On 7/26/05, Julian Foad <julianfoad@btopenworld.com> wrote:
> I don't see what you mean by "would help everybody". The Subversion code base
> doesn't refer to that (struct) name "svn_repos_parse_fns2_t", only to the
> existing typedef name "svn_repos_parser_fns2_t". Perhaps you just mean that it
> would restore consistency which is potentially helpful to users of various tools.
Yes, that's what I meant. But it will also help anyone who's reading
the header file, and trying to figure out which name is the right one.
Now that you've documented svn_repos_parser_fns2_t as being
deprecated, everyone will know which struct name to use.
> How about the attached patch, folks?
It's good. The SWIG bindings are already using svn_repos_parse_fns2_t
(without the "r") because they were confused by the typedef. It's
possible that some other people were using the wrong typedef as well.
Now our header file is consistent, and I don't think anyone will be
confused.
Should we fix svn_repos_parse_fns_t as well? We don't have to -- I can
tell the SWIG bindings to ignore the inconsistency in the old struct
definition, since it's deprecated. I'm happy with your patch either
way. It might be a good idea to briefly mention this in your commit
message: "We didn't fix the inconsistency in the definition of
svn_repos_parser_fns_t, since it's deprecated."
Cheers,
David
--
David James -- http://www.cs.toronto.edu/~james
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Tue Jul 26 16:55:25 2005