Re: Making POST (and PUT) requests
To: Henrik Frystyk Nielsen <email@example.com>
Subject: Re: Making POST (and PUT) requests
From: Erik Selberg <firstname.lastname@example.org>
Date: Tue, 23 Jan 1996 16:46:03 PST
From email@example.com Tue Jan 23 19: 46:05 1996
In-Reply-To: Henrik Frystyk Nielsen's message of Tue, 23 Jan 1996 18:36:50 -0500
>> second being in HTMIME_put_block which causes anything using the
>> content_length field (like POST requests) to stop reading after
>> you've _read_ content_length bytes (the content_length is used for
>> telling the server how many bytes you're writing). I'm assuming Henrik
>> will have patches for these in a day or so.
> This is a problem as long as you use the same anchor for both data objects. By
> using two anchors - one that represents the form data which you keep in memory
> and one which represents what you get back from the remote server then this is
> not a problem. This is how it works while using PUT or POST from a local file,
> for example and this is how I imagined it to work on memory buffers as well.
I got lost somewhere here. Sure, two anchors are great, but I'm
confused as to where the anchors you mean are located. For instance,
the current (well, 4.0B :) library seems to want to use the
request->anchor field for both the outgoing request as well as the
incoming, at least in relation to POST requests. Which are the actual
anchors you had in mind?