On Tue, Jun 7, 2016 at 4:54 PM, Pranit Bauva <pranit.bauva@xxxxxxxxx> wrote:
> Reimplement `bisect_clean_state` shell function in C and add a
> `bisect-clean-state` subcommand to `git bisect--helper` to call it from
> git-bisect.sh .
> [...]
> Signed-off-by: Pranit Bauva <pranit.bauva@xxxxxxxxx>
> ---
> diff --git a/builtin/bisect--helper.c b/builtin/bisect--helper.c
> @@ -78,11 +87,43 @@ static int write_terms(const char *bad, const char *good)
> +int mark_for_removal(const char *refname, const struct object_id *oid,
> + int flag, void *cb_data)
> +{
> + struct string_list *refs = cb_data;
> + char *ref = xstrfmt("refs/bisect/%s", refname);
Here you're allocating a string...
> + string_list_append(refs, ref);
> + return 0;
> +}
> +
> +int bisect_clean_state(void)
> +{
> + int result = 0;
> + struct string_list refs_for_removal = STRING_LIST_INIT_DUP;
> + for_each_ref_in("refs/bisect/", mark_for_removal, (void *) &refs_for_removal);
...and the allocated string gets inserted into a string_list which
itself duplicates the string (STRING_LIST_INIT_DUP), so this is
leaking the string you created with xstrfmt(), isn't it?
> + string_list_append(&refs_for_removal, "BISECT_HEAD");
> + result |= delete_refs(&refs_for_removal);
Not sure I understand the point of using |= here rather than merely =
since this is the only place 'result' is assigned in this function.
> + string_list_clear(&refs_for_removal, 0);
> + remove_path(git_path_bisect_expected_rev());
> + remove_path(git_path_bisect_ancestors_ok());
> + remove_path(git_path_bisect_log());
> + remove_path(git_path_bisect_names());
> + remove_path(git_path_bisect_run());
> + remove_path(git_path_bisect_write_terms());
> + /* Cleanup head-name if it got left by an old version of git-bisect */
> + remove_path(git_path_head_name());
> + /* Cleanup BISECT_START last */
> + remove_path(git_path_bisect_start());
The BISECT_START comment merely repeats what the code itself already
says. I realize that you merely copied this from the shell code, but
it isn't helpful in its current form. Much more helpful would be to
explain *why* this needs to be done last. Perhaps the commit message
of the commit which introduced the comment originally would give a
clue (I haven't checked).
> + return result;
> +}
> diff --git a/git-bisect.sh b/git-bisect.sh
> @@ -430,27 +430,7 @@ bisect_reset() {
> -bisect_clean_state() {
> - # There may be some refs packed during bisection.
This comment doesn't seem to be reproduced in the C version. Should it
be? Is it no longer relevant in the C version? What does it mean
exactly?
> - git for-each-ref --format='%(refname) %(objectname)' refs/bisect/\* |
> - while read ref hash
> - do
> - git update-ref -d $ref $hash || exit
> - done
> - rm -f "$GIT_DIR/BISECT_EXPECTED_REV" &&
> - rm -f "$GIT_DIR/BISECT_ANCESTORS_OK" &&
> - rm -f "$GIT_DIR/BISECT_LOG" &&
> - rm -f "$GIT_DIR/BISECT_NAMES" &&
> - rm -f "$GIT_DIR/BISECT_RUN" &&
> - rm -f "$GIT_DIR/BISECT_TERMS" &&
> - # Cleanup head-name if it got left by an old version of git-bisect
> - rm -f "$GIT_DIR/head-name" &&
> - git update-ref -d --no-deref BISECT_HEAD &&
> - # clean up BISECT_START last
> - rm -f "$GIT_DIR/BISECT_START"
> + git bisect--helper --bisect-clean-state || exit
> }
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html