Stefan Fuhrmann wrote: > Daniel Shahaf wrote: >> Julian Foad wrote on Tue, Apr 03, 2012 at 18:09:19 +0100: >>> For comments, please: >>> [[[Tweak the recently introduced svn_stringbuf_create_empty() function. >>> >>> Unlike with svn_string_create_empty() which produces an immutable >>> string, the resulting string buffer is often going to have data >>> appended to it. >> I don't see an issue with it. +1 >> > Well, the point was that those *won't* get data appended > but rather get resized later and *then* filled. IIRC, I used it > for all kinds of receive / decode buffers.
Lots of places don't use it that way, they simply append. > For now, however, the change is fine. Once I got through my > commit backlog, I may run the profiler again and see whether > there is an issue or not. Great, thanks. r1309730. - Julian
