On Fri, Jul 11, 2014 at 9:38 PM, Jim Jagielski <[email protected]> wrote: > > On Jul 11, 2014, at 12:29 PM, Yann Ylavic <[email protected]> wrote: > >>> >>> Maybe we could also leave skiplist as is and introduce a new type >>> (skipmap?) that would be ordered and that would take both key and >>> value as arguments (in the relevant functions)... >> > > Yeah... good idea.
OK, will propose that. I'll commit skiplist's size computation bugfix included in the current patch, and maybe the stack reuse (to avoid malloc()s for each insert()). Regarding apr_skiplist_destroy(), don't you think we should free() the given pointer when the skiplist was malloc()ated, and have apr_skiplist_clear() which does not? Last but not least, apr_skiplist_remove() to remove one (as is) or all the matching elements?
