[GLib] Open test results in default browser if MiniBrowser fails

[TingPing]

5a1edfb

[GLib] Open test results in default browser if MiniBrowser fails
https://bugs.webkit.org/show_bug.cgi?id=268015

Reviewed by Adrian Perez de Castro.

If something goes wrong, which happens during development, it is
useful to see results in a different browser rather than nothing.

* Tools/Scripts/webkitpy/port/gtk.py:
(GtkPort.show_results_html_file):

Canonical link: https://commits.webkit.org/273490@main

c076b70

Misc	iOS, tvOS & watchOS	macOS	Linux	Windows
✅ 🧪 style	✅ 🛠 ios	✅ 🛠 mac	✅ 🛠 wpe	✅ 🛠 wincairo
✅ 🧪 bindings	✅ 🛠 ios-sim	✅ 🛠 mac-AS-debug	✅ 🧪 wpe-wk2
✅ 🧪 webkitperl	✅ 🧪 ios-wk2	✅ 🧪 api-mac	✅ 🧪 api-wpe
❌ 🧪 webkitpy	✅ 🧪 ios-wk2-wpt	✅ 🧪 mac-wk1	✅ 🛠 gtk
	✅ 🧪 api-ios	✅ 🧪 mac-wk2	✅ 🧪 gtk-wk2
	✅ 🛠 tv	✅ 🧪 mac-AS-debug-wk2	✅ 🧪 api-gtk
	✅ 🛠 tv-sim
🛠 🧪 merge	✅ 🛠 watch
✅ 🛠 🧪 unsafe-merge	✅ 🛠 watch-sim

[webkit-early-warning-system]

EWS run on previous version of this PR (hash 4af85bc)

Misc	iOS, tvOS & watchOS	macOS	Linux	Windows
✅ 🧪 style	✅ 🛠 ios	✅ 🛠 mac	✅ 🛠 wpe	⏳ 🛠 wincairo
✅ 🧪 bindings	✅ 🛠 ios-sim	✅ 🛠 mac-AS-debug	⏳ 🧪 wpe-wk2
✅ 🧪 webkitperl	✅ 🧪 ios-wk2	⏳ 🧪 api-mac	✅ 🧪 api-wpe
🧪 webkitpy	⏳ 🧪 ios-wk2-wpt	✅ 🧪 mac-wk1	✅ 🛠 gtk
⏳ 🛠 🧪 jsc	✅ 🧪 api-ios	✅ 🧪 mac-wk2	🧪 gtk-wk2
	✅ 🛠 tv	⏳ 🧪 mac-AS-debug-wk2	🧪 api-gtk
	⏳ 🛠 tv-sim
	✅ 🛠 watch
	⏳ 🛠 watch-sim

[TingPing]

Personally I'd prefer it always ran in my default, but it was changed to do this not that long ago.

[aperezdc]

Wouldn't it be better to try webbrowser.open_new_tab() before the gio tool? In that case we could have that as common fallback for all the ports.

[TingPing]

webbrowser.open doesn't work on my system 🤷. gio correctly does what it should on Linux at least.

[TingPing]

xdg-open is broken in my container, but that is on me I guess. I can switch it to webbrowser.

[aperezdc]

webbrowser.open doesn't work on my system 🤷. gio correctly does what it should on Linux at least.

That's amusing, the webbrowser module works just fine here. Anyway, I have nothing against using gio, so patch LGTM anyway.

[TingPing]

All of the other ports do very port specific things for show_results_html_file so I think it would be glib specific anyway.

[webkit-early-warning-system]

EWS run on current version of this PR (hash c076b70)

Misc	iOS, tvOS & watchOS	macOS	Linux	Windows
✅ 🧪 style	✅ 🛠 ios	✅ 🛠 mac	✅ 🛠 wpe	✅ 🛠 wincairo
✅ 🧪 bindings	✅ 🛠 ios-sim	✅ 🛠 mac-AS-debug	✅ 🧪 wpe-wk2
✅ 🧪 webkitperl	✅ 🧪 ios-wk2	✅ 🧪 api-mac	✅ 🧪 api-wpe
❌ 🧪 webkitpy	✅ 🧪 ios-wk2-wpt	✅ 🧪 mac-wk1	✅ 🛠 gtk
	✅ 🧪 api-ios	✅ 🧪 mac-wk2	✅ 🧪 gtk-wk2
	✅ 🛠 tv	✅ 🧪 mac-AS-debug-wk2	✅ 🧪 api-gtk
	✅ 🛠 tv-sim
🛠 🧪 merge	✅ 🛠 watch
✅ 🛠 🧪 unsafe-merge	✅ 🛠 watch-sim

[mcatanzaro]

Personally I'd prefer it always ran in my default, but it was changed to do this not that long ago.

To change that, we'd first really need to figure out how to make local subresources work under flatpak. The test results page is going to be super messed up otherwise.

[TingPing]

webkitpy failure is unrelated:

[2410/2441] webkitpy.port.server_process_unittest.TestServerProcess.serial_test_basic erred:
  Traceback (most recent call last):
    File "/Volumes/Data/worker/WebKitPy-Tests-EWS/build/Tools/Scripts/webkitpy/port/server_process_unittest.py", line 125, in serial_test_basic
      self.assertEqual(line.strip(), b"stdout")
  AttributeError: 'NoneType' object has no attribute 'strip'

[webkit-commit-queue]

Committed 273490@main (5a1edfb): https://commits.webkit.org/273490@main

Reviewed commits have been landed. Closing PR #23177 and removing active labels.