On 5/8/05, Patrick Mayweg <mayweg@qint.de> wrote:
> Hi Philip,
>
> Philip Martin wrote:
>
> >"Peter N. Lundblad" <peter@famlundblad.se> writes:
> >
> >
> >
> >>Ouch! Putting vars of those sizes on the stack isn't very friendly. Just
> >>allocate it out of a pool instead.
> >>
> >>(Saying this without having investigated in detail, but I can't imagine it
> >>should be a problem.)
> >>
> >>
> >
> >Look at this:
> >
> >
> This is the only case, where a pool is not available.
>
> >svn_error_t *
> >svn_subst_translate_stream (svn_stream_t *s, /* src stream */
> > svn_stream_t *d, /* dst stream */
> > const char *eol_str,
> > svn_boolean_t repair,
> > const svn_subst_keywords_t *keywords,
> > svn_boolean_t expand)
> >{
> > char buf[SVN_STREAM_CHUNK_SIZE + 1];
> > const char *p, *interesting;
> >
> >There is no readily available pool, and neither svn_stream_t nor
> >svn_subst_keywords_t provides one. There is a void* baton in
> >svn_stream_t but svn_subst_translate_stream can't really make any
> >assumptions about it.
> >
> >
> All callers of this function have a pool available. If I could extend
> the interface by a pool parameter, I could change that case too.
> Patrick
Looking at the name of the function, that will need to happen before
1.2 final, since it seems like a public function. If we need a revved
API, it has to be done in a minor release.
We could ofcourse update all internal callers, but it won't fix the
problem for external callers.
bye,
Erik.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Sun May 8 15:29:42 2005