[py] Use XWayland for internal Python Firefox tests

[cgoldberg]

User description

🔗 Related Issues

Fixes #15584

💥 What does this PR do?

There are several issues related to setting window size/position when running Firefox under Wayland display server (Linux). This PR sets the MOZ_ENABLE_WAYLAND=0 environment variable in our PyTest configuration, so Firefox uses XWayland instead.

(This should fix several CI failures)

🔄 Types of changes

• Bug fix (backwards compatible)

---

PR Type

Bug fix

---

Description

• Set MOZ_ENABLE_WAYLAND=0 for Firefox in tests to use XWayland.

• Fixes issues with window size/position under Wayland on Linux.

• Resolves CI failures related to Firefox on Wayland.

---

Changes walkthrough 📝

	Relevant files
Bug fix	conftest.py Configure Firefox to use XWayland in tests                              py/conftest.py Added environment variable MOZ_ENABLE_WAYLAND=0 for Firefox. Ensures Firefox uses XWayland instead of Wayland. Fixes issues with window size/position under Wayland. +3/-0

---

Need help?

Type /help how to ... in the comments thread for any questions about Qodo Merge usage.

Check out the documentation for more information.

[qodo-code-review]

PR Reviewer Guide 🔍

Here are some key observations to aid the review process:

⏱️ Estimated effort to review: 1 🔵⚪⚪⚪⚪

🧪 No relevant tests

🔒 No security concerns identified

⚡ No major issues detected

[qodo-code-review]

PR Code Suggestions ✨

Explore these optional code suggestions:

Category	Suggestion	Impact
Possible issue	✅ Ensure environment variable timing Suggestion Impact: The commit partially addressed the suggestion's concern about environment variable timing by moving the MOZ_ENABLE_WAYLAND setting to the Remote driver section, ensuring it's set before Firefox is launched in that context. However, it didn't implement the conditional check to avoid overriding existing values. code diff: - os.environ['MOZ_ENABLE_WAYLAND'] = '0' + os.environ["MOZ_ENABLE_WAYLAND"] = "0" if driver_class == "Chrome": options = get_options(driver_class, request.config) if driver_class == "Edge": options = get_options(driver_class, request.config) if driver_class == "WebKitGTK": options = get_options(driver_class, request.config) - if driver_class.lower() == "WPEWebKit": + if driver_class == "WPEWebKit": options = get_options(driver_class, request.config) if driver_class == "Remote": options = get_options("Firefox", request.config) or webdriver.FirefoxOptions() options.set_capability("moz:firefoxOptions", {}) options.enable_downloads = True + # There are issues with window size/position when running Firefox + # under Wayland, so we use XWayland instead. + os.environ["MOZ_ENABLE_WAYLAND"] = "0" Setting environment variables after the process has started may not affect Firefox's behavior. Environment variables should be set before Firefox is launched. Consider checking if this environment variable is being set early enough to affect Firefox's startup behavior. py/conftest.py [157-159] # There are issues with window size/position when running Firefox # under Wayland, so we use XWayland instead. -os.environ['MOZ_ENABLE_WAYLAND'] = '0' +# Set this before Firefox is launched to ensure it takes effect +if 'MOZ_ENABLE_WAYLAND' not in os.environ: + os.environ['MOZ_ENABLE_WAYLAND'] = '0' [Suggestion has been applied] Suggestion importance[1-10]: 7 __ Why: The suggestion correctly identifies a potential issue with environment variable timing. Setting environment variables after process initialization may not affect Firefox's behavior, and the improved code adds a check to avoid overriding existing values, which is a good practice for environment configuration.	Medium
Update

[Delta456]

Is there a way to also test this environment variable in test files? It would be a good addition.

[cgoldberg]

@Delta456 you mean add tests that verify it works with this env variable set and broken without? I can add those.

[Delta456]

Yes those. They will be very useful

[cgoldberg]

@Delta456 I added some tests. Can you give it a review?

[Delta456]

Looks good to me though the Lint CI didn't start correctly. Can you restart that?

[cgoldberg]

That was actually a legitimate linting error. I just fixed it and it's running now.

[cgoldberg]

CI is failing because of #15608 (unrelated to this PR)