Fix Firefox version check #1363
Conversation
…ic comparison > 99 Signed-off-by: Kenneth T <6724477+kennethtran93@users.noreply.github.com>
Signed-off-by: Kenneth T <6724477+kennethtran93@users.noreply.github.com>
kennethtran93
added
Firefox
Codecov Report
@@ Coverage Diff @@
## 3.X.X-Branch #1363 +/- ##
=============================================
Coverage 40.99% 40.99%
=============================================
Files 34 34
Lines 2059 2059
Branches 330 330
=============================================
Hits 844 844
Misses 900 900
Partials 315 315 Continue to review full report at Codecov.
|
This comment was marked as outdated.
This comment was marked as outdated.
This comment has been minimized.
This comment has been minimized.
|
/get-build 2142469342 |
1 similar comment
|
/get-build 2142469342 |
|
It seems like the bot isn't behaving. |
This comment was marked as outdated.
This comment was marked as outdated.
|
No fix I'm afraid.
|
|
Same here, the problem persists. |
Signed-off-by: Kenneth T <6724477+kennethtran93@users.noreply.github.com>
This comment was marked as outdated.
This comment was marked as outdated.
|
clearly my CI workflow needs some updates as well. In the meantime, here's the manually compiled version for Firefox, from the recent commit (more parseInt) I did (and probably didn't need to but just in case). |
|
Sorted.
Thanks mate :) |
|
When would be the release of this? 👀 |
|
I would like to release as soon as I can, but I need to also implement a privacy user-consent first to keep up with the recommended web extension workflow. |
|
I've managed to install CAD 3.999.999 as an xpi. Works perfectly well. So happy to have localStorage and IndexedDB handled again within FF100. Not blocked by Firefox 100 because Not blocked but I do get of course in
When do you intend to release CAD 4.0? 3.999.999 works perfectly well, and I can imagine the number of users of FF100 being desperate! Thanks for this invaluable extension. |
|
It won't be a 4.0 yet. I use 3.999.999 for Dev and test works. It will probably be 3.7.0 or so. |
Good to know. That means installed 3.999.999 won't be updated by Firefox. I'll have to backup data/remove it first. |
FYI while a backup would be wise before and after the beta test, I was able to downgrade by simply removing the addon and installing the official version, and the existing data worked fine. |
|
I won't be advised by Firefox add-ons update that CAD 3.7.0 or so is available if I'm running CAD 3.999.999, which means that
either, considering
OK. But given CAD_CoreSettings and CAD_Expressions.json are easily importable and complete, given I'm a clean install maniac, I might follow my "academic" policy :=) though one thing to remember when importing CAD settings : import Expressions before CoreSettings, because if the latter has been set to Thanks for the information @xcasxcursex |
Did you have to use Firefox Nightly or did you find another way? I've been waiting for the updated version, but think I want to use this dev version instead of waiting. I noticed a bunch of websites remembering me and found it was because this extension wasn't working. |
|
I proceeded with Firefox 100.0, in fact because it's that version which revealed the CAD issue. What I had done: Downloaded from AMO 1- I'm no expert, not sure 3- I think CAD, independently of the Firefox's 3-digit compatibility issue, is not a 100% efficient with Firefox's Enhanced Tracking Protection given this feature auto-sets We then have as shown in my about:addons screenshot CAD 3.999.999 installed with the browser's warning yet fully functional on Firefox 100. Hope that helps. |
|
@kennethtran93 I'm sorry but I've held my tongue for long enough that I've reached a point where this needs to be said. I really don't think you should be putting in work for #754, while the plugin is non-functional for the entire firefox userbase. This kind of thing is EXACTLY why that kind of thing gets downvotes like it did. Your time is valuable and we don't want to see it lost to functionality of the plugin, in favour of a political agenda. I know you've already delayed that thing by a long, long time.... But I really feel like you ought to delay it a little longer. Late edit: and just for what it's worth, the reason I'm discussing this is because I thought maybe I should offer to help with CAD, so I decided to look at what's being worked on right now, to see if I would be of any use to you... And that is what I found. I'm still happy to help out, so long as it isn't with that, because I believe that action detracts from the fight against racism. |
|
Auto-updated to 3.70, 267 sites cleaned, seems to be working now :) Thanks! |
|
@Cade66 You can install an extension via the zip file as well. No need to extract something from the published version to put into the 3.999 version. Of course, the As for your 3rd point, we are still continuing to look into this and will hopefully have something in the near future. |
|
@xcasxcursex at this point in time, the PR that is linked to that request needs to be redone anyway. I foresee tons of reworkings on the front and back end, and just renaming the lists to something more action based may help as we try to implement either custom lists and/or more lists with specific actions. If you are still up for assisting with this project, regardless of the aforementioned item, there are still other issues and feature requests you can start to tackle and submit a PR for us to look at (like perhaps getting some dark styling going or pointing us to that direction). |
OK, and thanks for this precision. It was the first time I was dealing with an un-signed and moreover zipped extension so I sort of improvised.
What I had related in my 3rd point was my personal experience confirmed by experts' savvy explanations I've read occasionally : Firefox's Enhanced Tracking Protection is not a panacea yet is beneficial for a wider audience than that focusing on First-Party Isolation (FPI). FPI was and remains TOR's approach and remains the wisest and most efficient one, IMO. I took YouTube as an example but other domains have this "elevated privilege" within Enhanced Tracking Protection which is that their IndexedDB data --- whatever isolated it is, and it is -- is isolated to the point of not being removable, even with CAD, even with other cookie managers such as the Cookie Quick Manager extension: they are preserved. What's the deal? Take the following scenario: we want to keep only some of YouTube's cookies, not localStorage, not IndexedDB, and great: that's what CAD is done for. We allow cookies for Youtube.com (they'll remain after FF restart) and set CAD accordingly only for some of its cookies:
My experience is that, with Firefox's Enhanced Tracking Protection (browser.contentblocking.category = "strict" together with network.cookie.cookieBehavior = 5) CAD cannot delete Yutube's IndexedDB, but it does perform perfectly with browser.contentblocking.category = "custom" together with network.cookie.cookieBehavior = 1. The latter would be nonsense of course without Firefox's Enhanced Tracking Protection counterpart which is First-Party Isolation enabled. I'm not sure CAD will ever be able to defeat Firefox's Enhanced Tracking Protection site isolation when it comes to domains granted with this elevated privilege I was referring to, an approximate terminology for a number of concerned domains of which I haven't managed to find the exact list. We have of course installed CAD 3.7.0, signed!, and I'll really emphasize on the quality of this Recommended by Firefox extension. |
How i can enable these options? |
|
@KamiOh I'm aware of that, but thanks. |
@KamiOh those are specific cookies that the aforementioned user has explicitly kept but removed the rest. If you are looking to keep all cookies, then keep your expression as is, otherwise uncheck the 'keep all cookies' checkbox to see a list of individual cookies for that site. |
i can't found
|
if CONSENT is not shown, this means that the site has not created a cookie with that name yet. Sites overtime may change how they store/set cookies and other data. Keep in mind that cookie names may not necessarily represent what you are looking for. If you still want to have that specific cookie name checked in your entry, there are a couple ways of doing so.
At this point, there's no plan in the near future to implement a way to manually add a (non-existent) cookie name to the list, because for most users, there is no need to. For all anybody knows, that CONSENT cookie may have been used at one point in time but not anymore, so trying to exclude a cookie from being deleted when it is not even created yet is a bit of an extra work in which it may never be created. Also that cookie may not even be used at that site as well, but created from another web extension that another person has on their browser (good example is our own cookie |
Fixes #1329.
Version number check (primarily in Firefox) is now done by number instead of string. This fixes the issue of '100' >= '79' being
falseinstead oftrue.🤦 ...I wonder why I didn't do this in the first place.