, which is worse than it simply being a little slower. We may want to revisit this decision in the future if: - builds with python become more common - we find more uses for python support that tip the cost-benefit analysis But for now this patch sticks with the angle-bracket solution, and converts all existing cocci patches. This fixes only one missed case in the current code, though it makes a much better difference for some new rules I'm adding (converting "!hashcmp()" to "hasheq()" misses over half the possible conversions using the old form). [1] https://public-inbox.org/git/alpine.DEB.2.21.1808240652370.2344@hadrien/ Helped-by: Julia Lawall <julia.lawall@lip6.fr> Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-08-23i18n: fix mistakes in translated stringsLibravatar Jean-Noël Avila1-1/+1
Fix typos and convert a question which does not expect to be replied to a simple advice. Signed-off-by: Jean-Noël Avila <jn.avila@free.fr> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-08-23append_signoff: use size_t for string offsetsLibravatar Jeff King1-2/+2
The append_signoff() function takes an "int" to specify the number of bytes to ignore. Most callers just pass 0, and the remainder use ignore_non_trailer() to skip over cruft. That function also returns an int, and uses them internally. On systems where size_t is larger than an int (i.e., most 64-bit systems), dealing with a ridiculously large commit message could end up overflowing an int, producing surprising results (e.g., returning a negative offset, which would cause us to look outside the original string). Let's consistently use size_t for these offsets through this whole stack. As a bonus, this makes the meaning of "ignore_footer" as an offset (and not a boolean) more clear. But while we're here, let's also document the interface. Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-08-23sequencer: ignore "---" divider when parsing trailersLibravatar Jeff King1-0/+2
When the sequencer code appends a signoff or cherry-pick origin, it uses the default trailer-parsing options, which treat "---" as the end of the commit message. As a result, it may be fooled by a commit message that contains that string and fail to find the existing trailer block. Even more confusing, the actual append code does not know about "---", and always appends to the end of the string. This can lead to bizarre results. E.g., appending a signoff to a commit message like this: subject body --- these dashes confuse the parser! Signed-off-by: A results in output with a final block like: Signed-off-by: A Signed-off-by: A The parser thinks the final line of the message is "body", and ignores everything else, claiming there are no trailers. So we output an extra newline separator (wrong) and add a duplicate signoff (also wrong). Since we know we are feeding a pure commit message, we can simply tell the parser to ignore the "---" divider. Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>