eamondaly wrote: > > I'm trying to add a cleaner to Script_OnFlush so that our pages > will validate properly at http://validator.w3.org/: > > sub Script_OnFlush { > my $data = $Response->{BinaryRef}; > > # This is experimental. Allows our docs to validate HTML 4.01 by > # converting '&' to '&' in hrefs. > > $$data =~ s{href=["']([^"']+?)["']}{ my $url = $1; > $url =~ s/&/&/g; > qq{href="$url"} > }sgie; > } >
If you look at the Apache::ASP::Response::Flush routine, you will find that Script_OnFlush gets called BEFORE these ASP Flush() fixup actions: XSLT FormFill Clean SessionQueryParse CompressGzip SendHeaders --- write output --- While, having it run before is fairly arbitrary, it seems to make sense that all of the above actions would happen after Script_OnFlush() so they could also work on your Script_OnFlush output. What you want is the final event after the above translations ( well before Clean and CompressGzip I imagine. ) We might call this: sub Script_OnFlushFixup() so the order would be: Script_OnFlush() XSLT FormFill SessionQueryParse Script_OnFlushFixup() *** Clean CompressGzip SendHeaders --- write output --- What it sounds like you really want though is for SessionQueryParse to handle & "correctly" and have it auto escape & to & ... my only concern with this is I do not know the browser compability issues with this if there are any... do ALL browsers treat & in URLs correctly as & when the URLs are in HTML tag attributes? I know some browsers do, but I am concerned with the prospect of changing the behavior for what seems to work fine already. -- Josh _________________________________________________________________ Joshua Chamas Chamas Enterprises Inc. NodeWorks Founder Huntington Beach, CA USA http://www.nodeworks.com 1-714-625-4051 --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
