Giter VIP home page Giter VIP logo

github-wc-polyfill's Introduction

github-wc-polyfill's People

Contributors

justoff avatar seahoh avatar

Stargazers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

Watchers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

github-wc-polyfill's Issues

Argh... GitLab is broken again (regex/named capture groups)

Today must be a horrible day for everybody using source hosting sites which are hellbent on making life miserable to anyone not using Chrome®™ or a Chrome®™-compatible browser. Now GitLab is breaking again because they decided it was a good idea to deploy named capture groups on their regexes for no good reason at all.

Apparently the polyfills injected by the addon cause GitLab to load one of their scripts... which breaks because it now contains unsupported regexes. The error console logs this:

SyntaxError: invalid regexp group commons-pages.admin.topics.edit-pages.admin.topics.new-pages.groups.epics.new-pages.groups.epics.sho-59a8e774.9fc6fd37.chunk.js:1:285
	onStopRequest jar:file:///home/tomman/.mozilla/seamonkey/7b88ug1a.default/extensions/[email protected]!/bootstrap.js:430:7

Since SeaMonkey doesn't support named capture groups (yet), the entire tower of cards falls down there, and GitLab goes back to its usual behavior for anyone not using A Blessed Browser™ (i.e. everything broken, can't read anything but the first post on a issue, etc.)

When will the madness stop!? Maybe it's time to declare JavaScript and Google war criminals against mankind at the United Nations :/

Extension won't install out-of-the-box in latest (and last) official Basilisk browser

Official Basilisk browser by MCP is one this extension natively supports (along with Pale Moon & Seamonkey);

https://forum.palemoon.org/viewtopic.php?f=61&t=27845
https://www.basilisk-browser.org/releasenotes.shtml
https://www.basilisk-browser.org/download.shtml

The EoS (End-of-Service) binary officially released (64-bit only, requires at minimum Win7 SP1) has an appVersion of 52.9.2022.01.27, but because of

<em:maxVersion>52.9.2021.*</em:maxVersion>

this extension won't install there (without manually tinkering with the install.rdf file 😉 ...); thus, maxVersion should be now raised to 52.9.2022.* to allow for such an installation to proceed 😄 ...

Kindest regards.

Support for multi-process mode in Basilisk/Serpent?

First I would like to thanks you for your work.
I like using the multi-process mode of Basilisk. to do this you can set
browser.tabs.remote.force-enable ; true
in about:config options.
However it seems The extension stops working correctly if this option is enabled, ie github homepage no longer display activity, post menu no longer fills etc.
I understand that multi-process mode is still experimental in Basilisk, but this plugin would be more useful if it supported this mode. I could not find old issue about that. So if you think it is irrelevant , just close it.

To reproduce, just download the latest build of Basilisk from roytam1 and in a new clean profile, add the browser.tabs.remote.force-enable=true boolean option in the about:config panel. Install your plugin and try it, it will fail unless you disable the previous option.

Fix bootstrap typo

There's a typo in your bootstrap code:

 if (ctype.indexOf("text/html") == -1 && ctype.indexOf("text/javascrip") == -1) {

should be text/javascript

incompatible with waterfox

Waterfox classic does not create accessibility.lastLoadDate by default
without this parameter, the extension does not work.
if you create this parameter, then everything will work fine, but it will stop working multi-process
please fix it.

It does not work in the Light browser

When installing the extension in the Light browser, an error comes out.

GitHub/GitLab Web Components Pollyfill could not be installed because it is not compatible with Light 48.0.

Is there any solution? Right now I'm in Github in Light, and it does not completely charge well.

Не работает в Firefox ESR 52.9.0

Время от времени дополнение не срабатывает и на гитхабе продолжают наблюдаться серые прямоугольники вместо текста как тут например https://github.com/sunjw/jstoolnpp/tree/master/sub-proj

p5BZlHZcpmN3KaU63JJUybXR7db

p.s. Посредством этого дополнения можно решить вопрос с просмотром превью на гитхабе (кнопка превью заработала, но превью не появилось - "Nothing to preview") и аттачем файлов ? Или для этого нужны другие костыли ?

Don't use an encoded & minified blob, but use a resource instead

This would make the extension a lot more transparent to evaluate.
Adding resource:// to the CSP is 100% safe (actually marginally safer than doing the hash dance) since those are browser internals.
Local resources also don't have to be minified which makes it fully readable for anyone needing to vet/evaluate the extension before installation/approval.

Argh.. GitHub is broken again

Yup... GH has done it again.... must be driving you bonkers..

I see this error in the console:

TypeError: (intermediate value).attachShadow is not a function[Learn More]
chunk-index2-a8e52977.js:1:2276

As for symptoms:

  • Drop down on a comment (to edit the comment, for instance) not working.
  • No notification showing for a newly pushed branch on the "Pull" page.
  • etc

ShadowRoot needed?

SeaMonkey 2.53.6 Linux/x86

Sometimes (eg, Commits list, go to commit, go back) Alt+<- (Back) fails: a blue progress indicator extends across the top of the page but no content appears under the page header.

This appears in the console:

ReferenceError: ShadowRoot is not defined[Learn More]  index.js:234:12

SM developer tools can't find index.js.

I added this minimal ShadowRoot polyfill for github.githubassets.com using NoScript:

if (!window.ShadowRoot) {
  window.ShadowRoot = function (init) {
    var _doc = new Document();
    Object.defineProperty(this, 'activeElement', {
      get: function () {
        return _doc.activeElement;
      }
    });
    Object.defineProperty(this, 'fullscreenElement', {
      get: function () {
        return _doc.fullscreenElement;
      }
    });
    Object.defineProperty(this, 'innerHTML', {
      get: function () {
        return _doc.innerHTML;
      },
      set: function (h) {
        _doc.innerHTML = h;
      }
    });
    Object.defineProperty(this, 'getSelection', function () {
      return _doc.defaultView.getSelection();
    });
    Object.defineProperty(this, 'elementFromPoint', function (x, y) {
      return _doc.elementFromPoint(x, y);
    });
    init = init || {
    };
    this.delegatesFocus = init.delegatesFocus;
    this.host = null;
    this.mode = init.mode || 'open';
    return _doc.documentElement;
  }
}
if (!Boolean(HTMLElement.prototype.shadowRoot)) {
  HTMLElement.prototype.shadowRoot = null;
}
if (!Boolean(HTMLElement.prototype.attachShadow)) HTMLElement.prototype.attachShadow = function (shadowInit) {
  var shad = new ShadowRoot(shadowInit);
  var shadowElt = shad.documentElement;
  this.shadowRoot = (shadowInit.mode === 'closed') ? null : shadowElt;
  Object.defineProperty(shad, 'host', this);
  return shadowElt;
}

I've no idea how much ShadowRoot functionality is being used, but this seems to fix the problem.

Nullish coalescing operator (??) is back, with a vengeance :-(

Browser: Latest (UXP-based) Serpent 52.9.0 (2022-02-25) (32-bit)
Browser profile: Both dirty (current) and fresh/clean one (with default settings), only this extension installed
Extension version: Latest release, i.e. v1.12.16

Since about three hours ago (ca. 202203111600 UTC), GitHub are again serving up scripts containing the Nullish coalescing operator (??), which is something they also did in the recent past, as part of previous, closed, #44 ... 😡

#56 =>

e1

SyntaxError: expected expression, got '?'[Learn More]    8933-d8b0634d2391.js:49:2049

https://github.githubassets.com/assets/8933-d8b0634d2391.js

From a quick search, areas/functions affected are:

  1. Comment Previews NOT working
  2. Names of emojis missing when selecting reaction
  3. Actually reacting results in the whole page getting reloaded, ending up in the top of that page
  4. ellipsis (...) buttons (e.g. in comments list view) are NOT working
  5. GitHub favicons on tabs not respecting browser's browser.display.prefers_color_scheme about:config setting
  6. Possibly many more, I didn't search thoroughly...
OT, but maybe not...

As in the past, my Chromium 69/78 based, alternative (to UXP), browsers on my platform also broke, typing this now on the sole browser (Chromium 86 based) that currently works with GitHub here 😞 ; please, re-think the possibility of hosting this repo on a more browser-agnostic platform... 😉 )

I don't know whether this is part of another A/B test, but I'm already affected myself, being located in South-Eastern Europe... 😠

Modify minVersion

I use Serpent52, which is a fork of Basilisk, whose version number is always 52.9.0. In order to make the add-on work, the version requirements line in the install.rdf should be modified to: <em:id>{ec8030f7-c20a-464f-9b0e-13a3a9e97384}</em:id> <em:minVersion>52.9</em:minVersion>

Incompatible with Waterfox Classsic

It says "Pollyfill is Incompatible with Waterfox 56.5." which is Waterfox Classic 2021.07. Please explain how it's supposed to work and in which browsers.
Polly has a toolbar icon where we select component for a web page. I don't see an icon here.
Thanks.

A loading issue 1.0.4

Maybe this additional code is needed for bundles/webcomponents-ce.js, and why not use bundles/webcomponents-sd-ce.js ?

(function() {
  'use strict';
  var polyfillsLoaded = false;
  var flushFn;

  function ready() {
    // bootstrap <template> elements before custom elements
    if (window.HTMLTemplateElement && HTMLTemplateElement.bootstrap) {
      HTMLTemplateElement.bootstrap(window.document);
    }
    flushFn && flushFn();
    polyfillsLoaded = true;
    window.WebComponents.ready = true;
    document.dispatchEvent(new CustomEvent('WebComponentsReady', { bubbles: true }));
  }

  if (window.customElements && customElements.polyfillWrapFlushCallback) {
    customElements.polyfillWrapFlushCallback(function (flushCallback) {
      flushFn = flushCallback;
      polyfillsLoaded && flushFn();
    });
  }

  window.WebComponents = window.WebComponents || {};
  window.WebComponents.ready = window.WebComponents.ready || false;
  // this script may come between DCL and load, so listen for both, and cancel load listener if DCL fires
  window.addEventListener('load', ready);
  window.addEventListener('DOMContentLoaded', function() {
    window.removeEventListener('load', ready);
    ready();
  })
}).call(this);

Reaction tooltips missing on "Releases" page / Missing tooltips for Editor's toolbar buttons

Browser: Latest (UXP-based) Serpent 52.9.0 (2022-02-25) (32-bit)
Browser profile: Fresh/clean one (with default settings), only this extension installed
Extension version: Latest release, i.e. v1.12.15

STR

  1. Make sure you're already signed-in, as this GitHub feature is reserved for GitHub members...
  2. Load e.g. https://github.com/SeaHOH/github-wc-polyfill/releases
    In the "reactions" area, just below "Assets", you'll see that (at the time of this writing) 6 members have reacted with 👍 and another with a ❤️ emoji.
  3. Place the cursor on top of either reaction;
    expected behaviour: A tooltip containing the usernames of the members that reacted similarly:

tooltip

(the above was taken with a Chromium-derived browser)

actual behaviour: No tooltip is being displayed:

tooltip2

(St52+gh-wc-pf-1.12.15)

NB: "Reaction" tooltips are being displayed correctly on Comment reactions...

Addendum:

It later emerged that the GitHub editor's toolbar buttons are similarly affected, i.e.

  1. Scroll down this page to find the "New Comment" input field (you also need to be signed-in).
  2. Hover with the cursor over one of the editor's toolbar buttons,

c3

expected behaviour: A tooltip pops-up, explaining the function of selected button:

c4

(Chromium-derived browser)

actual behaviour: Tooltip missing:

c2

(St52+gh-wc-pf-1.12.15)

taking too long to load

entry https://github.com/AdguardTeam/AdguardFilters without logging in (when i am logged in it works fine)
choose
issues
Sort by - Recently updated
issues - closed
= error displays
unicorn
This page is taking too long to load.
Sorry about that. Please try refreshing and contact us if the problem persists.

Youtube Polyfill

i think it might be acting up, since about half a month ago youtube keeps freezing by doing things such as putting your mouse such that the little text bubble that says "search" appears, or by being on tab 1 and having youtube on tab 2(the tab bar looks such: 1 2) and then closing 1 making you enter the youtube tab, i have had to restart my computer because that locked it up, i think is the polyfill because some dude on the palemoon forums managed(after many times) to get an error message and it mointed to the polyfill
thanks in advance

github - color labels

(translation Polish>English)
I don't know if this is a good place to submit - I'm using "github-wc-polyfill"
some labels are illegible (black text on dark color)
if you find it irrelevant, just close my application
thanks for the addon "github-wc-polyfill"

(original Polish)
nie wiem czy to dobre miejsce na zgłoszenie - używam "github-wc-polyfill"
niektóre etykiet są nieczytelne (czarny tekst na ciemnym kolorze)
jeśli uznasz to za nieistotne to po prostu zamknij moje zgłoszenie
dziękuje za dodatek "github-wc-polyfill"

screenshot of https://github.com/AdguardTeam/AdguardFilters/issues
gihabbugcol

"Copy the full SHA" no longer works...

Platform: (forked) UXP
Browser: (latest) Serpent 52.9.0 (2022-02-25) (32-bit)
Extension version: 1.2.14b2
Browser profile: Fresh (with default settings), with only this extension installed

STR:

  1. Visit/load e.g. :

https://github.com/JustOff/github-wc-polyfill/commits/master

  1. Move cursor to the "Copy the full SHA" button of top commit,

CopySHA

, click button

  1. Expected behaviour: Full (40 chars) commit HASH copied to the clipboard (with a visual indication of a successful copy action)

  2. Actual result: Clipboard is empty (or contains whatever string was put there, via whatever, immediately previous, copy action)...

  3. In contrast, on a Chromium-based browser sanctioned by GitHub, this is the visual result of clicking the button in question:

CopySHA2

and the clipboard becomes populated with 1291c0573ddaa116c48386110ffe5c6a4fe751a3 ...
Nothing I could spot stands out in Web/Error Console... 😞

[regression]Closed issue #46 seems to have come back...

Browser: Latest (UXP-based) Serpent 52.9.0 (2022-02-25) (32-bit)
Browser profile: Fresh/clean one (with default settings), only this extension installed
Extension version: Latest release, i.e. v1.12.15

p1

STR

  1. Load in a tab, e.g. 8446565 (#48)
  2. Click the Load diff area; diff will display correctly 👍 .
  3. Now normally reload that same tab; depending on the time of day (i.e. completely random 😸 ), you'll either get a misplaced and disabled Load diff button like below:

p2

or a correctly aligned Load diff button, but disabled nonetheless 😠 :

p3

Since we know that the code that addressed #46 is already inside v1.12.15, and with me not being a coder in the slightest, I'd hazard a guess that some type of race-condition is taking place that prevents that code from actually functioning... Perhaps it has to be the last to run (or is every polyfill inside the extension injected asynchronously?) ?

As already noted in #46, CTRL+F5 will, again, restore things...

PS: The SeaHOH fork is also plagued by the same issue currently (at least for me, that is...).

JavaScript-Error for local Gitlab-installations with relative URL

If Gitlab is installed locally and under a relative URL e.g. https://example.com/gitlab the actual content of the page is not loaded, depending on the page, the content-area stays blank or a loading-indicator is shown forever. The JavaScript-console contains an error:

ReferenceError: customElements is not defined[Weitere Informationen]  main.5565962b.chunk.js:44:75230
	F2LA https://www.example.com/gitlab/assets/webpack/main.5565962b.chunk.js:44:75230
	c https://www.example.com/gitlab/assets/webpack/runtime.f3c0b0d6.bundle.js:1:556
	HVBj/< https://www.example.com/gitlab/assets/webpack/main.5565962b.chunk.js:44:82973
	HVBj https://www.example.com/gitlab/assets/webpack/main.5565962b.chunk.js:44:82892
	c https://www.example.com/gitlab/assets/webpack/runtime.f3c0b0d6.bundle.js:1:556
	[182] https://www.example.com/gitlab/assets/webpack/pages.projects.merge_requests.show.505b2803.chunk.js:1:71262
	c https://www.example.com/gitlab/assets/webpack/runtime.f3c0b0d6.bundle.js:1:556
	a https://www.example.com/gitlab/assets/webpack/runtime.f3c0b0d6.bundle.js:1:430
	r https://www.example.com/gitlab/assets/webpack/runtime.f3c0b0d6.bundle.js:1:293
	<anonym> https://www.example.com/gitlab/assets/webpack/pages.projects.merge_requests.show.505b2803.chunk.js:1:2
Source-Map-Fehler: SyntaxError: JSON.parse: unexpected character at line 1 column 1 of the JSON data
Ressourcen-Adresse: https://www.example.com/gitlab/assets/webpack/main.5565962b.chunk.js
Source-Map-Adresse: main.5565962b.chunk.js.map  (unbekannt)
Source-Map-Fehler: SyntaxError: JSON.parse: unexpected character at line 1 column 1 of the JSON data
Ressourcen-Adresse: https://www.example.com/gitlab/assets/webpack/runtime.f3c0b0d6.bundle.js
Source-Map-Adresse: runtime.f3c0b0d6.bundle.js.map  (unbekannt)
Source-Map-Fehler: SyntaxError: JSON.parse: unexpected character at line 1 column 1 of the JSON data
Ressourcen-Adresse: https://www.example.com/gitlab/assets/webpack/pages.projects.merge_requests.show.505b2803.chunk.js
Source-Map-Adresse: pages.projects.merge_requests.show.505b2803.chunk.js.map

Note: This is the same error as without installed extension.

Commenting on StackOverflow no longer works

I understand this add-on primarily focuses on Github and Gitlab, but I was wondering if you could also have a look at stackoverflow.com.

Recently I believe they've updated their (or third-party?) javascript code, after which adding comments and using the "Show # more comments"-button don't work anymore.
Also the preview-window for posting answers is at a bare minimum. No layout-buttons, nothing. If you'd carefully prepare your answer with Markdown syntax somewhere else, copy-paste it on StackOverflow and hit the "Post Your Answer"-button, then the answer might still be posted, but I haven't tested it.

I'm using palemoon-28.10.4a1.win32-git-20220226-ba47fad4d-uxp-c403014cb-xpmod-sse.7z and the "Error Console" reports:

SyntaxError: expected expression, got '.'
https://cdn.sstatic.net/Js/full.en.js?v=a230c14f22c0   (Line: 1, Column: 73115)
arams.delete("page"),n.searchParams.delete("tab");const s=n?.toString().replace(/#.*$/,"");window.location.href=s+"#tab-

Extension breaks on GitHub "404" pages...

Browser: Latest (UXP-based) Serpent 52.9.0 (2022-02-25) (32-bit)
Browser profile: Fresh/clean one (with default settings), only this extension installed
Extension version: Latest release, i.e. v1.12.15

STR

  1. One of my bookmarks is

https://github.com/coletdjnz/yt-dlp-dev/commits/itv-fix

FTR, that URI did have content until very recently, when the repo's owner deleted the itv-fix branch...
If you visit that URI now, you'll get a familiar "Page not found"

404 - This is not the web page you are looking for.

result...

  1. Try to access your GH account's avatar button, on the top-right, e.g. to sign out/change an account setting, etc.

expected result: The familiar dropdown should open:

p1

(Chromium-derived browser)

actual result: The dropdown informs:

Sorry, something went wrong.

p2

(St52+gh-wc-pf-1.12.15)

Errors are shown in Web Console:

TypeError: window.customElements is undefined[Learn More]  vendors-node_modules_github_clipboard-copy-element_dist_index_esm_js-node_modules_github_filt-19a0ad-b0aa79d964b0f0b0715d.js:1:14977
ReferenceError: customElements is not defined[Learn More]  element-registry-f2d13d7793cb74e9c058.js:1:231
TypeError: window.customElements is undefined[Learn More]  behaviors-df903b53ab15748a5077.js:1:4217
TypeError: window.customElements is undefined[Learn More]  vendors-node_modules_delegated-events_dist_index_js-node_modules_github_catalyst_lib_index_js-6e358f-064c9c0434d0fad1cfc8.js:1:8547

Thanks for your efforts 👍 , I feel somehow guilty, though, for hitting (and reporting) many bugs recently... 😜 😄

none sha512

on "github" pages (UXP last build ,github-wc-polyfill last)
Error Console : None of the “sha512” hashes in the integrity attribute match the content of the subresource.
please add sha512 support
is it possible to support sha512

GitHub: "Load diff" misplaced and non-clickable on many commits

Browser: Serpent v52.9.0 (2022-01-19) (32-bit)
[it's platform should be on par with last official Basilisk 52.9.2022.01.27 (64-bit)]
Extension version: github-wc-polyfill-v1.12.13

STR:

Load, e.g. , 3c2384b:

st-ld

Load diff is misplaced (but this is something already pointed out here) 😉 , but is also overlaid with the message:

Some generated files are not rendered by default. Learn more.

Sadly, it appears that now you can only click on either "L" or "o" (of the whole Load diff string) to display the actual commit content, but those two letters are superimposed by the word "not" 😞 ...
The screenshot is from a dirty profile, but I have replicated the issue on a fresh St52 profile, with only said extension installed...

Toying a little with uBlock Origin 1.16.4.30, I removed the overlay with
github.com##.width-full.height-full.left-0.top-0.position-absolute.flex-justify-center.flex-items-center.d-flex and I thus got:

st-ld-2

which now allows me to generate the diff by clicking on the "Lo" area, but after a tab reload, Load diff appears greyed-out and non-clickable:

st-ld-3

I have to disable uBO on github.com, reload the tab a second time to get a non-clickable "Lo" area (due to the overlay), then re-enable uBO and reload the tab a third time to get a clickable "Lo" area (without the overlay); this is impractical ...

My instinct tells me this issue is caused by limitations in UXP's CSS/JS parsers (can't handle well the CSS code served by GitHub), so I'm not certain it could be addressed in the extension itself, but it sure is a nuisance... 😠

Any tips to mitigate the issue (e.g., some userstyle installable in Stylem/Stylus?) would be highly appreciated!

FWIW, this is how things look on a Chromium-78-based browser:

ch78-ld

... but this is no surprise, the modern web is tailored to perfectly suit only Chromium (I consider current Firefox to be just another Chromium fork, give or take...).

As always, utmost gratitude extended to the maintainer of this extension for his unwavering efforts, especially considering current RL tensions in his neck of the woods ❤️ ...

Palemoon devs going loco - what will the alternatives be?

This is not necessarily concerning github-wc-polyfill but a more general issue or question.

I just dug through the history of the 3 main palemoon devs and I am pretty shocked,
including their behaviour on the webforum or sneakily modifying extensions without
announcing this. But, skipping these 3 epic trolls, what alternatives do we have?
Will we be forced back into the Google empire? I am still using palemoon right now
mostly due to inertia, but I am not going to waste my time with the 3 insane palemoon
core devs anymore (Tobin is the nuttiest among these three by far really).

IF you have suggestions it would be useful if you could perhaps mention them in
some of the github repositories so people can check. I am also banned from the
palemoon forum - Moonchild can not take criticism. Dude has some mental issue.

So the TL;DR: please consider linking to viable alternatives in particular aside
from chrome and firefox if you know about them. Thanks for reading.

Congratulations for this great extension

Hi JustOff,

I want to thank you personally for this great extension !
Thanks to you JustOff, github work back again !!
👍
BTW : You can close this issue because it's not a real one 😄
God bless you abundantly in all areas of your life.

GitLab broke again (SyntaxError: invalid regexp group)

It just seems GitLab just landed its latest load of Chromeisms™, breaking the site for us SeaMonkey users (at the very least), even with your great addon.

The error for today is:

SyntaxError: invalid regexp group 21.4fceb368.chunk.js:1:37724

...which points to the following regex:
Re=/^&(?<iid>\d+)$/

I'm no regex guru so I can't tell what's wrong, but is this one even workable around?

UPDATE: Checked what that regexp does: it looks for digits preceded by an & character (i.e.: &12345 would match), while the ? part means that this capture group is named (the name being "iid"). That's only supported on Firefox starting at version 78, so I dunno what would be a possible workaround (fwiw, this regex works fine online regex testers like regex101, even on SeaMonkey)

Add IceApe-UXP and BOC/UXP to install.rdf

I do use the polyfill add-on on the IceApe-UXP bower built by @roytam1
The browser being UXP based, it works flawlessly, same deal with the BOC/UXP (bnavigator.exe) also often build by from roytam1.

Could those two browsers be added to the install.rdf file so that people do not have to edit it manually every time? This is what I do for now.
It would look something like this:

    <!-- ICEAPE-UXP -->
    <em:targetApplication>
        <Description>
            <em:id>{9184b6fe-4a5c-484d-8b4b-efbfccbfb514}</em:id>
            <em:minVersion>52.9.*</em:minVersion>
            <em:maxVersion>52.*</em:maxVersion>
        </Description>
    </em:targetApplication>

    <!-- BNAVIGATOR (BOC/UXP) -->
    <em:targetApplication>
      <Description>
        <em:id>{a3210b97-8e8a-4737-9aa0-aa0e607640b9}</em:id>
        <em:minVersion>0.9.0a1</em:minVersion>
        <em:maxVersion>2.*</em:maxVersion>
      </Description>
    </em:targetApplication> 

I do not know your position on roytam builds, I completely understand if you do not want to support those forks.

GitLab (Next) now broken in UXP-based browsers :-(

I've thanked you many times 👍 for creating and maintaining this excellent extension, without which UXP-based browsers (official releases and forks alike 😉 ) would be rendered useless for GitHub/GitLab frequenters... Let this be once more... 😄

Browser: Serpent 52.9.0 (32-bit), buildID=20211008025107
NB: Its UXP platform should be on par/better than the latest official Basilisk 52.9.2021.09.27 (64-bit) release...
Extension version: 1.2.7 (latest)

When one visits a GitLab URL, by default one's being served the "GitLab Next" rendition of that URL:

GLNext_logo

Until a few days ago, v1.2.7 of github-wc-polyfill had no issues working its wonders on GL Next pages, however, this isn't true any more... 😭 E.g., upon loading

https://gitlab.com/gitlab-org/gitlab/-/issues/333598

I get:

GLNext

By the looks of it, this is a repetition of #28, i.e. they're again pushing newer RegExp syntax that isn't palatable to current UXP... 👎 😞 😠

I'm not sure whether this new breakage could be fixed inside the extension (a la #28) and, personally, am losing faith this cat-and-mouse game can go on for much longer... With what's been going on lately in MCP-land and GitLab's determination to bow down willy-nilly to what Google Chrome implements, things look grim... 😿

INTERIM WORKAROUND:

"Current" GitLab appears to work fine with github-wc-polyfill-1.2.7, at least at the time of this writing;
you can opt to revert to Current when visiting GL pages by loading
https://next.gitlab.com/
and moving the slider to the left; your selection is stored in a browser cookie named gitlab_canary, its content value should be false; I've used a browser extension (CookieKeeper 1.9.3.1) to prolong that cookie's expiration time and also "protect" it,
CKGL
so, thankfully, I'm now getting a compatible GitLab iteration... Prone to future breakage, sadly, when they
a. break "Current" even more... 😞
b. retire "Current" completely, when current "Next" becomes the "Current" of the future... 😠

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.