| 00:50.46 | blast007 | basically it would depend on the feature and how invasive it was and if it fit into the plan for the release |
| 01:10.26 | *** join/#bzflag Pimpi (~frank@pD9E100BA.dip0.t-ipconnect.de) |
| 01:10.54 | BulletCatcher | Normally, any individual commit that actually changes the over-the-wire protocol should include an increment of the protocol number. |
| 01:10.56 | BulletCatcher | The comments in src/date/buildDate.cxx list all of the files to be changed when that happens. |
| 01:11.54 | BulletCatcher | If anyone has any ideas for reducing the number of distinct changes required when the protocol number increments, we're interested. |
| 01:12.56 | *** join/#bzflag khonkhortisan (~Kyle@c-24-16-240-200.hsd1.wa.comcast.net) |
| 01:19.39 | blast007 | so something about my src/other/ removal and using the external dependency package has made bzadmin on Windows unhappy.. |
| 01:46.41 | *** join/#bzflag khonkhortisan (~Kyle@c-24-16-240-200.hsd1.wa.comcast.net) |
| 02:17.14 | BulletCatcher | bzadmin on Linux seems to work normally, FWIW. |
| 02:27.04 | blast007 | might be some memory corruption because it's throwing an access violation and the 'LINES' variable for pdcurses is being set to some large negative value.. |
| 02:28.16 | blast007 | doesn't crash in debug, but does in release mode, so there's probably some uninitialized memory |
| 03:59.03 | *** join/#bzflag khonkhortisan (~Kyle@c-24-16-240-200.hsd1.wa.comcast.net) |
| 04:56.31 | *** join/#bzflag AAA_awright (~aaa@2607:5300:60:7d4::1) |
| 08:40.03 | *** join/#bzflag Constitution (~josh@bzflag/developer/Constitution) |
| 10:14.28 | *** join/#bzflag I_Died_Once (~I_Died_On@unaffiliated/idiedonce/x-1828535) |
| 11:22.06 | *** join/#bzflag I_Died_Once (~I_Died_On@unaffiliated/idiedonce/x-1828535) |
| 15:13.15 | blast007 | when run in a debugger, the LINES variable has a value of 0xCDCDCDCD inside CursesUI::CursesUI "Used by Microsoft's C/C++ debug malloc() function to mark uninitialized heap memory, usually returned from HeapAlloc()" |
| 15:14.28 | blast007 | and that's after initscr() is called |
| 15:37.06 | *** join/#bzflag MonsieurApple (~MonsieurA@unaffiliated/mrapple) |
| 15:37.06 | *** join/#bzflag alezakos (~kongr45gp@95.46.199.28) |
| 15:37.12 | *** join/#bzflag mdskpr (~quassel@209.222.8.90) |
| 15:39.03 | BulletCatcher | You may need to call use_env() before initscr(). |
| 15:40.15 | BulletCatcher | use_env(true) looks right. |
| 15:41.29 | blast007 | "filter() and use_env() are no-ops in PDCurses." |
| 15:42.13 | BulletCatcher | Maybe you have found the case where they should do more than that. :-p |
| 15:43.58 | BulletCatcher | If bzadmin worked when src/other was there, then that should help find what is different now. |
| 15:44.54 | blast007 | int LINES = 0; in initscr.c (outside of any function definition) and it's set to the correct value inside Xinitscr() (which gets called by initscr() ) |
| 15:44.57 | blast007 | yeah |
| 15:45.17 | blast007 | I'll start by putting PDCurses back from the old src/other/ stuff |
| 15:52.03 | *** join/#bzflag kierra (~quassel@209.222.8.90) |
| 16:44.47 | *** join/#bzflag zuii (~akos@BC06BF93.catv.pool.telekom.hu) |
| 17:26.16 | *** join/#bzflag zuii (~akos@BC06BF93.catv.pool.telekom.hu) |
| 17:46.51 | *** join/#bzflag dngor (~abuse@p3m/dngor) |
| 17:54.02 | *** join/#bzflag alezakos (~kongr45gp@bzflag/developer/alezakos) |
| 18:51.58 | *** join/#bzflag zuii (~akos@BC06BF93.catv.pool.telekom.hu) |
| 19:29.28 | *** join/#bzflag Constitution (~josh@bzflag/developer/Constitution) |
| 19:34.48 | *** part/#bzflag Guest58597 (~quassel@209.222.8.90) |
| 19:36.44 | *** join/#bzflag kierra (~quassel@unaffiliated/kierra) |
| 21:49.18 | Notify | 03BZFlag:24.16.240.200 * 8894 /Installing_on_Linux: suse uses zypper |
| 23:00.50 | *** join/#bzflag brlcad (~sean@66-118-151-70.static.sagonet.net) |
| 23:10.11 | *** join/#bzflag Notify (~notify@66-118-151-70.static.sagonet.net) |