On Mon, 19 Jul 2004, Joe Orton wrote: > On Sat, Jul 17, 2004 at 03:22:35PM -0000, [EMAIL PROTECTED] wrote: > > niq 2004/07/17 08:22:35 > > > > Modified: modules/filters mod_deflate.c > > Log: > > Fix previous patch to deal correctly with multiple empty brigades before > > we know if there's any content, and not re-process the headers. > > Is there no simpler fix for this e.g. first thing the filter does is "if > (APR_BRIGADE_EMPTY(bb)) return APR_SUCCESS;". And to avoid the
Yes, that should work (but leaves us to reprocess the whole thing next time round). Is that better or worse? Or come to think of it, we can both return APR_SUCCESS and set a flag. > re-process issue just ap_remove_output_filter(f) if finding an EOS-only > brigade? Do you recollect the discussion around when that patch went in? I don't in full, but I had a nagging recollection of someone having proposed a simpler solution but found it didn't work. Just enough to persuade me to preserve the loop-over-buckets test. -- Nick Kew
