The Un-Official Proxomitron Forum

Full Version: No menu and other issues on Scout site
You're currently viewing a stripped down version of our content. View the full version with proper formatting.
While trying to discover why my normal filters don't seem to be working on this forum site, I've discovered that the Proxomitron menu doesn't show up. In fact reloading the page with ?prx-command=dbug.. shows no matches. The view source seems to be missing much of what is on the page. Firebug shows all the html elements, but "view source" only reveals part of it. How is this page getting by Proxomitron? Is it correctible? The log window does show many matches.

using sidiki 10-23-10, firefox 3.5.8, win xp pro
Firefox 5 appears to be getting an xml version of the page.
The Proxomitron's log window shows "Content-Type: text/xml; charset=utf-8".

A solution might be to fake the user-agent. An Opera UA gets "Content-Type: text/html; charset=utf-8".

[^/]   $SET(0=f_ua_op.)

I added it to User-Agents, but everything still looks the same. No menu. No dbug matches. I do still see the text/xml in the first log response. I thought proxomitron could filter anything from http. Is faking an Opera browser the only way to do it?
Sorry, add it to Exceptions-U.

xml is one of the content types for which filtering must be enabled, .

Also, content types for which filtering must be enabled should not send the "accept-encoding" header. Unfortunately, the Proxomitron does not always decompress filter forced types and only decompressed files are filtered.

I think, Sidki's set will do this for you.

## ============================================================================
## 4.2  OTHER
## ============================================================================
## In this section you can add entries which control filtering directly,
## rather than applying a keyword or flag.
## Examples:
##   Force filtering of an RSS feed:
##*type\=atom $SET(hRealCT=filter)$FILTER(1)

I think I'd try faking the user-agent first.
Putting it in Exceptions-U worked. Menu is inserted and the html can now be filtered. Faking the user agent is fine for my current purpose. But what if the site only served xml? Or is this method rare and unlikely to be an issue with others? I read the section on $FILTER but am not sure how to force xml filtering other than just putting $TYPE(xml) in the URL Match box. Is that insufficient or different from what seems to be an option in sec 4.2? It's not evident to me how xml could be forced like RSS in 4.2 because I'm not sure what some of that means.

(Jul. 09, 2011 01:27 PM)JJoe Wrote: [ -> ]I think I'd try faking the user-agent first.

Probably best for me. I had looked earlier at the xml files and couldn't grasp them in the structural sense that Firebug presents for the page, so it seems like it would be difficult to work with.
I no longer allow Fx 4 to identify as Fx. I have it set to identify as Chrome 11. I use the extension User Agent Switcher. Great extension! I downloaded a list of user agents from the author's forum.

I have less problems with Fx4 (including the one you describe) if I have Fx identify as Chrome. Upon closing Fx, the extension automatically reverts back to identifying as Fx because of a bug in Java (which causes Fx to crash upon starting if not identifying as itself). So, user agent has to be set each session from Tools drop down list. You can switch user agent for Fx at any time, and switch back at any time ,so you can just switch right before you go to a site using xml instead of html and switch back as soon as you are through with that site. Unfortunately, Chrome is now the darling of the internet so better to just identify as Chrome all the time I have found.
(Jul. 10, 2011 03:08 AM)zoltan Wrote: [ -> ]Or is this method rare and unlikely to be an issue with others?

Let's hope so. I didn't expect to see all that xml and for just Firefox.

(Jul. 10, 2011 03:08 AM)zoltan Wrote: [ -> ]I read the section on $FILTER but am not sure how to force xml filtering other than just putting $TYPE(xml) in the URL Match box.

$TYPE limits filtering and $TYPE(xml) is not one of the choices.

htm - Web pages
css - Cascading style sheets
js - JavaScript
vbs - VB Script
oth - Anything else

With sidki's set

Code:  $SET(hRealCT=filter)$FILTER(1)

should force filtering of files from .
Oh, that was easy enough. I didn't realize the $SET(hRealCT=filter)$FILTER(1) example wasn't for RSS only.

As for the add-on switcher, I'll keep that in mind as an option, but I'd prefer to have the agent faked automatically per site (as Proxomitron seems able to do) rather than doing it manually each time.

I'll try faking Chrome, however, just to experiment with the behavior. I've noticed that with my customized config, what works fine in firefox, does not with chrome on 2 sites (youtube and google). I won't switch unless forced though because I hate Chrome's UI.
Reference URL's