rik / au-revoir-utm Goto Github PK

Hello,
I've got a problem on the website fredzone.com
I receive articles by rss with an url like this http://www.fredzone.org/la-terre-a-t-elle-ete-ensemencee-par-des-particules-extraterrestres-554#utm_source=feed&utm_medium=feed&utm_campaign=feed and the page don't load. But without utm_source, it's working.

An idea to fix that ?

Open this site in Firefox

with UTM turned on:
http://www.independent.co.uk/voices/41-per-cent-trans-transgender-trans-women-prisoners-sex-offenders-false-study-statistic-this-is-why-a8072431.html

Notice that CPU and memory both shoot up.

It seems to be trying to make requests that are then failing, as part of the general run of the page. I've attached an example.
ExampleRequest.txt

From looking at the source (3 lines of code, but whatever) I can tell that this addon won't work as expected. It only removes the utm stuff if the query string starts with an utm parameter (i.e. like "http://something.com?utm_...", but not "...?page=main&utm_..."), and it removes ALL parameters from the URL, not just the utm params.

Please fix this, as it seems like a quite useful addon otherwise.

the link in this issue #1
the Advertising link on the bottom left when you visit https://www.google.com

and so on.

the firefox addon simply does not clean any links. I disabled every other extension to test this.

Hi.

It seems that there is no license information. A LICENSE or COPYING file would be useful to know the license easily.

Regards.

Hi, first-off, great add-on, super useful!

Just a small bug I found - when opening links such as the following in Firefox (which are created by Nuzzel.com's newsletters for example), then au-revoir stops that page from ever loading. When disabling au-revoir the links load normally.

The issue appears to be the last part of the URL that Nuzzel adds- "%26utm_medium%3Demail%26utm_source%3Dnuzzel"

When this string is removed from the URL, it opens as normal.

I.e., http://www.example.org/?%26utm_medium%3Demail%26utm_source%3Dnuzzel

When replacing "%26" with "&", then the link works as normal, even with au-revoir enabled.

Anyone else also experiencing this or any thoughts on what in au-revoir could be causing this?

Can the addon be included to remove other tracking parameters ?
On mozilla, for the MHT archive addon there so many variations

https://addons.mozilla.org/en-US/firefox/addon/mozilla-archive-format/?src=search
https://addons.mozilla.org/en-US/firefox/addon/mozilla-archive-format/?src=hp-dl-featured
https://addons.mozilla.org/en-US/firefox/addon/mozilla-archive-format/?src=cb-dl-featured
and so on ...

On PCWORLD

https://www.pcworld.com/article/257569/the_roots_of_social_networking.html#tk.hp_fv
https://www.pcworld.com/article/257569/the_roots_of_social_networking.html#tk.hp_ess
https://www.pcworld.com/article/257569/the_roots_of_social_networking.html#tk.hp_new

Hello,

The tracking stuff is not removed on Ulule, example with this link :
https://fr.ulule.com/macronie2/?ul_campaign=presale_83971&ul_source=shared-from-Ulule-mail-order.completed.user-on---http.referer--&ul_medium=uluid_123456-unknown-201901011010

Possible to add it ?

Thank you

Original link: https://geektimes.ru/post/298161/?utm_source=habrahabr&utm_medium=rss&utm_campaign=298161#habracut

Correct link: https://geektimes.ru/post/298161/#habracut

But the addition adds an extra sign equally after the hash.

au-revoir-utm 1.1.0

When I enabled this extension, e10s is disabled. Is there anything to change in the extension to keep using it with e10s?

Thanks

Hello. The current version of the addon fails to work properly on this URL.

The URL will just load without removing any tracking:

https://vuejs.org/v2/guide/components-edge-cases.html?utm_campaign=Vue.js%20News&utm_medium=email&utm_source=Revue%20newsletter#Programmatic-Event-Listeners

Reloading does not change anything. But what does help is to reload without cache (i.e. by hitting shift+cmd+R). By doing that, all utm parameters will be removed properly:

https://vuejs.org/v2/guide/components-edge-cases.html#Programmatic-Event-Listeners

This might be a shot in the dark, but could this issue be related to Service Workers?

Can the addon be modified so as to prevent tracking ?

Currently, your add-on produces a lot of error messages in the error console on every page load.

Error: The Components object is deprecated. It will soon be removed.
Source file: resource://jid1-yrtlfkrtmbratq-at-jetpack/api-utils/lib/cuddlefish.js -> resource://jid1-yrtlfkrtmbratq-at-jetpack/api-utils/lib/sandbox.js -> resource://jid1-yrtlfkrtmbratq-at-jetpack/api-utils/data/content-proxy.js
Line: 7

Error: Exposing chrome JS objects to content without exposedProps is insecure and deprecated. See https://developer.mozilla.org/en/XPConnect_wrappers for more information.
Source file: resource://jid1-yrtlfkrtmbratq-at-jetpack/api-utils/lib/cuddlefish.js -> resource://jid1-yrtlfkrtmbratq-at-jetpack/api-utils/lib/sandbox.js -> resource://jid1-yrtlfkrtmbratq-at-jetpack/api-utils/data/worker.js
Line: 130

This is coming from an old version of the Add-ons SDK. Please repack your add-on with the newest version (currently 1.12) and reupload to addons.mozilla.org.

Currently, au-revoir-utm has a listener installed on all requests through onBeforeRequest - it then searches for utm_ in the URLs before deciding to redirect. Am I wrong should au-revoir-utm only care about HTML documents?

onBeforeRequest.addListener() allows to filter by types. You could specify main_frame and sub_frame in there, and I think that would be enough to cover all HTML documents and improve performance, avoiding calls to the add-on for requests it does not care about.

Like this: http://bigbrowser.blog.lemonde.fr/2012/09/05/langue-morte-comment-dit-on-minijupe-en-latin/#xtor=RSS-32280322?utm_source=twitterfeed&utm_medium=twitter

When au-revoir-utm is activated in my Nightly, it causes the console to say:

The Web Console logging API (console.log, console.info, console.warn, console.error) 
has been disabled by a script on this page.

Which, you will agree, is quite annoying when you're working on the Web. :)

It seems that the ref query parameter is also generally used for marketing/tracking campaigns, as a "utm-like parameter".

Also generally I wonder what is the effectiveness of the method offered by this plugin, as utm_ parameters, start with this prefix (as query parameters of a URL), only by convention.

So tracking from URL parameters, can still probably be done, by using any other name of parameters (non utm_ prefixed)?

It's a pretty cool project and implementation in any case (and nicely cleans urls!),
cheers!