BLECK! Ok this is my 'fault'. Please note that EXACTLY this proposed code exists and is used in the ssl core filter of httpd.
Before you go blasting to smithereens, could we please discuss the merits of this approach? It's clear that every brigade application at some point will FINALLY consume the brigade, and apr-util aught to be the standard of how this is done, not each and every arbitrary application. In the SSL case, although the data isn't transmitted over the wire, it is slurped, exactly as much as is useful at that moment, and then destroyed as it has been crypted into a SSL bytestream for the wire. Persisting any of those plaintext buckets (except obviously any immortal buckets) after the crypt operation becomes a memory hog. Are we leaving this up to each and every app to reinvent? Any thoughts? Bill At 03:08 PM 6/22/2004, [EMAIL PROTECTED] wrote: >jorton 2004/06/22 13:08:03 > > Modified: include apr_buckets.h > Log: > * include/apr_buckets.h: Remove some vapourware. > > Revision Changes Path > 1.157 +0 -11 apr-util/include/apr_buckets.h > > Index: apr_buckets.h > =================================================================== > RCS file: /home/cvs/apr-util/include/apr_buckets.h,v > retrieving revision 1.156 > retrieving revision 1.157 > diff -d -w -u -r1.156 -r1.157 > --- apr_buckets.h 13 Feb 2004 09:55:26 -0000 1.156 > +++ apr_buckets.h 22 Jun 2004 20:08:03 -0000 1.157 > @@ -701,17 +701,6 @@ > apr_off_t point, > apr_bucket **after_point); > > -#if APR_NOT_DONE_YET > -/** > - * consume nbytes from beginning of b -- call apr_bucket_destroy as > - * appropriate, and/or modify start on last element > - * @param b The brigade to consume data from > - * @param nbytes The number of bytes to consume > - */ > -APU_DECLARE(void) apr_brigade_consume(apr_bucket_brigade *b, > - apr_off_t nbytes); > -#endif > - > /** > * Return the total length of the brigade. > * @param bb The brigade to compute the length of > > >
