On Wed, Mar 14, 2007 at 11:15:27PM -0700, Brendan Cully wrote: > On Thursday, 01 March 2007 at 16:23, Miroslav Lichvar wrote: > > Ok, here is a patch that removes buffy_size completely and adds the > > check_box_size option. > > Looks nice, thanks. But...
Thanks for looking into this. > > Index: commands.c > > @@ -842,11 +837,9 @@ > > > > if (need_buffy_cleanup) > > { > > -#ifdef BUFFY_SIZE > > tmp = mutt_find_mailbox (buf); > > if (tmp && !tmp->new) > > mutt_update_mailbox (tmp); > > -#else > > /* fix up the times so buffy won't get confused */ > > if (st.st_mtime > st.st_atime) > > { > > @@ -856,7 +849,6 @@ > > } > > else > > utime (buf, NULL); > > -#endif > > } > > this looks a little like it should be a conditional on > option(CHECKMBOXSIZE). I think it's ok either way. Without the conditional it's a bit closer to allowing runtime switching of the option and not losing new mail flags on the mailboxes. > > Index: mx.c > > @@ -436,7 +430,7 @@ > > else if (mutt_strcmp (MMDF_SEP, tmp) == 0) > > magic = M_MMDF; > > safe_fclose (&f); > > -#ifndef BUFFY_SIZE > > + > > /* need to restore the times here, the file was not really accessed, > > * only the type was accessed. This is important, because detection > > * of "new mail" depends on those times set correctly. > > @@ -444,7 +438,6 @@ > > times.actime = st.st_atime; > > times.modtime = st.st_mtime; > > utime (path, ×); > > -#endif > > } > > else > > { Same as the first case. > > @@ -786,9 +777,7 @@ > > case M_MBOX: > > case M_MMDF: > > rc = mbox_sync_mailbox (ctx, index_hint); > > -#ifdef BUFFY_SIZE > > tmp = mutt_find_mailbox (ctx->path); > > -#endif > > break; > > > > case M_MH: > > @@ -815,10 +804,8 @@ > > mutt_error ( _("Could not synchronize mailbox %s!"), ctx->path); > > #endif > > > > -#ifdef BUFFY_SIZE > > if (tmp && tmp->new == 0) > > mutt_update_mailbox (tmp); > > -#endif > > return rc; > > } > > and probably these too? This should be ok, mutt_find_mailbox returns NULL when CHECKMBOXSIZE is disabled. -- Miroslav Lichvar