Re: [xsl] Document() function: GET vs. POST

Subject: Re: [xsl] Document() function: GET vs. POST
From: "Joe Fawcett" <joefawcett@xxxxxxxxxxx>
Date: Sat, 12 Nov 2005 11:40:15 -0000
----- Original Message ----- From: "Colin Paul Adams" <colin@xxxxxxxxxxxxxxxxxx>
To: <xsl-list@xxxxxxxxxxxxxxxxxxxxxx>
Sent: Saturday, November 12, 2005 11:24 AM
Subject: Re: [xsl] Document() function: GET vs. POST



"Joe" == Joe Fawcett <joefawcett@xxxxxxxxxxx> writes:

Joe> Well normally you'd use a separate protocol to specify that a Joe> post was needed. document("httpPost:://.....

I don't think that can be right at all.
Firstly, the word before the colon (not a double-colon as you typed),
is a URI scheme, not a protocol. The two do not map one-to-one.
Not even for http, necessarily.

Secondly, inventing new URI schemes is not a good idea as it decreases
interoperability.
--
Colin Adams
Preston Lancashire

Fine ,I'll start correcting every typo and instance of poor grammar that appears on this list.

What I was saying is that if you use a custom resolver you can specify a new scheme as well as override an old one. Whichever way it is done it's obviously going to decrease interoperability.

Joe

Current Thread