On Fri, Sep 18, 2015 at 06:25:24PM -0400, Eric Sunshine wrote:
> On Fri, Sep 18, 2015 at 03:12:50PM -0700, Brian Norris wrote:
> > --- a/git-send-email.perl
> > +++ b/git-send-email.perl
> > @@ -1136,7 +1136,7 @@ sub smtp_auth_maybe {
> >  
> >     # Check mechanism naming as defined in:
> >     # https://tools.ietf.org/html/rfc4422#page-8
> > -   if ($smtp_auth !~ /^(\b[A-Z0-9-_]{1,20}\s*)*$/) {
> > +   if ($smtp_auth && $smtp_auth !~ /^(\b[A-Z0-9-_]{1,20}\s*)*$/) {
> >             die "invalid smtp auth: '${smtp_auth}'";
> >     }
> 
> Thanks, makes sense. I wonder if moving the check to the point where
> $smtp_auth is actually used (despite the noisier diff) would be cleaner,
> like this:
> 
> --- 8< ---
> diff --git a/git-send-email.perl b/git-send-email.perl
> index c5a3f76..2a5ceda 100755
> --- a/git-send-email.perl
> +++ b/git-send-email.perl
> @@ -1134,12 +1134,6 @@ sub smtp_auth_maybe {
>               Authen::SASL->import(qw(Perl));
>       };
>  
> -     # Check mechanism naming as defined in:
> -     # https://tools.ietf.org/html/rfc4422#page-8
> -     if ($smtp_auth !~ /^(\b[A-Z0-9-_]{1,20}\s*)*$/) {
> -             die "invalid smtp auth: '${smtp_auth}'";
> -     }
> -
>       # TODO: Authentication may fail not because credentials were
>       # invalid but due to other reasons, in which we should not
>       # reject credentials.
> @@ -1154,6 +1148,12 @@ sub smtp_auth_maybe {
>               my $cred = shift;
>  
>               if ($smtp_auth) {
> +                     # Check mechanism naming as defined in:
> +                     # https://tools.ietf.org/html/rfc4422#page-8
> +                     if ($smtp_auth !~ /^(\b[A-Z0-9-_]{1,20}\s*)*$/) {
> +                             die "invalid smtp auth: '${smtp_auth}'";
> +                     }
> +
>                       my $sasl = Authen::SASL->new(
>                               mechanism => $smtp_auth,
>                               callback => {
> --- 8< ---

By moving the patch into the sub-subroutine (is this a lambda? I'm not
too familiar with my perl), I think you change the order of the checks.
So, previously, initial password auth would happen after the
$smtp_auth format check. With your patch, it comes first.

My patch:

  $ git send-email --smtp-auth="@" ...
  ...
  Send this email? ([y]es|[n]o|[q]uit|[a]ll): y
  invalid smtp auth: '@' at ./git-send-email.perl line 1140.

Your patch:

  $ git send-email --smtp-auth="@" ...
  ...
  Send this email? ([y]es|[n]o|[q]uit|[a]ll): y
  Password for xxxx:
  invalid smtp auth: '@' at ./git-send-email.perl line 1155.

Seems like the former is a little better, so you don't have to waste
time with your password too many times.

Regards,
Brian
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Reply via email to