[Edbrowse-dev] Refresh http header

[Edbrowse-dev] Refresh http header

[Adam Thompson]

[-- Attachment #1: Type: text/plain, Size: 761 bytes --]

Hi,

How difficult would it be to add support for the pseudo-standard Refresh http
header in edbrowse? As far as I can work out this works exactly like the meta
Refresh mechanism, just in the http response headers.

I only ask since a number of websites,
most importantly for me my isp's internet monitoring pages,
use this header in place of conventional redirects.
I'm not sure why some people prefer it,
but it's supported by the major browsers.
It'd also be good if this mechanism (and the meta equivalent)
were disableable using the hr command if they're not already (I don't use this
command often thus can't remember).

I'd have a look myself, however I've got various bits of university work to do
as well as revising for my final exams.

Cheers,
Adam.

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 473 bytes --]

[Edbrowse-dev] Refresh http header

[Karl Dahlke]

> Refresh: 30;url=http://the-brannons.com/

Sure, but, this is I think exactly the same header I use for code 302.
I guess you're saying sometimes we get this header with code 200,
and are suppose to jump on that.
Well I'm not looking for that, I'm watching the codes first,
but I might be able to check for the header even on code 200.
Not sure if I should infer in this case code 301, 302 or 303.
I'll let you know.

Karl Dahlke

Re: [Edbrowse-dev] Refresh http header

[Adam Thompson]

[-- Attachment #1: Type: text/plain, Size: 1043 bytes --]

On Tue, Apr 22, 2014 at 02:14:32PM +0000, Karl Dahlke wrote:
> > Refresh: 30;url=http://the-brannons.com/
> 
> Sure, but, this is I think exactly the same header I use for code 302.
> I guess you're saying sometimes we get this header with code 200,
> and are suppose to jump on that.

My understanding is that with codes 301,
302 and 303 the url is provided in the Location header rather than this one.
As Chris says, this is only a pseudo-standard header created by netscape which has been supported
by all the big browsers and unfortunately some sites rely on it.


> Well I'm not looking for that, I'm watching the codes first,
> but I might be able to check for the header even on code 200.
> Not sure if I should infer in this case code 301, 302 or 303.

As I said this should behave exactly the same as the meta refresh mechanism does apparently.
As for whether the html or http refresh header should be used if both are present,
I'd probably go with the html and display a warning if they differ.

Cheers,
Adam.

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 473 bytes --]

[Edbrowse-dev] Refresh http header

[Karl Dahlke]

Is this http codes 301 and 302?
If so then it is already implemented.
If it is some other construct then I guess I don't know about it,
but it probably wouldn't be difficult to implement,
if I could just pretend like it was http code 302.

Karl Dahlke

Re: [Edbrowse-dev] Refresh http header

[Chris Brannon]

Karl Dahlke <eklhad@comcast.net> writes:

> Is this http codes 301 and 302?

No.  This is an HTTP header created by Netscape years ago, and it has
never been included in the standards!
It serves the same place as a <meta refresh> tag in HTML.  These two
things are equivalent:

Refresh: 30;url=http://the-brannons.com/
in HTTP headers, and
<meta http-equiv="refresh" content="30; url=http://the-brannons.com" />
in HTML.

We could extract the header in libcurl and then pass it along when
parsing the HTML, I suppose.  Here's the question.  Which takes
precedence: the Refresh header or the meta tag?

-- Chris