| Message ID | 20240303220600.2491792-2-gitster@pobox.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | [1/1] clean: further clean-up of implementation around "--force" | expand |
Junio C Hamano <gitster@pobox.com> writes: > @@ -950,13 +950,8 @@ int cmd_clean(int argc, const char **argv, const char *prefix) > argc = parse_options(argc, argv, prefix, options, builtin_clean_usage, > 0); > > - /* Dry run won't remove anything, so requiring force makes no sense */ > - if (dry_run) > - require_force = 0; > - > - if (require_force != 0 && !force && !interactive) > - die(_("clean.requireForce is true and neither -f nor -i given:" > - " refusing to clean")); > + if (require_force != 0 && !force && !interactive && !dry_run) > + die(_("clean.requireForce is true and -f not given: refusing to clean")); > > if (force > 1) > rm_flags = 0; An obvious alternative way to clean-up the logic is to do this instead: if (dry_run || interactive)) require_force = 0; if (require_force != 0 && !force) die(_("clean.requireForce is true and ...")); But as I wrote, the most important improvement done by Sergey's patch was to remove the dual meaning of the "force" variable so that it indicates if the "--force" option was given and nothing else, while the "require_force" variable indicates if clean.requireForce was given and nothing else. From that point of view, the conditional tweaking done to require_force in the above alternative makes the code worse, relative to Sergey's patch, and certainly to its follow up, my patch about "--interactive".
Junio C Hamano <gitster@pobox.com> writes: > We clarified how clean.requireForce interacts with the --dry-run > option in the previous commit, both in the implementation and in the > documentation. Even when "git clean" (without other options) is > required to be used with "--force" (i.e. either clean.requireForce > is unset, or explicitly set to true) to protect end-users from > casual invocation of the command by mistake, "--dry-run" does not > require "--force" to be used, because it is already its own > protection mechanism by being a no-op to the working tree files. > > The previous commit, however, missed another clean-up opportunity > around the same area. Just like in the "--dry-run" mode, the > command in the "--interactive" mode does not require "--force", > either. This is because by going interactive and giving the end > user one more step to confirm, the mode itself is serving as its own > protection mechanism. > > Let's take things one step further, unify the code that defines > interaction between `--force` and these two other options. Just > like we added explanation for the reason why "--dry-run" does not > honor `clean.requireForce`, add the same explanation for > "--interactive". Finally, add some tests to show the interaction > between "--force" and "--interactive" (we already have tests that > show interaction between "--force" and "--dry-run"). Looks fine to me, including the patch itself. Thanks, -- Sergey Organov
diff --git a/Documentation/config/clean.txt b/Documentation/config/clean.txt index b19ca210f3..c0188ead4e 100644 --- a/Documentation/config/clean.txt +++ b/Documentation/config/clean.txt @@ -1,3 +1,3 @@ clean.requireForce:: A boolean to make git-clean refuse to delete files unless -f - or -i is given. Defaults to true. + is given. Defaults to true. diff --git a/Documentation/git-clean.txt b/Documentation/git-clean.txt index 662eebb852..082d033438 100644 --- a/Documentation/git-clean.txt +++ b/Documentation/git-clean.txt @@ -37,7 +37,7 @@ OPTIONS --force:: If the Git configuration variable clean.requireForce is not set to false, 'git clean' will refuse to delete files or directories - unless given -f or -i. Git will refuse to modify untracked + unless given -f. Git will refuse to modify untracked nested git repositories (directories with a .git subdirectory) unless a second -f is given. @@ -45,6 +45,8 @@ OPTIONS --interactive:: Show what would be done and clean files interactively. See ``Interactive mode'' for details. + Configuration variable clean.requireForce is ignored, as + this mode gives its own safety protection by going interactive. -n:: --dry-run:: diff --git a/builtin/clean.c b/builtin/clean.c index 41502dcb0d..29efe84153 100644 --- a/builtin/clean.c +++ b/builtin/clean.c @@ -950,13 +950,8 @@ int cmd_clean(int argc, const char **argv, const char *prefix) argc = parse_options(argc, argv, prefix, options, builtin_clean_usage, 0); - /* Dry run won't remove anything, so requiring force makes no sense */ - if (dry_run) - require_force = 0; - - if (require_force != 0 && !force && !interactive) - die(_("clean.requireForce is true and neither -f nor -i given:" - " refusing to clean")); + if (require_force != 0 && !force && !interactive && !dry_run) + die(_("clean.requireForce is true and -f not given: refusing to clean")); if (force > 1) rm_flags = 0; diff --git a/t/t7300-clean.sh b/t/t7300-clean.sh index 611b3dd3ae..1f7201eb60 100755 --- a/t/t7300-clean.sh +++ b/t/t7300-clean.sh @@ -407,6 +407,12 @@ test_expect_success 'clean.requireForce and -f' ' ' +test_expect_success 'clean.requireForce and --interactive' ' + git clean --interactive </dev/null >output 2>error && + test_grep ! "requireForce is true and" error && + test_grep "\*\*\* Commands \*\*\*" output +' + test_expect_success 'core.excludesfile' ' echo excludes >excludes &&
We clarified how clean.requireForce interacts with the --dry-run option in the previous commit, both in the implementation and in the documentation. Even when "git clean" (without other options) is required to be used with "--force" (i.e. either clean.requireForce is unset, or explicitly set to true) to protect end-users from casual invocation of the command by mistake, "--dry-run" does not require "--force" to be used, because it is already its own protection mechanism by being a no-op to the working tree files. The previous commit, however, missed another clean-up opportunity around the same area. Just like in the "--dry-run" mode, the command in the "--interactive" mode does not require "--force", either. This is because by going interactive and giving the end user one more step to confirm, the mode itself is serving as its own protection mechanism. Let's take things one step further, unify the code that defines interaction between `--force` and these two other options. Just like we added explanation for the reason why "--dry-run" does not honor `clean.requireForce`, add the same explanation for "--interactive". Finally, add some tests to show the interaction between "--force" and "--interactive" (we already have tests that show interaction between "--force" and "--dry-run"). Signed-off-by: Junio C Hamano <gitster@pobox.com> --- Documentation/config/clean.txt | 2 +- Documentation/git-clean.txt | 4 +++- builtin/clean.c | 9 ++------- t/t7300-clean.sh | 6 ++++++ 4 files changed, 12 insertions(+), 9 deletions(-)