On Mon, Dec 06 2021, Junio C Hamano wrote:
> Ævar Arnfjörð Bjarmason <avarab@xxxxxxxxx> writes:
>
>> In preceding commits the rest of the vreportf() users outside of
>> usage.c have been migrated to die_message(), leaving only the
>> git_die_config() function added in 5a80e97c827 (config: add
>> `git_die_config()` to the config-set API, 2014-08-07).
>>
>> Let's have its callers call error() themselves if they want to emit a
>> message, which is exactly what git_die_config() was doing for them
>> before emitting its own die() message.
>
> I do not quite get this. If git_die_config() has been showing the
> message for them, and if the existing callers can just use error(),
> why not git_die_config() call error() on behalf of these callers?
>
>> diff --git a/builtin/fast-import.c b/builtin/fast-import.c
>> index 2b2e28bad79..4e2432bb491 100644
>> --- a/builtin/fast-import.c
>> +++ b/builtin/fast-import.c
>> @@ -3456,9 +3456,10 @@ static void git_pack_config(void)
>> }
>> if (!git_config_get_int("pack.indexversion", &indexversion_value)) {
>> pack_idx_opts.version = indexversion_value;
>> - if (pack_idx_opts.version > 2)
>> - git_die_config("pack.indexversion",
>> - "bad pack.indexversion=%"PRIu32, pack_idx_opts.version);
>> + if (pack_idx_opts.version > 2) {
>> + error("bad pack.indexversion=%"PRIu32, pack_idx_opts.version);
>> + git_die_config("pack.indexversion");
>> + }
>
> This is exactly what triggered the question above, and the pattern
> repeats elsewhere, too.
>
>> @@ -2550,18 +2552,12 @@ void git_die_config_linenr(const char *key, const char *filename, int linenr)
>> key, filename, linenr);
>> }
>>
>> -NORETURN __attribute__((format(printf, 2, 3)))
>> -void git_die_config(const char *key, const char *err, ...)
>> +NORETURN
>> +void git_die_config(const char *key)
>> {
>> const struct string_list *values;
>> struct key_value_info *kv_info;
>>
>> - if (err) {
>> - va_list params;
>> - va_start(params, err);
>> - vreportf("error: ", err, params);
>> - va_end(params);
>
> I get that we do not want to expose vreportf() to this caller, and I
> agree with the goal, but wouldn't it be the matter of calling
> get_error_routine() and calling it with err and params here, instead
> of losing the whole block? Is that insufficient to avoid toucing
> all the callers?
I'll fix that in the incoming re-roll. This really didn't belong in this
series but a later one.
FWIW that change was because like this we'll accurately report the
source of the "error" when the function learns to spew <file>:<line> at
the user.
But for now I'll just have it do as you suggest...
