On Sat, Nov 15, 2014 at 12:40 AM, Jeff Trawick <[email protected]> wrote: > Summary: I don't think it is a big deal in practice but I think it can be > confusing that the code that just needs to determine if the end has been > found can discover errors that otherwise would be unnoticed.
I agree, the checks belong to ap_scan_script_header*(). Maybe we need (yet) another ap_scan_script_header_parse(), reentrant (optionaly taking care of folding), and returning a relevant value for the caller to know. All cgi handlers (by the existing functions) would use it. I will undo these change for now and let all the data reach ap_scan_script_header_err_brigade_ex(), that's indeed better. Likewise wrt r1638818. Thanks for the review, Yann.
