| Message ID | Y5n4o3s0DW8ilXDA@coredump.intra.peff.net (mailing list archive) |
|---|---|
| State | Accepted |
| Commit | 285da4321a9351d4c99dd1685a4aa2dfb47ff62e |
| Headers | show |
| Series | minor ref-filter error-reporting bug-fixes | expand |
diff --git a/ref-filter.c b/ref-filter.c index f40bc4d9c9..733b0149e8 100644 --- a/ref-filter.c +++ b/ref-filter.c @@ -489,7 +489,7 @@ static int person_email_atom_parser(struct ref_format *format, struct used_atom else if (!strcmp(arg, "localpart")) atom->u.email_option.option = EO_LOCALPART; else - return strbuf_addf_ret(err, -1, _("unrecognized email option: %s"), arg); + return err_bad_arg(err, atom->name, arg); return 0; }
The error message for a bogus argument to %(authoremail), etc, is: $ git for-each-ref --format='%(authoremail:foo)' fatal: unrecognized email option: foo Saying just "email" is a little vague; most of the other atom parsers would use the full name "%(authoremail)", but we can't do that here because the same function also handles %(taggeremail), etc. Until recently, passing atom->name was a bad idea, because it erroneously included the arguments in the atom name. But since the previous commit taught err_bad_arg() to handle this, we can now do so and get: fatal: unrecognized %(authoremail) argument: foo which is consistent with other atoms. Signed-off-by: Jeff King <peff@peff.net> --- ref-filter.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)