SVN_REVNUM_T_FMT anyone?, WAS: RE: [PATCH]: fix #661 (Bad user supplied rev # produces scary error)

[Sander Striker <st...@apache.org>]

> From: David Kimdon [mailto:dwhedon@dualmedia.fr]On Behalf Of David
> Kimdon
> Sent: 01 May 2002 07:58

> Hi,
> 
> This patch fixes the spooky error message of issue #661.

[...]
> +  return 
> +    svn_error_createf
> +    (SVN_ERR_FS_NO_SUCH_REVISION, 0, 0, fs->pool,
> +     "reference to non-existent revision `%lu' in filesystem `%s'",
                                              ^^
> +     (unsigned long) rev, fs->path);
         ^^^^^^^^^^^^^

I've seen this a lot in the codebase.  What I would really like
is an SVN_REVNUM_T_FMT define, which we can use similar like the
APR_xxx_T_FMT defines (example: printf("%" APR_OFF_T_FMT, some_apr_off_t_var); ).
We can loose the cast with this and changing the type (lets not do that!),
will only be a one place modification.

Sander

PS.  This has nothing to do with the presented patch.


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org

Re: SVN_REVNUM_T_FMT anyone?, WAS: RE: [PATCH]: fix #661 (Bad user supplied rev # produces scary error)

[Karl Fogel <kf...@newton.ch.collab.net>]

"Sander Striker" <st...@apache.org> writes:
> I've seen this a lot in the codebase.  What I would really like
> is an SVN_REVNUM_T_FMT define, which we can use similar like the
> APR_xxx_T_FMT defines (example: printf("%" APR_OFF_T_FMT,
> some_apr_off_t_var); ).
> We can loose the cast with this and changing the type (lets not do that!),
> will only be a one place modification.

Ahhh, good idea.  Done in revision 1836.

-K

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org