Message ID | pull.1250.v2.git.1654778272871.gitgitgadget@gmail.com (mailing list archive) |
---|---|
State | New, archived |
Headers | show |
Series | [v2] ls-files.c: add --object-only option | expand |
"ZheNing Hu via GitGitGadget" <gitgitgadget@gmail.com> writes: > From: ZheNing Hu <adlternative@gmail.com> > > `git ls-files --stage` default output format is: > > [<tag> ]<mode> <object> <stage> <file> > > sometime we want to find a path's corresponding objectname, "sometime" -> "When", perhaps. If you really want to say that, "Sometimes, " is also good, though. By the way, I do not think you are "want to find a path's corresponding objectname" at all with this feature. The output from "ls-files -s" will have many object names, one per each path if the index is merged, and if you discard the path, you no longer can tell which object name corresponds to which path. > we will parse the output and extract objectname from it > again and again. Why is that a problem? "again and again" is over-exaggerating; you'd munge each line just once. It would help readers if you say WHY you want to find object names. Perhaps you want to find the set of objects that are registered in the index, regardless of their paths? In any case, the paragraph needs a rewrite. > So introduce a new option `--object-only` which can only > output objectname when giving `--stage` or `--resolve-undo`. "which can only" makes it sound like you are complaining about its limitation. I read these two lines to mean "git ls-files -s --object-only" does not even give me the stage information, but that would make the command completely useless, so I am assuming that is not what you meant to say. The same comment applies for resolve-undo, which is merely "what 'ls-files -s' may have given before you resolved". If you borrowed a feature from another existing command, say that explicitly, which will allow your commit to gain confidence by reviewers and future readers by showing that you care about overall consistency in the system. Add a new option `--object-only` that omits the mode and filename from the output, taking inspiration from the option with the same name in the `git ls-tree` command. or something like that, perhaps. How does/should this interact with the `--deduplicate` option? If we are not giving stages and truly giving only object names (which I doubt is what we want, by the way), then we can and should deduplicate the output when the option is given. If we have two identical blobs at different paths, or two identical blobs at the same path but at different stages, shouldn't we get only a single copy of output for that blob, as we are not showing paths nor stages, right? How does/should this behave when --stage is not given? I have a suspicion that this whole thing is misdesigned. Instead of making it piggy back on --stage, don't you want to make it an independent option? I.e. git ls-files --object-only with no other option would behave like git ls-files -s | sed -e 's/^[0-6]* \([0-9a-f]*\) .*/\1/' and it is an error to combine it with -s or --deduplicate. If the purpose is to learn the set of objects registered in the index, then it might even make sense to make it an equivalent to git ls-files -s | sed -e 's/^[0-6]* \([0-9a-f]*\) .*/\1/' | sort -u as duplicates or order of the entries is no use for such a use case. It entirely depends on WHY you want to find object names, and that is why I asked it much earlier in this message. And I do not think it makes any sense to give resolve-undo information without paths nor stages at all. Please do not tie this with that mode. In short - this probably is better done as a separate independent mode "--object-only", rather than a piggy-back feature on top of existing other features like "-s" and "--resolve-undo". - the new mode should be made mutually incompatible with "-s" and "--resolve-undo". There may be other options that this should be incompatible, like "--tag" and "--full-name". Thanks.
Junio C Hamano <gitster@pobox.com> 于2022年6月10日周五 03:51写道: > > "ZheNing Hu via GitGitGadget" <gitgitgadget@gmail.com> writes: > > > From: ZheNing Hu <adlternative@gmail.com> > I read these two lines to mean "git ls-files -s --object-only" does > not even give me the stage information, but that would make the > command completely useless, so I am assuming that is not what you > meant to say. The same comment applies for resolve-undo, which is > merely "what 'ls-files -s' may have given before you resolved". > > If you borrowed a feature from another existing command, say that > explicitly, which will allow your commit to gain confidence by > reviewers and future readers by showing that you care about overall > consistency in the system. > > Add a new option `--object-only` that omits the mode and > filename from the output, taking inspiration from the option > with the same name in the `git ls-tree` command. > > or something like that, perhaps. > Yes, this message will be better. I think it omits not only mode, filename, but also tag, stage, eol info, debug message. > How does/should this interact with the `--deduplicate` option? > > If we are not giving stages and truly giving only object names > (which I doubt is what we want, by the way), then we can and should > deduplicate the output when the option is given. If we have two > identical blobs at different paths, or two identical blobs at the > same path but at different stages, shouldn't we get only a single > copy of output for that blob, as we are not showing paths nor > stages, right? > I have think about it for a long time, I think deduplicate is used for removing duplicates entries which caused by one path can have different stage. But we now care about a output format just like %(objectname), if we need to deduplicate it, when we use --format="%(objectname) %(path)" later, do we need to deduplicate its output too? I think we should disable --deduplicate when we are using --object-only. > How does/should this behave when --stage is not given? > > I have a suspicion that this whole thing is misdesigned. Instead of > making it piggy back on --stage, don't you want to make it an > independent option? I.e. > > git ls-files --object-only > > with no other option would behave like > > git ls-files -s | sed -e 's/^[0-6]* \([0-9a-f]*\) .*/\1/' > > and it is an error to combine it with -s or --deduplicate. If the > purpose is to learn the set of objects registered in the index, then > it might even make sense to make it an equivalent to > > git ls-files -s | > sed -e 's/^[0-6]* \([0-9a-f]*\) .*/\1/' | > sort -u > > as duplicates or order of the entries is no use for such a use > case. > > It entirely depends on WHY you want to find object names, and that > is why I asked it much earlier in this message. > My origin requirement is to do a app which can move one file to another file in a bare git-repo, so I need to get first file object-name for second file to update-index. It can parsed by the app of course, but I think such kind of work left to git itself can help other app programers. Maybe you are right that --object-only or --format should not be sub-option of --stage or --resolve-undo, I will think about how to implement it later. > And I do not think it makes any sense to give resolve-undo > information without paths nor stages at all. Please do not tie this > with that mode. > > In short > > - this probably is better done as a separate independent mode > "--object-only", rather than a piggy-back feature on top of > existing other features like "-s" and "--resolve-undo". > > - the new mode should be made mutually incompatible with "-s" and > "--resolve-undo". There may be other options that this should be > incompatible, like "--tag" and "--full-name". > By the way, if we need --format for git ls-files, which atoms should we keep? I think those atoms are undoubtedly necessary to keep %(tag) %(objectmode) %(objectname) %(stage) %(path) git ls-files --stage just like git ls-file --format="%(tag)%(obejctmode) %(objectname) %(stage)\t%(path)" but for these follow atoms, I am not sure if we need them? %(eofinfo) %(debug) %(eol) %(ctime) %(ctime:sec) %(ctime:nsec) %(mtime) %(mtime:sec) %(mtime:nsec) %(dev) %(ino) %(uid) %(gid) %(size) %(flags) Thanks
ZheNing Hu <adlternative@gmail.com> writes: > I think those atoms are undoubtedly necessary to keep > > %(tag) > %(objectmode) > %(objectname) > %(stage) > %(path) I am not sure what you mean by "keep". You cannot keep what you do not have yet ;-) If ls-files needs (that is a big if; it is a plumbing to be used by whatever program that want to assemble the pieces, and it shouldn't have to learn such assembly itself) to support "--format", so that people can reinvent its "-s" output (but why? There already is "-s" output available), then the above would be necessary (assuming that via the "--format" the user will be able to supply inter-field spaces and tabs properly). I do not know if there are other things available in output other than the "-s" option produces offhand, but if there are, they would need to be added for completeness.
diff --git a/Documentation/git-ls-files.txt b/Documentation/git-ls-files.txt index 0dabf3f0ddc..9736b02b565 100644 --- a/Documentation/git-ls-files.txt +++ b/Documentation/git-ls-files.txt @@ -13,7 +13,7 @@ SYNOPSIS [-c|--cached] [-d|--deleted] [-o|--others] [-i|--|ignored] [-s|--stage] [-u|--unmerged] [-k|--|killed] [-m|--modified] [--directory [--no-empty-directory]] [--eol] - [--deduplicate] + [--deduplicate] [--object-only] [-x <pattern>|--exclude=<pattern>] [-X <file>|--exclude-from=<file>] [--exclude-per-directory=<file>] @@ -88,6 +88,10 @@ OPTIONS When any of the `-t`, `--unmerged`, or `--stage` option is in use, this option has no effect. +--object-only: + When giving `--stage` or `--resolve-undo` , only output `<object>` + instead of `[<tag> ]<mode> <object> <stage> <file>` format. + -x <pattern>:: --exclude=<pattern>:: Skip untracked files matching pattern. diff --git a/builtin/ls-files.c b/builtin/ls-files.c index e791b65e7e9..2fef5f40a3f 100644 --- a/builtin/ls-files.c +++ b/builtin/ls-files.c @@ -26,6 +26,7 @@ static int show_deleted; static int show_cached; static int show_others; static int show_stage; +static int object_only; static int show_unmerged; static int show_resolve_undo; static int show_modified; @@ -241,10 +242,15 @@ static void show_ce(struct repository *repo, struct dir_struct *dir, if (!show_stage) { fputs(tag, stdout); } else { + const char *object_name = repo_find_unique_abbrev(repo, &ce->oid, abbrev); + if (object_only) { + printf("%s%c", object_name, line_terminator); + return; + } printf("%s%06o %s %d\t", tag, ce->ce_mode, - repo_find_unique_abbrev(repo, &ce->oid, abbrev), + object_name, ce_stage(ce)); } write_eolinfo(repo->index, ce, fullname); @@ -274,6 +280,10 @@ static void show_ru_info(struct index_state *istate) for (i = 0; i < 3; i++) { if (!ui->mode[i]) continue; + if (object_only) { + printf("%s%c", find_unique_abbrev(&ui->oid[i], abbrev), line_terminator); + continue; + } printf("%s%06o %s %d\t", tag_resolve_undo, ui->mode[i], find_unique_abbrev(&ui->oid[i], abbrev), i + 1); @@ -635,6 +645,8 @@ int cmd_ls_files(int argc, const char **argv, const char *cmd_prefix) DIR_SHOW_IGNORED), OPT_BOOL('s', "stage", &show_stage, N_("show staged contents' object name in the output")), + OPT_BOOL(0, "object-only", &object_only, + N_("only show staged contents' object name in the output")), OPT_BOOL('k', "killed", &show_killed, N_("show files on the filesystem that need to be removed")), OPT_BIT(0, "directory", &dir.flags, @@ -734,6 +746,10 @@ int cmd_ls_files(int argc, const char **argv, const char *cmd_prefix) die("ls-files --recurse-submodules does not support " "--error-unmatch"); + if (object_only && !show_stage && !show_resolve_undo) + die(_("ls-files --object-only only used with --stage " + "or --resolve-undo")); + parse_pathspec(&pathspec, 0, PATHSPEC_PREFER_CWD, prefix, argv); diff --git a/t/t2030-unresolve-info.sh b/t/t2030-unresolve-info.sh index f691e6d9032..cdab953980c 100755 --- a/t/t2030-unresolve-info.sh +++ b/t/t2030-unresolve-info.sh @@ -32,6 +32,31 @@ check_resolve_undo () { test_cmp "$msg.expect" "$msg.actual" } +check_resolve_undo_object_only() { + msg=$1 + shift + while case $# in + 0) break ;; + 1|2|3) BUG "wrong arguments" ;; + esac + do + path=$1 + shift + for stage in 1 2 3 + do + sha1=$1 + shift + case "$sha1" in + '') continue ;; + esac + sha1=$(git rev-parse --verify "$sha1") && + printf "%s\n" $sha1 + done + done >"$msg.expect" && + git ls-files --resolve-undo --object-only >"$msg.actual" && + test_cmp "$msg.expect" "$msg.actual" +} + prime_resolve_undo () { git reset --hard && git checkout second^0 && @@ -194,4 +219,12 @@ test_expect_success 'rerere forget (add-add conflict)' ' test_i18ngrep "no remembered" actual ' +test_expect_success '--resolve-undo with --object-only' ' + prime_resolve_undo && + check_resolve_undo_object_only kept fi/le initial:fi/le second:fi/le third:fi/le && + git checkout second^0 && + echo switching clears && + check_resolve_undo cleared +' + test_done diff --git a/t/t3004-ls-files-basic.sh b/t/t3004-ls-files-basic.sh index a16e25c79bd..6c81ead140e 100755 --- a/t/t3004-ls-files-basic.sh +++ b/t/t3004-ls-files-basic.sh @@ -52,4 +52,30 @@ test_expect_success SYMLINKS 'ls-files with absolute paths to symlinks' ' test_cmp expect actual ' +test_expect_success 'git ls-files --stage with --object-only' ' + git init test && + test_when_finished "rm -rf test" && + echo a >test/a.txt && + echo b >test/b.txt && + git -C test add a.txt b.txt && + oid1=$(git -C test hash-object a.txt) && + oid2=$(git -C test hash-object b.txt) && + git -C test ls-files --stage --object-only >actual && + cat >expect <<-EOF && + $oid1 + $oid2 + EOF + test_cmp expect actual +' + +test_expect_success 'git ls-files --object-only without --stage or --resolve-undo' ' + git init test && + test_when_finished "rm -rf test" && + echo a >test/a.txt && + echo b >test/b.txt && + git -C test add a.txt b.txt && + test_must_fail git -C test ls-files --object-only 2>stderr && + grep "fatal: ls-files --object-only only used with --stage or --resolve-undo" stderr +' + test_done