RE: [Kannel 0000010]: HTTP requests get urlencoded twice (orevenmore) if HTTP 302 (redirect) responses are received

"Rene Kluwen" <[email protected]>
Newsgroups gmane.comp.mobile.kannel.devel
Message-ID <[email protected]>
I think you answered your question yourself :).
Yes, the querystring should be included in subsequent requests.

-- Rene...

-----Original Message-----
From: [email protected] [mailto:[email protected]]On Behalf Of
Bruno David Rodrigues
Sent: dinsdag 25 maart 2003 17:08
To: [email protected]
Subject: RE: [Kannel 0000010]: HTTP requests get urlencoded twice
(orevenmore) if HTTP 302 (redirect) responses are received


Citando Rene Kluwen <[email protected]>:

> About 302 (temporary) redirecting responses:
>
> After a GET request, the client (kannel) should redirect the SAME
request
> to the URL pointed to in the Location: header.
> This means, the query string should not re-urlencoded.

Section 3.2.2 says that URL includes the querystring, so going back to my
first
question - should the original querystring be added to the new location ?

Imagine this:
original URL: http://foobar/foo?phone=%p&text=%a
Location header: http://barfoo/bar?foo=true&bar=false

Which URL should kannel use ? Simply the location ? should it add
"phone...%a"
to the second URL (http://barfoo....bar=false&phone=%p&text=%a) ?
I don't see anything relevant in rfc :(


> After a POST request, the same request should be POSTed the the new URL.
> Specs tell that this should require user interaction, "since this might
> change the conditions under which the request was issued". But I figure,
> this will be a little hard, as Kannel is a daemon.

No, it says "MUST NOT go without user intervention". As kannel is a
daemon, it
shouldn't follow locations if original method is a post. Unless we have a
config
value "yes-i-ve-read-rfc-and-i-want-to-follow-post-redirections = true" ;)


> The above also goes when the status code is 301 (Moved permanently). But
> Kannel should save the new URI's and use them for future connections.

Should we have a url cache system ? Should we bother with it ? does IE do
it ? ;)


> The 303 status (See Other) is an exception. The new URI given in the
> response should be retrieved using a GET method.

This one is cute. You do your job at first location then server replies
"thanks
for your job, now go there for more information"

> -- Rene...

Thx


--
<br/>
 15:56:22 up 122 days, 17:10,  1 user,  load average: 0.00, 0.01, 0.41
BOFH excuse #172:
pseudo-user on a pseudo-terminal
smime.p7s (application/x-pkcs7-signature, 3 KB) - not displayed
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.