| 00:20.03 | *** join/#bzflag infobot (ibot@rikers.org) |
| 00:20.03 | *** topic/#bzflag is https://BZFlag.org || http://ohloh.net/p/bzflag || https://bzflag.org/help || Channel Logs: http://infobot.rikers.org/%23bzflag/ || 2.4.12 released! https://forums.bzflag.org/viewtopic.php?f=62&t=19720 |
| 00:20.03 | *** mode/#bzflag [+v infobot] by ChanServ |
| 00:21.14 | *** join/#bzflag BZnotify (BZnotify@gateway/service/github.com/x-uqwfklayjqdvtdxr) |
| 00:21.15 | BZnotify | [13bzflag] 15blast007 pushed 2 new commits to 062.4: 02https://git.io/vp4kj |
| 00:21.15 | BZnotify | 13bzflag/062.4 1400d7a63 15Scott Wichser: Fix up the update/notify list in the "Making a Release" section. |
| 00:21.15 | BZnotify | 13bzflag/062.4 146c201f6 15Scott Wichser: Fix up some include paths. |
| 00:21.15 | *** part/#bzflag BZnotify (BZnotify@gateway/service/github.com/x-uqwfklayjqdvtdxr) |
| 00:28.04 | *** join/#bzflag BZnotify (BZnotify@gateway/service/github.com/x-ukmljsmhosofbcoh) |
| 00:28.04 | BZnotify | [13bzflag] 15blast007 pushed 2 new commits to 062.4: 02https://git.io/vp4Il |
| 00:28.04 | BZnotify | 13bzflag/062.4 14f8a9742 15Vladimir Jimenez: Remove credits page from help menu |
| 00:28.04 | BZnotify | 13bzflag/062.4 148610361 15blast007: Merge pull request #107 from allejo/feature/nuke-credits... |
| 00:28.04 | *** part/#bzflag BZnotify (BZnotify@gateway/service/github.com/x-ukmljsmhosofbcoh) |
| 01:38.43 | *** join/#bzflag BZnotify (BZnotify@gateway/service/github.com/x-nlaivfczpduxrvdo) |
| 01:38.43 | BZnotify | [13bzflag] 15blast007 closed pull request #91: upgrade to VC2017 (062.4...062.4.VC15_2017) 02https://git.io/vdat5 |
| 01:38.43 | *** part/#bzflag BZnotify (BZnotify@gateway/service/github.com/x-nlaivfczpduxrvdo) |
| 02:23.33 | blast007 | curious if the numpad keys work with PR #108 on macOS. How it works on Windows and Linux now is that for key binding, the keypad keys will be treated differently than the top row and equivilent symbol keys. So it should map as, say, Kp5 instead of 5. |
| 02:24.02 | blast007 | at lesat if numlock is on, which I'm not sure what macOS defaults to (as they don't have a numlock on their keyboards) |
| 02:24.56 | *** mode/#bzflag [-v Foo_man_choo] by ChanServ |
| 02:25.11 | blast007 | the numpad keys should still type in text boxes too |
| 02:28.31 | macsforme | will give it a try |
| 02:29.46 | blast007 | then I'd be curious if, within the mac bins, we can include the licenses for any third-party libraries that we bundle binaries for |
| 02:32.13 | blast007 | even if we have to do something like putting the files within the Xcode/ directory in the bzflag repo |
| 02:57.15 | macsforme | sure |
| 03:00.01 | blast007 | this is what I had grabbed from the Windows dependencies: http://static.bzexcess.com/licenses/ |
| 03:00.35 | blast007 | (one or two of them are just the readme files, which include license info in part of it) |
| 03:11.22 | macsforme | oh, you put them in the dependencies repo? |
| 03:13.05 | blast007 | I didn't add additional files, no |
| 03:13.50 | blast007 | I just copy them into a 'licenses' folder as part of the build batch file |
| 03:15.18 | macsforme | so where should Xcode copy them from? |
| 03:15.59 | blast007 | if it's easier for now, just include them Xcode directory |
| 03:19.39 | blast007 | I don't know if Xcode can include a file from a location outside our source tree or if the license files for our deps are even in a predictable location on macOS |
| 03:32.12 | macsforme | the c-ares source repository has a LICENSE.md file, but it looks like it's not installed with the library, and the SDL 2 framework doesn't seem to contain an obvious license file |
| 03:35.34 | blast007 | just throw the ones you need into the Xcode directory then |
| 03:54.46 | *** join/#bzflag infobot (ibot@rikers.org) |
| 03:54.46 | *** topic/#bzflag is https://BZFlag.org || http://ohloh.net/p/bzflag || https://bzflag.org/help || Channel Logs: http://infobot.rikers.org/%23bzflag/ || 2.4.12 released! https://forums.bzflag.org/viewtopic.php?f=62&t=19720 |
| 03:54.56 | *** mode/#bzflag [+v infobot] by ChanServ |
| 04:03.24 | *** join/#bzflag nadir (uid134094@gateway/web/irccloud.com/x-ixcmsacsxieduwgk) |
| 05:25.02 | *** join/#bzflag I_Died_Once (~I_Died_On@unaffiliated/idiedonce/x-1828535) |
| 06:50.33 | *** join/#bzflag I_Died_Once_ (~I_Died_On@unaffiliated/idiedonce/x-1828535) |
| 17:44.12 | *** mode/#bzflag [+v spldart] by ChanServ |
| 19:32.34 | macsforme | so on macOS, we have five libraries already on the OS that we link to dynamically, one external library that we we link to statically, and one external library that we link to dynamically and copy into the bundle |
| 19:33.20 | macsforme | which of those should we include licenses for? just the two that are not already on the users' systems? |
| 19:34.41 | blast007 | only the ones we're distributing with our binaries, so the two external libs |
| 19:40.22 | *** join/#bzflag nadir (uid134094@gateway/web/irccloud.com/x-qscngdcynogreomm) |
| 19:41.34 | macsforme | k |
| 19:43.19 | blast007 | which two is it? I assume one is SDL2 |
| 20:05.52 | macsforme | yes, and c-ares |
| 20:06.05 | blast007 | k, was thinking that might be the other one |
| 20:06.44 | *** join/#bzflag BZnotify (BZnotify@gateway/service/github.com/x-svoadlfjnezerqxn) |
| 20:06.44 | BZnotify | [13bzflag] 15macsforme pushed 1 new commit to 062.4: 02https://git.io/vp4XM |
| 20:06.44 | BZnotify | 13bzflag/062.4 148eca4e6 15Joshua Bodine: Added license files for SDL 2 and c-ares to the Xcode project so they will be copied into the application bundle. |
| 20:06.44 | *** part/#bzflag BZnotify (BZnotify@gateway/service/github.com/x-svoadlfjnezerqxn) |
| 20:07.49 | blast007 | thanks |
| 20:08.52 | macsforme | that will do it for now, but in the future we should do it in a way that doesn't require us to store those files redundantly in two places |
| 20:09.07 | blast007 | agreed |
| 20:09.18 | blast007 | this was mostly a holdover until 2.6 |
| 20:09.45 | macsforme | perhaps when we are making the dependencies package work on macOS too |
| 20:09.53 | blast007 | that's my hope |
| 20:10.07 | blast007 | does Xcode make it possible/easy to have a dependencies subfolder that we can drop pre-build binaries into? |
| 20:10.18 | blast007 | pre-built* |
| 20:11.13 | macsforme | yes... I even did it the other day for one of my other projects, and it works fine |
| 20:11.21 | blast007 | sweet :) |
| 20:11.33 | blast007 | so yes, that will be the goal going forward |
| 20:19.13 | macsforme | as far as your numpad fix, on macOS you still can't drive with the keypad keys unless you press clear (which is NumLock on macOS), and then it works |
| 20:19.45 | macsforme | except the keypad enter key works either way |
| 20:22.05 | macsforme | when I go to change the binding in the menu, pressing keypad 0 gives "Insert," then after pressing clear it gives keypad 0 |
| 20:22.29 | macsforme | so that is backwards, right? |
| 20:30.37 | macsforme | no I guess that wouldn't be backwards... but I thought by default keypad 0 would just give keypad 0, and then with numlock on you would get just 0... why should that give Insert at all? |
| 20:34.33 | blast007 | does macOS default to having numlock turned off? |
| 20:35.06 | blast007 | if so, that woule explain that behavior. with numlock off the keypad 0 key would be Insert. |
| 20:35.14 | blast007 | would* |
| 20:36.03 | macsforme | yes |
| 20:37.40 | macsforme | but numpad 0 doesn't give insert on macos, regardless of numlock state |
| 20:38.50 | blast007 | 'when I go to change the binding in the menu, pressing keypad 0 gives "Insert,"' |
| 20:38.52 | macsforme | you get either keypad 0, or NUM + keypad 0 |
| 20:39.09 | macsforme | yes, in bzflag |
| 20:39.29 | blast007 | oh, you mean in normal mac software |
| 20:39.53 | macsforme | because you hard-coded that behavior: https://github.com/BZFlag-Dev/bzflag/commit/82250476d2aa5a4ce3d8dab741d8aefd7ce92235#diff-e696a66bf13e750a8c5827cd44236405R199 |
| 20:40.34 | macsforme | is that a Windows thing? |
| 20:42.40 | blast007 | Windows and Linux treats keypad 0 as Insert when numlock is off |
| 20:42.43 | blast007 | SDL2 doesn't though |
| 20:43.13 | blast007 | so that was more of a workaround for SDL's handling of the numpad |
| 20:43.24 | macsforme | hmm |
| 20:44.03 | blast007 | (for instance, in my Gnome terminal, if I turn numlock off, I can use 2 4 6 8 as arrow keys) |
| 20:44.35 | blast007 | I don't know what layer of the system handles that |
| 20:44.49 | macsforme | well, I guess this works... mac users would just have to re-bind those keys, or turn numlock on first... |
| 20:46.29 | macsforme | it's just not what I would expect for a macOS application |
| 20:47.34 | blast007 | xev (which dumps out X events) shows https://gist.github.com/blast007/a4fc3b965fbfde2c87625cc1b2d4f3b2 |
| 20:52.01 | blast007 | would you prefer if macOS always treated it as if numlock was on? |
| 20:52.44 | blast007 | if BZFlag on macOS* |
| 21:11.14 | macsforme | I think I'm okay with what you have, now that I understand the big picture |
| 21:13.56 | macsforme | this will at least allow players to bind those keys to whatever they need |
| 21:19.55 | blast007 | k |
| 21:37.04 | *** join/#bzflag galileo_ (~QUJAAAAAA@d216-108-171-73.nt.northwestel.net) |
| 22:08.16 | *** join/#bzflag BZnotify (BZnotify@gateway/service/github.com/x-clsbstvhworxanmd) |
| 22:08.16 | BZnotify | [13bzflag] 15blast007 pushed 4 new commits to 062.4: 02https://git.io/vp49G |
| 22:08.16 | BZnotify | 13bzflag/062.4 147779135 15Scott Wichser: Allow both a special button type and an ASCII value to be assigned to a single keyboard key. With this, treat the number pad keys uniquely for key mapping while still allowing them to type in text inputs. |
| 22:08.16 | BZnotify | 13bzflag/062.4 14eecca83 15Scott Wichser: Give priority to the special button type when converting a key to a string. |
| 22:08.16 | BZnotify | 13bzflag/062.4 145b1cf28 15Scott Wichser: Mention fix in ChangeLog. |
| 22:08.16 | *** part/#bzflag BZnotify (BZnotify@gateway/service/github.com/x-clsbstvhworxanmd) |
| 22:08.54 | blast007 | allejo: was there anything else you were working on for 2.4.14? |
| 22:10.53 | allejo | not for .14. i'll probably target a .16 come summer time when I have some more free time |
| 22:13.26 | macsforme | I am looking at two more issues... there is a mismatched glBegin/glEnd error with certain settings, so trying to diagnose/bisect that now... I also am checking an issue with multisampling not initially working when the client is started in windowed mode |
| 22:41.57 | blast007 | macsforme: https://github.com/BZFlag-Dev/bzflag/blob/2.4/include/bzfgl.h#L73-L95 <-- does that help you at all? |
| 22:59.54 | macsforme | yes, I was just trying to find the previous unclosed pair |
| 23:01.29 | macsforme | the error was at SegmentedShotStrategy.cxx:326, and I believe that the previous glBegin() was at RadarRenderer.cxx:1187... but I don't see the problem there yet |
| 23:16.51 | blast007 | and leadingShotLine is set to either 0, 1, or 2, right? Looks like if it was anything else it would skip running a glEnd() in SegmentedShotStrategy.cxx |
| 23:17.43 | macsforme | the issue is between SegmentedShotStrategy.cxx:326 and SegmentedShotStrategy.cxx:347 |
| 23:18.50 | macsforme | this player's configuration has "set leadingShotLine 30" |
| 23:19.04 | blast007 | yeah, that's not the length, that's the mode |
| 23:20.24 | blast007 | move 328-330 to the end of that series of 'if/else if' clauses as an else clause instead of an 'else if' |
| 23:20.56 | blast007 | (if 0, lagging, if 2, both, else leading) |
| 23:21.30 | macsforme | okay, good call |
| 23:22.08 | macsforme | I was comparing configurations trying to isolate the issue, but was getting nowhere :P |
| 23:24.44 | blast007 | I believe they were wanting to edit 'linedradarshots' instead :) |
| 23:29.22 | *** join/#bzflag BZnotify (BZnotify@gateway/service/github.com/x-erphtnpjvglomddq) |
| 23:29.22 | BZnotify | [13bzflag] 15macsforme pushed 1 new commit to 062.4: 02https://git.io/vp4QF |
| 23:29.22 | BZnotify | 13bzflag/062.4 14fcd3367 15Joshua Bodine: Protect against bad values for "leadingShotLine" in the configuration. |
| 23:29.22 | *** part/#bzflag BZnotify (BZnotify@gateway/service/github.com/x-erphtnpjvglomddq) |
| 23:31.01 | macsforme | can anyone else try turning multisampling on and then launching the client with -window, and tell me whether multisampling works in the initial window? |
| 23:32.52 | macsforme | what I think I'm seeing is that if you start in windowed mode, multisampling doesn't work until you toggle fullscreen at least once |
| 23:44.03 | blast007 | I can't even see a difference between 8x and off on my Linux+Intel system, so I'm not even sure it's doing anything there |
| 23:49.11 | macsforme | does it look like it's always on or always off? could be a driver override |
| 23:51.25 | blast007 | okay, I guess it is changing |
| 23:52.09 | blast007 | the fonts didn't look different, but the bottom of the mountains did have a difference when I switched it |
| 23:52.25 | blast007 | let me see if I can find somewhere that makes it more noticible |
| 23:54.05 | macsforme | it should be pretty clear on actual map geometry, unless your screen DPI is super high |
| 23:54.07 | blast007 | yeah, it's definitely not doing anti-aliasing when I start it windowed |
| 23:55.20 | macsforme | and then toggling fullscreen fixes it? |
| 23:55.24 | blast007 | yep |