On Wed, Apr 7, 2010 at 4:59 PM, Hyrum K. Wright <hyrum_wri...@mail.utexas.edu> wrote: > > > On Thu, Mar 25, 2010 at 2:28 PM, Hyrum K. Wright > <hyrum_wri...@mail.utexas.edu> wrote: >> >> On Mar 9, 2010, at 4:37 PM, Hyrum K. Wright wrote: >> >> > >> > On Mar 9, 2010, at 2:00 PM, William A. Rowe Jr. wrote: >> > >> >> On 3/9/2010 11:48 AM, Jeff Trawick wrote: >> >>> On Tue, Mar 9, 2010 at 11:19 AM, Hyrum K. Wright >> >>> <hyrum_wri...@mail.utexas.edu> wrote: >> >>>> In using the apr_hash datastructure in Subversion, we've found that >> >>>> we >> >>>> often only want the key or value from a hash. Furthermore, casting >> >>>> the various return parameters has proven cumbersome. To solve this >> >>>> problem, we've introduced three helper functions to return the key, >> >>>> key length, and value from a hash iterator. >> >>>> >> >>>> We've found these functions quite useful, so I'm including a patch to >> >>>> add them to APR proper. >> >>> >> >>> IMO these functions are a natural addition; any concerns from the >> >>> crowd? >> >>> >> >>>> The patch is against trunk, but if possible, >> >>>> I'd like to see these APIs backported to 1.4.x and 1.5.x. >> >>> >> >>> too late for 1.4.x >> >> >> >> But not 1.5 - sounds like a great idea. Only change I'd suggest is >> >> _key_len >> >> or even _keylen rather than _klen for an exported public function. >> > >> > I'm fine with that. >> > >> > I don't think I've got the appropriate karma to commit to APR, so >> > whoever commits this patch can make that change (or would it be worse the >> > hassle for me to create a new patch?). >> > >> > I'm also planning a followup which const-ifies the apr_hash_index_t >> > params to these functions, as well as apr_hash_this(). Thoughts? >> >> Ping. It's been a while since anybody has looked at this. Any plans to >> commit the original (or modified original) patch? > > Unfortunately, this still hasn't gotten any response, so I've filed it in > the issue tracker as issue 49065: > https://issues.apache.org/bugzilla/show_bug.cgi?id=49065
finally committed care to confirm (or fix) the applicability of your patch to the 1.5.x branch?
