gecko-dev

Interestingly, they are there: https://github.com/mozilla/gecko-dev/blob/771bc161e016e2bd1f... (Confirmed on a recent-ish Firefox browser) They look to have been there for at least 5 years, dunno when they were added before that.

Standard operating procedure for both the Chrome [https://chromium.googlesource.com/chromium/src/+/HEAD/docs/s...] and Firefox [https://www.mozilla.org/en-US/about/governance/policies/secu...] bug tracking systems. But the fix itself is public in both the Chrome [https://chromium.googlesource.com/chromium/src.git/+/36dbbf3...] and Firefox [https://github.com/mozilla/gecko-dev/commit/ac605820636c3b96...] source repos, and it makes pretty clear what the bug is.

Thanks, that's useful - I added that to the document: https://til.simonwillison.net/css/dialog-full-height#user-co... ... which also lead me to the html.css stylesheet in the Firefox source code, a really useful reference: https://searchfox.org/mozilla-central/source/layout/style/re... Here's a GitHub mirror with a commit log view showing changes over time: https://github.com/mozilla/gecko-dev/commits/master/layout/s...

this is the change that fixed it https://github.com/mozilla/gecko-dev/commit/7a85a111b5f42cdc...

For AudioData and VideoEncoder, it should be possible to enable webcodecs by setting dom.media.webcodecs.enabled to true in about:config. - https://github.com/mozilla/gecko-dev/blob/af09c55c0e6253ba8f... - https://github.com/mozilla/gecko-dev/blob/af09c55c0e6253ba8f...

> the Mozilla endpoints to Cloudflare DNS Oh, I did not know that Mozilla had an endpoint on Cloudflare's 1.1.1.1 service, but there it is! https://github.com/mozilla/gecko-dev/blob/c09764753ea40725eb... https://mozilla.cloudflare-dns.com/dns-query Does Mozilla publish their terms somewhere? I'm curious how their endpoint is any different.

It’s not on the brink, it’s that Firefox shims GA[1] with Enhanced Tracking Protection, which is on by default.[2] analytics.usa.gov uses GA. So you really can’t rely on usage figures that don’t represent the truth of your situation. [1] https://github.com/mozilla/gecko-dev/blob/413b88689f3ca2a30b... [2] https://blog.mozilla.org/en/products/firefox/firefox-now-ava...

All the code you need to build Firefox for Linux, macOS, or Windows is in one Mercurial repo: https://hg.mozilla.org/mozilla-central/ The repo is also mirrored on GitHub for convenience, though PRs are not accepted through GitHub: https://github.com/mozilla/gecko-dev The build instructions start with a script (“mach bootstrap”) that will download and install all the blessed tools and SDKs needed to build: https://firefox-source-docs.mozilla.org/setup/index.html Depending on your network and computer, it can be possible to download the source and compile and run your own Firefox build in less than 20 minutes.

The source is mirrored on GitHub here: https://github.com/mozilla/gecko-dev Code search is here: https://searchfox.org/mozilla-central/ I’ve never seen a better search tool. It’s become pretty straightforward to build in the past few years. You pretty much just need your system SDKs (which are listed), python, and git or hg (which is Python anyway!). I promise you that you don’t need CI until you have a patch that works locally.

That page is very out of date. They did later adopt a build step and switch to using JSX in much of the FF DevTools codebase. As an example, here's the Editor main component (same "latest" commit you pointed to): https://github.com/mozilla/gecko-dev/blob/aec3a901e6f6b3041b... The codebase is a mish-mash of very old-style React+Redux code, with newer files and usages. (I know this because I work at Replay.io, which started as a fork of the FF DevTools source, and we spent all of last year modernizing our fork :) Migrated the entire codebase to TS, ripped out all remaining uses of the ancient `React.DOM` helper functions and converted those to JSX, modernized the 2015-era Redux logic, etc: https://github.com/replayio/devtools/pulls?q=is%3Apr+sort%3A... )

Firefox's Devtools uses React and it doesn't use JSX. Example: https://github.com/mozilla/gecko-dev/blob/aec3a901e6f6b3041b... Their reasoning: https://firefox-source-docs.mozilla.org/devtools/frontend/re...

Yes. https://github.com/mozilla/gecko-dev/tree/32c74afbb24dce4b5d...

I used Mozilla for many years and loved it, despite it being a real hog in the early oughts. It's funny to think back about all the handwringing about XUL performance and whatnot that happened in places like Slashdot. For many years every time I got a new machine I would make the effort to switch my link hover cursor to the old Netscape/Mozilla one [1][2] just because I liked it. There is a 22 year old Bugzilla issue for this that I had bookmarked [3]. I don't know how to do this now and it would probably look weird on my HDPI display, but it was something I liked to do when I was younger and more motivated to mess with my setup. These days I just accept whatever is given to my by Apple or Ubuntu. [1] https://i.imgur.com/1YNFW6r.png [2] https://github.com/mozilla/gecko-dev/blob/098bef52b444fde064... [3] https://bugzilla.mozilla.org/show_b

> Gecko is base-60 a.k.a. sexagesimal Thanks for that. I was very confused as to why the value was about 7% larger than 16,777,216 (2²⁴), and not exactly that. But assuming a binary representation (!) having the subdivisions be slightly fewer than a power of 2 (60 being slighly less than 64) means that the whole number limit will be slightly over a power of two, meaning that 60 × 17,895,697 should be a power of 2 (give or take a bit of rounding). And if you do the multiplication, it's 1,073,741,820, which is... huh. 2³⁰ (- 4). Why not 2³¹ or 2³²? Looks like they could at least double the CSS length limit there, even while continuing to allow negative lengths. Edit: As we can see from the parent-linked source on line 36, there is: # define nscoord_MAX nscoord((1 << 30) - 1) which confirms the limit, but doesn't say why 30 is used ins

Gecko is base-60 a.k.a. sexagesimal (https://en.wikipedia.org/wiki/Sexagesimal) which is highly divisible. In Firefox, a pixel is sort of a CSS minute :) Blink and WebKit are less sexy and use tetrasexagesimal a.k.a. 1/64 (https://trac.webkit.org/wiki/LayoutUnit). 1/60 was tried originally, but it was switched to 1/64 for performance and to avoid precision loss. The reason it's faster is interesting: floating point values are also stored as number with a base-2 exponent, so to convert to/from floats, you can use shifts instead of needing division. The differences between 1/60 and 1/64 were important 10 years ago but are pretty minor now. If you're interested, the code is at https://github.com/mozilla/gecko-dev/blob/d36cf98aa85f24ceef... and https://source.chromium.org/chromium/chromium/src/+/main:thi...