Re: [PATCH] No hash table for load_props(), follow-up for r34630

On Wed, 2008-12-10 at 00:48 -0800, Greg Stein wrote:
> s/You can just/You can't just/
>
> Look at the code you're changing here. Think about what it is doing
> before writing a patch.

I did run make && make check.

/me looking into the code again.

> On Wed, Dec 10, 2008 at 00:47, Greg Stein <gstein_at_gmail.com> wrote:
> > Euh... did you even *try* to run any portion of the test suite after
> > this change? Really, this isn't going to work at all. I don't even
> > have to try applying it.
> >
> > The callers need the hash table filled with the props. You can just
> > make the hash table inside the function, then throw it away. All those
> > callers won't end up with any data!!
> >
> > The hash table has to be an *out* parameter.
> >
> > -g
> >
> > On Wed, Dec 10, 2008 at 00:40, Bhuvaneswaran A <bhuvan_at_collab.net> wrote:
> >> This is a follow-up patch for r34630 to not hash table to load_props()
> >> function and avoid making hash before calling this function. The hash
> >> table is created within the function.
> >>
> >> {{
> >> This is a follow-up fix for r34630. Make hash table inside the
> >> load_props() function, instead of passing it as an argument.
> >>
> >> Suggested by: gstein
> >>
> >> * subversion/libsvn_wc/props.c
> >> (load_props): Remove the argument apr_hash_t *hash and define it
> >> within the function.
> >> (svn_wc__load_props, svn_wc__wcprop_list, modified_props): Adjust the
> >> callers.
> >> }}
> >> --
> >> Regards,
> >> Bhuvaneswaran
> >>
> >> ------------------------------------------------------
> >> http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=982077
> >

-- 
Regards,
Bhuvaneswaran
------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=982084