- From: Henrik Frystyk Nielsen <frystyk@w3.org>
- Date: Wed, 14 Jul 1999 10:55:30 -0400
- To: Marco Ariano <marco.ariano@pipercreek.com>
- Cc: www-lib@w3.org
At 14:28 12/07/1999 -0600, Marco Ariano wrote: >I don't dispute your claim about states and SSL but... > >I thought I'd tried things with and without the call to >HTEventList_loop() with no difference. >Since I had WWW_WIN_ASYNC defined, I get some kind of event loop as a >side effect of calling > HTProfile_newPreemptiveClient( "Secure_Client", "1.0" ); >anyway don't I? No, the preemptive version uses blocking sockets where you don't need an eventloop. The equivalent profile for non-blocking sockets is called HTProfile_newClient( "Secure_Client", "1.0" ); I still think this may be the reason why you are only getting the first chunk of the data (if I understood you correctly). Henrik >Here's why I think there's something not quite right. The work-around >I've decided to use for now is to call a modified version of >HTLoadToFile() (since that worked). I've ripped out the file >confirmation stuff and just write to "get.out" every time. I replaced >the call to HTLoadAbsolute with HTGetFormAbsolute since that's what I >really want to do, but only after I convinced myself that HTLoadAbsolute >was working. The only real difference that I see between doing this and >calling HTGetFormAbsolute in the first place is the change to the stream >stack. Without the change, things don't work. > >I tried digging a bit deeper for a while but time constraints forced me >to use the work-around. > >thanks, >Marco > -- Henrik Frystyk Nielsen, World Wide Web Consortium http://www.w3.org/People/Frystyk
Received on Wednesday, 14 July 1999 10:55:38 UTC