| RAOF | > <@grayson-g:matrix.org> Frame CrashReporter update/question: I'm currently trying to set a check if a log file exists. I've got `inotify` implemented so I can see when a file is created. I need this to run on a thread. I was told that I could use `register_fd_handler()` to watch for events from `inotify` (if I'm... (full message at https://libera.ems.host/_matrix/media/r0/download/libera.chat/6dd7b8acf1757ceaaa0448e210d207acf68d8f70) | 01:11 |
|---|---|---|
| RAOF | I mean, you can also just spin up a thread and do blocking calls if you want, but I think you've got easy access to something that'll run your callback when a fd becomes readable. | 01:13 |
| grayson-g[m] | RAOF: Awesome! I’ll take a look at this in the morning | 01:13 |
| alan_g[m] | The mirserver headers are not public: so Server::the_main_loop isn't available | 07:10 |
| RAOF | Aha! That's what I was missing. | 07:29 |
| alan_g[m] | There is no rule against more functors, but that would be an addition to MirAL for the next Mir release | 07:49 |
| -GitHub[m]:#mir-server- **[MirServer/ubuntu-frame]** AlanGriffiths opened [pull request #79](https://github.com/MirServer/ubuntu-frame/pull/79): Don't try to use evdi | 08:48 | |
| -GitHub[m]:#mir-server- | 08:48 | |
| -GitHub[m]:#mir-server- > Mir platform detection doesn't correctly detect that evdi won't work: Just say "no" | 08:48 | |
| sespiros | Hello everyone, I am doing security updates for the wpewebkit package we have in the Ubuntu archive and I am having some trouble testing it. I install cog, libwpebackend-fdo-1.0-1, libwpe and wpewebkit and try to simply run cog but it segfaults. saviq mentioned that the the team your team is using the snap (wpe-webkit-mir-kiosk), but I didn't have much luck running cog from the snap either, that seems to run but I get no window (running on | 08:49 |
| sespiros | Wayland). I would be super grateful if anyone can give me pointers on how to properly configure and run it. | 08:49 |
| sespiros | s/the the team// | 08:49 |
| alan_g[m] | Welcome sespiros, this happens to be an example snap we use a lot for demos. | 08:52 |
| alan_g[m] | https://mir-server.io/docs/howto-run-your-iot-gui-on-your-desktop | 08:52 |
| alan_g[m] | We don't maintain the snap though, nor know much about what's in it | 08:53 |
| alan_g[m] | The most I've ever done is fork the snap recipe and rebuild to pick up current content | 08:56 |
| alan_g[m] | sespiros I can confirm the snap doesn't run against 22.04's Mutter: `interface 'xdg_toplevel' has no event 2` seems to be the problem. (Which probably has to do with mismatched expectations of how xdg-shell is implemented) | 09:18 |
| sespiros | Thanks alan_g I will check the guide too. hmm I will try running it in focal maybe. Well the snap is not my target, it seems from the repo that they source the libraries directly from upstream and not the Ubuntu archive but I was hoping to use the snap to get insights on how to run the Ubuntu archive versions. It may well be the case that the different libraries that are required for it to work (the ones we have in the archive), do not | 09:24 |
| sespiros | actually work together and probably nobody has tested those ever. FYI from a security point of view both the ones in the Ubuntu archive but also the ones that the snap sources are out of date and contain several open security vulnerabilities. | 09:24 |
| sespiros | And just for context, webkit in general receives limited support from the Canonical security team but our attention was brought to this by the community; the same person who opened https://gitlab.com/glancr/wpe-webkit-snap/-/merge_requests/11 | 09:26 |
| alan_g[m] | That's less than ideal, but I'm not sure there's anything more I can do. The snap maintainer is "tobias" on the snapcraft forums, but he has limited time to do anything with it. (As you've probably seen by the lack of action on the MR) | 09:33 |
| -GitHub[m]:#mir-server- **[MirServer/mir]** AlanGriffiths closed [pull request #2435](https://github.com/MirServer/mir/pull/2435): Miroil sync | 13:52 | |
| grayson-g[m] | alan_g: How would you feel about merging Frame's wallpaper and the CrashReporter into a single internal client? (called `InternalClient` or `StartupClient` or something) | 15:42 |
| grayson-g[m] | Behavior would be: | 15:42 |
| grayson-g[m] | 1) Upon launch, render the gradient wallpaper as before | 15:42 |
| grayson-g[m] | 2) If a crash log is detected, swap the wallpaper to the error background color and display the log text | 15:42 |
| grayson-g[m] | * alan_g: How would you feel about merging Frame's wallpaper and the CrashReporter into a single internal client? (called `InternalClient` or `StartupClient` or something)... (full message at https://libera.ems.host/_matrix/media/r0/download/libera.chat/94ef99e354ed81c3cb28e72a1c5512cd67a4df6a) | 15:42 |
| alan_g[m] | <grayson-g[m]> "alan_g: How would you feel about..." <- Behavior would be:... (full message at https://libera.ems.host/_matrix/media/r0/download/libera.chat/b2f884a6574eecded16478f30a6add97c2ba67b9) | 15:59 |
| sophie-w | Is it possible to add `CTEST_OUTPUT_ON_FAILURE=1` or something like that to CI so I can see the output when cmake tests fail? Currently it seems nothing is shown except that the test failed (for example, here | 17:11 |
| sophie-w | https://pipelines.actions.githubusercontent.com/serviceHosts/c970ea87-beb6-485e-bd4a-f9b0c1bb1dd9/_apis/pipelines/1/runs/18902/signedlogcontent/13?urlExpires=2022-07-29T16%3A47%3A47.5226506Z&urlSigningMethod=HMACV1&urlSignature=mMbLAqXa%2BP3q%2FekGx9CgiQXASZ6pxRE4VOpXZ9I62SM%3D and search symbols-up-to-date) | 17:11 |
| -GitHub[m]:#mir-server- **[MirServer/mir]** bors[bot] merged [pull request #2518](https://github.com/MirServer/mir/pull/2518): Add tools/scan_symbols.py script | 17:14 | |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!