| 00:06.56 | blast007 | wouldn't even need that else since the break would leave the loop immediately |
| 00:07.36 | blast007 | starting to test that here |
| 00:25.26 | blast007 | macsforme: I fixed the issue where it didn't restore the resolution on exit in Linux. Making sure that works with SDL 1 too. |
| 00:26.51 | macsforme | ah, true |
| 00:29.37 | macsforme | okay that's good... I was just looking at the F4 issue... so far, I saw that it creates a window resize event when minimizing, but purging that event doesn't correct the issue... still investigating further |
| 00:39.27 | BZNotify | bzflag: 03macsforme 11synchronized pull request #251 "Fix multiple issues with SDL 2 window creation" (06https://git.io/JfnZO) |
| 01:24.40 | blast007 | okay, so I guess SDL 1 (or at least our SDL 1 code) does not like multiple displayc because it just displays spanned across both displays, acting like it's one giant display |
| 01:27.39 | blast007 | leaves it real broken if I change display modes and then exit too :D |
| 01:27.53 | blast007 | (this isn't anything with your PR - this is our stock code too) |
| 01:28.30 | blast007 | do we want to get rid of all non-SDL2 platform code for 2.4.22? |
| 01:44.55 | BZNotify | bzflag: 03blast007 11synchronized pull request #251 "Fix multiple issues with SDL 2 window creation" by 03macsforme (06https://git.io/JfnZO) |
| 02:57.00 | macsforme | that was simple :-P |
| 03:00.31 | blast007 | :) |
| 03:00.47 | blast007 | now I think it's just the iconify issue |
| 03:06.05 | macsforme | even if I purge all events right after minimizing (which gets rid of one resize event), there is still another resize event coming through afterwards... I haven't found where yet, but I think that's the issue |
| 03:08.12 | macsforme | and obviously, purging all events isn't the long-term solution... we may just need another event filter to skip resize events until the window is restored |
| 03:25.13 | *** join/#bzflag I_Died_Once (~I_Died_On@unaffiliated/idiedonce/x-1828535) |
| 04:18.17 | *** join/#bzflag infobot (ibot@96.86.209.99) |
| 04:18.17 | *** topic/#bzflag is BZFlag Support and Development || Latest version: 2.4.20 || https://www.bzflag.org || https://www.bzflag.org/help/ || https://www.openhub.net/p/bzflag || https://logs.bzexcess.com/freenode/%23bzflag/ || New release! https://forums.bzflag.org/viewtopic.php?f=62&t=20395 |
| 04:18.17 | *** mode/#bzflag [+v infobot] by ChanServ |
| 04:22.30 | *** join/#bzflag CaptainRoberts[m (generalpur@gateway/shell/matrix.org/x-vireadpsczsnysqb) |
| 04:37.39 | *** join/#bzflag Sgeo (~Sgeo@ool-18b982ad.dyn.optonline.net) |
| 04:45.00 | *** join/#bzflag blast007[m] (blast007ma@gateway/shell/matrix.org/x-cqjdgksyyzevoajc) |
| 08:31.28 | *** join/#bzflag infobot (ibot@96-86-209-99-static.hfc.comcastbusiness.net) |
| 08:31.28 | *** topic/#bzflag is BZFlag Support and Development || Latest version: 2.4.20 || https://www.bzflag.org || https://www.bzflag.org/help/ || https://www.openhub.net/p/bzflag || https://logs.bzexcess.com/freenode/%23bzflag/ || New release! https://forums.bzflag.org/viewtopic.php?f=62&t=20395 |
| 08:31.28 | *** mode/#bzflag [+v infobot] by ChanServ |
| 11:27.30 | *** join/#bzflag the_map (~the_map@unaffiliated/the-map/x-1795707) |
| 11:27.30 | *** join/#bzflag ruskie (ruskie@sourcemage/mage/ruskie) |
| 11:27.30 | *** join/#bzflag blast007 (~blast007@bzflag/developer/Blast) |
| 16:21.04 | macsforme | these are the SDL events triggering the unwanted window restoration, and the SDL events following a correct window restoration for comparison: https://pastebin.com/gYuPSmrR |
| 16:24.15 | macsforme | the two options I see are 1) filter and block any SDL_WINDOW_EVENT_RESIZED events occurring between SDL_WINDOWEVENT_MINIMIZED and SDL_WINDOWEVENT_SHOWN, or 2) change the window resolution back to the native screen resolution before minimizing, and put it back to the correct size when restoring |
| 17:32.35 | *** join/#bzflag Sgeo (~Sgeo@ool-18b982ad.dyn.optonline.net) |
| 17:41.55 | macsforme | hmmm, also need to filter out the resize after an alt-tab to a different application... |
| 18:29.50 | macsforme | the issue (in both cases) is that the unwanted resize event comes before the minimize and hidden events... long enough beforehand that the resize is already processed by the game before SDL puts the other events in the event queue, from what I can tell |
| 19:07.51 | blast007 | macsforme: if we switch to windowed before minimizing, does that work? |
| 19:08.29 | blast007 | I can't recall if I had the issue when I was in windowed mode |
| 19:14.39 | *** mode/#bzflag [-v short_circuit] by ChanServ |
| 19:15.15 | blast007 | I'll have to play around with SDL_PumpEvents and delays in my platform code rewrite and see if that helps |
| 19:16.04 | blast007 | I had previously tried putting an SDL_PumpEvents() between every other SDL_* function call out of desperation, but I hadn't ever added any delays |
| 19:51.02 | Sgeo | Why would I have UDP issues some times but not other times? |
| 19:51.14 | Sgeo | Got kicked for not using UDP, but was playing last night just fine |
| 21:17.43 | blast007 | are you using wifi? |
| 21:29.45 | blast007 | macsforme: I'm also going to try using KDE instead of Gnome and see if there's a difference in the behavior for any of this stuff |
| 21:34.05 | Sgeo | blast007, yes |
| 21:45.08 | blast007 | could just be a bad connection |