Firefox changed the way the timeZone is returned

[mawis]

Describe the bug

Since I updated to NixOS 23.05 I noticed, that several Elm programmed sites were broken in Firefox. I finally now had the time to dig deeper in that and I found the underlying problem: The way the timezone is returned by Firefox has changed with NixOS 23.05. When running the command Intl.DateTimeFormat().resolvedOptions().timeZone in the Firefox console, up until NixOS 22.11 the returned value on my system was Europe/Berlin, but with NixOS 23.05 it's no just CET.

I checked with Firefox on Ubuntu, MacOS and Windows. All still return Europe/Berlin in the current version of Firefox, so I think the problem has been introduced by NixOS and not by Firefox.

This change breaks the (at least) the timezone-data package of the Elm ecosystem, as it is now returning a TimeZone.error instead of a Time.Zone.

BTW: chromium on NixOS is still returning the expected value of Europe/Berlin on my system.

Steps To Reproduce

Steps to reproduce the behavior:

1. Open Firefox
2. Open the tools for web developers (Ctrl-Shift-I)
3. Execute the following expression: Intl.DateTimeFormat().resolvedOptions().timeZone

Expected behavior

Firefox should return the timezone as Europe/Berlin as it does on other platforms and as chromium does as well.

Screenshots

Result on NixOS 23.05/Firefox 115 is:

Result on Windows 10 is:

Notify maintainers

@lovesegfault @Hexa

Metadata

Please run nix-shell -p nix-info --run "nix-info -m" and paste the result.

 - system: `"x86_64-linux"`
 - host os: `Linux 6.1.41, NixOS, 23.05 (Stoat), 23.05.2385.48e82fe1b1c`
 - multi-user?: `yes`
 - sandbox: `yes`
 - version: `nix-env (Nix) 2.13.3`
 - channels(root): `"nixos-23.05, old-22.05, unstable"`
 - channels(matthias): `""`
 - nixpkgs: `/nix/var/nix/profiles/per-user/root/channels/nixos`

[bobby285271]

Maybe this is #238025?

[mawis]

Thanks @bobby285271

I first thought it's different, as I do get a timezone (and not UTC). But setting the TZ environment variable fixed my problem as well, so I guess it's a duplicate. Thanks for pointing out.