Gesturefy: Gestures aren't working on https://addons.mozilla.org/

[ryanbr]

Gestures aren't working on https://addons.mozilla.org/ or any extension pages like https://addons.mozilla.org/en-US/firefox/addon/gesturefy/ It seems to just ignore any gestures.

[Robbendebiene]

I also stated this on the mozilla addon page. Addons are denied to inject scripts into the mozilla addons page. (btw I have no idea why..) Same problem as described here: #5

[ryanbr]

Okay another url, https://upload.wikimedia.org/wikipedia/commons/5/5d/PI_constant.svg gesture doesn't work here :/

[Robbendebiene]

Thank you. I completely forgot about svgs, but it should be possible to fix this.

[Robbendebiene]

It looks like I was a bit too optimistic about that..
The short conclusion is: it's not or only partially possible without changing the svg.

In detail: Normally I append a container to the websites body which then lies above all the content. This gives me the possibility to draw everywhere on the screen what I want.

For pure not embeded svgs however there is no such thing as a body. The root element is called "svg" and normally does not allow to append any kind of html elements nor does it support the canavas (the api I use for drawing the gesture). Nevertheless for both problems there is kind of a solution.

For the first problem exisits a foreignObject which I scaled to the screen size. Even though this won't work perfectly if the svg is somehow postponed.

let container = document.createElementNS('http://www.w3.org/2000/svg',"foreignObject");
container.setAttribute("width", window.innerWidth);
container.setAttribute("height", window.innerHeight);
container.appendChild(canvas);

Second Problem: Here I created a new HTML document object which then allowed me to create a canvas object, because xml/svg would just create a normal element named canvas without its funcionality.

let canvas = document.implementation.createHTMLDocument().createElement("canvas");

However the area where I can show elements is still limited to the svgs size. So for your example it's a container of 440 x 100 pixes. I could change the sizes to the maximum window size and put the displayed svg elements in a down scaled group. But this would mean that I have to change the svg element, even though it wouldn't be visible untill the user downloads the svg.

Another possibities would be to completely remove/clear the current document object and create a new html document with html, head, body and append the svg element to the body. drawback: when the user wants to save the svg it will then save an html document.

I have to admit, that I like none of the solutions, because in both I have to modify the original svg :|

[sorz]

I think gestures without display in svg is acceptable.
Actually I has disabled all path/action/... display functions. :)

[happysurf]

Big news about that, now is sufficient create this boolean string:
privacy.resistFingerprinting.block_mozAddonManager and then set to true.
Now all extensions works on Mozilla websites.

Tip details here:
https://www.ghacks.net/2017/10/27/how-to-enable-firefox-webextensions-on-mozilla-websites/

[Robbendebiene]

Closed in an issue clean up process due to the lack of an appropriate or upcoming API.