| 18 Aug 2025 |
emily | and it doesn't even use pyside2 | 17:12:48 |
emily | but e.g. renderdoc is just using pyside2 | 17:13:02 |
Grimmauld (any/all) | Oh god renderdoc.. That thing is super broken already, iirc it breaks dlopen... But yes people will be mad if we break renderdoc | 17:14:07 |
emily | will they really | 17:15:07 |
emily | how many people are truly using a graphics debugger | 17:15:11 |
emily | I mean not that I don't think it's cool and worth keeping working but it doesn't exactly seem mass market | 17:15:26 |
K900 | Why does renderdoc need qwe to begin with | 17:15:29 |
emily | it doesn't | 17:15:33 |
emily | it needs pyside2 | 17:15:34 |
K900 | Then just yeet qwe from pyside2 | 17:15:44 |
emily | yes that's what I'm saying | 17:15:50 |
emily | what we should probably do is add a flag | 17:15:57 |
K900 | I bet the only user of that is qutebrowser-qt5 which we can also yeet now that amjoseph self-immolated | 17:15:57 |
emily | for backportability | 17:15:59 |
emily | i.e. pyside2 grows withWebEngine ? true,, we set it to false for all known users that don't seem to obviously want a browser engine, then that's backportable | 17:16:28 |
emily | then on unstable we swap it to false or delete the argument | 17:16:36 |
K900 | I guess we could | 17:18:02 |
emily | i mean mostly it seems bad to break WebEngine in pyside2 on stable | 17:25:24 |
emily | and there are only a handful of users anyway | 17:25:37 |
emily | withPySide2 = plotpy.override {
pyqt6 = pyside2;
qt6 = qt5;
};
| 17:26:57 |
K900 | The FUCK | 17:27:33 |
emily | it's in a "known failing tests" passthru | 17:28:11 |
emily | but lol | 17:28:12 |
emily | Grimmauld (any/all): (have we successfully sniped you into doing the marking btw) | 17:28:30 |
aloisw | They have at least three, Konqueror still exists. | 17:52:14 |
emily | I thought Konqueror was kinda dead. | 17:54:57 |
emily | like not really maintained. | 17:55:00 |
emily | is it on Qt 6? | 17:55:03 |
K900 | Barely | 17:55:31 |
aloisw | It runs at least. | 17:56:05 |