On Wed, 23 Jan 2002, Justin Erenkrantz wrote: > This type confusion just seems horrific. I don't buy into this > distinction at all. An apr_off_t represents an offset into a file. > apr_size_t should be the size of the bucket/brigade. If we want > to do it correctly, I'd say that brigade lengths should be 64-bit > ints (apr_int64_t).
The apr_size_t versus apr_off_t distinction has always been a point of internal contention for me. On the one hand, I see the need to support large files cleanly. But size_t buckets in off_t brigades doesn't seem particularly clean to me, either. Sigh. Just to stir things up, here's an idea. [disclaimer: haven't thought this all the way through, but it seems somewhat reasonable off the top of my head] What if we set the bucket -and- brigade size limit to be a size_t's worth. If you want to send a large file down the filter stack, you make multiple calls down the filter stack, one apr_size_t's worth at a time. Thoughts? --Cliff -------------------------------------------------------------- Cliff Woolley [EMAIL PROTECTED] Charlottesville, VA
