Closed Vangelis66 closed 1 year ago
martok
commented
1 year ago selector-observer doesn't seem to spot the elements in one place (mouseenter, issue-links.ts ), but it works in another (mouseover, mouseleave from hovercards ).
First guess is a timing difference. For once, it's pretty clearly not a webcomponents problem, there are none anywhere in that tree :)
Vangelis66
commented
1 year ago GitHub docs entry for the "links" manifesting this regression in palefill-1.24:
Also, what I called useful popup in the issue's title is technically referred to as hovercard ...
(I wasn't aware of the exact terminology when I opened this issue, luckily I was close enough 😜 ...)
jrfnl
commented
1 year ago Came here to report the exact same thing. Also noticed the mouse-overs when hovering over a contributor's name don't show anymore.
Using PaleMoon 31.4.1-win64.
In the console, the following message shows on page load, but may very well not be related:
Content Security Policy: Directive ‘child-src’ has been deprecated. Please use directive ‘worker-src’ to control workers, or directive ‘frame-src’ to control frames respectively. 1 (unknown)Content Security Policy: Directive ‘child-src’ has been deprecated. Please use directive ‘worker-src’ to control workers, or directive ‘frame-src’ to control frames respectively. 1 (unknown)
It's not related, that's been there forever. Looks like it was un-deprecated and UXP doesn't know that. "Living standards", yay.
jrfnl
commented
1 year ago Thank you @martok for working on this! I look forward to testing it.
nicolaasjan
commented
1 year ago Vangelis66
commented
1 year ago I can confirm 🎉 !
It probably serves nothing now 😉 , but yesterday night I did some git-bisection with regards to this issue and found out that:
code snapshot palefill-v1.23-3-git-20221107-g2e6e6f8 is the "last GOOD", while
immediately next code snapshot palefill-v1.23-4-git-20221201-g677dfa9 is the "first BAD" 😞 ;
thus, 677dfa9 was responsible for the regression (and not DarkTree per se); which coincides with 395dbc6 as being the fix (the former removed GH polyfill github-compat-838cedbb.js, while the latter re-introduced some "needed" parts of it) ...
Much appreciated martok ! 😄
Extension version:
palefill-v1.24Description
I was elated that the new update fixed #60 (:+1: ), but soon my joy was somewhat dimmed :disappointed:, because I stumbled upon a new regression (adding to #64) on a GitHub feature I extensively use...
STR
#60(or#30; these are issues [hopefully] resolved by the commit).Expected behaviour:
A popup window (hovercard) is being displayed, with details about the issue (or PR, if the hovered "number" corresponds to a PR):
Above screengrab is with a Chromium 86 fork (similar result with
palefill-1.23, minus the relative timestamp)Actual behaviour:
No popup window is being displayed - I have to middle-click on
#60and open the linked issue in another tab, to see what it's about :disappointed: ...With a commit:
2a9bc64Expected behaviour:
A popup window is being displayed, with details about that commit:
Above screengrab from Chromium 86 fork (similar result with
palefill-1.23, minus the relative timestamp).Actual behaviour:
No popup window is being displayed - I have to middle-click on
2a9bc64and open the linked commit in another tab, to see what it's about :disappointed: ...Hopefully I'm not coming across as ungrateful and/or entitled :wink: , I'm not a coder but (I think) I've formed a good idea of how much time-consuming all those "fixes" must be, as a result the maintainer's (and other contributors') efforts are indeed highly appreciated by me :heart: ...
For #64, a "workaround" was cited, hopefully something will come up for this issue, too... :smiley_cat:
Addition: As spotted later by @jrfnl :
Addition 2: Clarification about terminology in below (yet hidden [?]) comment...