Phillip Wood wrote:
> Hi Victoria
>
> On 21/03/2022 20:34, Victoria Dye via GitGitGadget wrote:
>> From: Victoria Dye <vdye@xxxxxxxxxx>
>>
>> The explicit '--refresh' option was needed in the past when '--quiet',
>> 'reset.quiet', and/or 'reset.refresh' disabled the index refresh in 'reset
>> --mixed'. Those options have since either been deprecated or made to always
>> refresh the index by default, leaving only '--[no-]refresh' determining
>> whether the index is refreshed or not.
>>
>> Because there is nothing other than '--no-refresh' to disable index refresh,
>> we do not need a '--refresh' option to counteract some other refresh
>> disabling.
>
> '--refresh' could be used to countermand a previous '--no-refresh' (typically when using an alias that includes '--no-refresh'). Usually we have '--foo' even when it is enabled by default e.g. 'commit --verify'. I think the code below does actually support '--refresh' as parse_options() is smart enough to recognize option names beginning with '--no-' and creates an inverse option by removing the prefix rather than adding '--no-' as it normally does.
>
Makes sense, I didn't think of the alias use-case/generally wanting to
counteract the negative version of the option. Thanks!
>> To ensure users don't use what is effectively a no-op, remove '--refresh'
>> from the set of 'reset' options, as well as its usage in 'git stash'.
>>
>> Signed-off-by: Victoria Dye <vdye@xxxxxxxxxx>
>> ---
>> Documentation/git-reset.txt | 3 +--
>> builtin/reset.c | 6 +++---
>> builtin/stash.c | 4 ++--
>> t/t7102-reset.sh | 5 ++---
>> 4 files changed, 8 insertions(+), 10 deletions(-)
>>
>> diff --git a/Documentation/git-reset.txt b/Documentation/git-reset.txt
>> index df167eaa766..ba8dece0c03 100644
>> --- a/Documentation/git-reset.txt
>> +++ b/Documentation/git-reset.txt
>> @@ -107,9 +107,8 @@ OPTIONS
>> --quiet::
>> Be quiet, only report errors.
>> ---refresh::
>> --no-refresh::
>> - Proactively refresh the index after a mixed reset. Enabled by default.
>> + Disable refreshing the index after a mixed reset.
>> --pathspec-from-file=<file>::
>> Pathspec is passed in `<file>` instead of commandline args. If
>> diff --git a/builtin/reset.c b/builtin/reset.c
>> index 54324217f93..d9427abc483 100644
>> --- a/builtin/reset.c
>> +++ b/builtin/reset.c
>> @@ -392,7 +392,7 @@ static int git_reset_config(const char *var, const char *value, void *cb)
>> int cmd_reset(int argc, const char **argv, const char *prefix)
>> {
>> int reset_type = NONE, update_ref_status = 0, quiet = 0;
>> - int refresh = -1;
>> + int refresh = 1;
>> int patch_mode = 0, pathspec_file_nul = 0, unborn;
>> const char *rev, *pathspec_from_file = NULL;
>> struct object_id oid;
>> @@ -400,8 +400,8 @@ int cmd_reset(int argc, const char **argv, const char *prefix)
>> int intent_to_add = 0;
>> const struct option options[] = {
>> OPT__QUIET(&quiet, N_("be quiet, only report errors")),
>> - OPT_BOOL(0, "refresh", &refresh,
>> - N_("skip refreshing the index after reset")),
>> + OPT_SET_INT(0, "no-refresh", &refresh,
>> + N_("skip refreshing the index after reset"), 0),
>
> I'm confused why we need to change the option type here - am I missing something?
>
This was to explicitly prevent a user from specifying "--refresh", only
allowing "--no-refresh". But, based on your earlier comments, there's still
some value in having "--refresh", so I'll drop this patch entirely.
>
> Best Wishes
>
> Phillip
