| 00:21.05 | *** join/#utah infobot (ibot@rikers.org) |
| 00:21.05 | *** topic/#utah is Utah related discussion including user groups | logs @ http://ibot.rikers.org/%23utah/ | See also ##lds |
| 00:21.05 | *** mode/#utah [+o infobot] by ChanServ |
| 01:17.26 | *** join/#utah aeryndunham_ (~jacki@170.250.131.39) |
| 01:45.55 | *** join/#utah aeryndunham_ (~jacki@170.250.131.39) |
| 01:48.12 | aeryndunham_ | Evening, friends |
| 01:53.57 | jholtom | good evening |
| 03:05.07 | *** join/#utah RobertReynolds (~RobertRey@2605:a601:53b:f800:3dd6:1caa:d584:72c1) |
| 08:27.36 | *** join/#utah RobertReynolds (~RobertRey@2605:a601:53b:f800:3dd6:1caa:d584:72c1) |
| 10:28.07 | *** join/#utah RobertReynolds (~RobertRey@2605:a601:53b:f800:3dd6:1caa:d584:72c1) |
| 11:18.32 | *** join/#utah RuShan (~RuShan@114.sub-174-208-5.myvzw.com) |
| 12:28.36 | *** join/#utah RobertReynolds (~RobertRey@2605:a601:53b:f800:3dd6:1caa:d584:72c1) |
| 13:04.25 | *** join/#utah BitBandit (~BitBandit@50-198-187-221-static.hfc.comcastbusiness.net) |
| 14:29.02 | *** join/#utah RobertReynolds (~RobertRey@2605:a601:53b:f800:3dd6:1caa:d584:72c1) |
| 15:02.57 | *** join/#utah SunSparc (~jonathan@45.56.57.85) |
| 15:06.29 | *** join/#utah RobertReynolds (~RobertRey@128.187.116.7) |
| 16:09.35 | *** join/#utah aeryndunham_ (~jacki@170.250.131.39) |
| 17:00.24 | spangborn | Looks like that east coast storm is gonna make my visit to Boston a bit chilly |
| 17:13.05 | TodPunk | It's good for your bones. |
| 18:42.41 | jlp_znc | I remember describing a chilly day to a friend once as "a bit nippy." He asked if there were little dogs out there, nipping at your heels. |
| 18:44.35 | jlp_znc | perl's Tcl::pTk library is pissing me off. Works great on the Mac, but on Linux it dumps core on exit. Apparently it is trying to free some hash that is already freed, and it confuses the tcl interpreter. |
| 18:44.56 | jlp_znc | DeleteInterpProc calls tclPanic, which calls abort(). |
| 18:45.10 | jlp_znc | I can't seem to catch the SIGABRT, either. |
| 18:46.56 | jlp_znc | stack trace is here: http://pastebin.com/nszVGDmb if anyone has any ideas. |
| 18:49.57 | Migs | does anyone have any exciting Pi Day plans that don't involve a Raspberry Pi? |
| 18:50.12 | jlp_znc | I might eat some pie. |
| 18:51.14 | programmerq | I might make some pie in my deep cast iron skillet |
| 18:52.53 | jlp_znc | might try to get the "I Like Pi" badge on Sporacle: https://www.sporcle.com/achievements/badges/i-like-pi/ |
| 19:03.22 | *** join/#utah aeryndunham_ (~jacki@170.250.131.39) |
| 20:10.03 | levi | jlp_znc: I like that you're using Tcl from Perl and telling us about it straight. Takes a lot of courage to admit that sort of thing, man. |
| 20:12.11 | levi | Apparently my IT department has identified pastebin as a HIGH RISK SERVICE and won't let me view your stack trace there, though. |
| 20:12.27 | joshh | lol |
| 20:46.36 | Migs | use pastebin.ru instead |
| 20:46.46 | Migs | surely that's lower risk |
| 20:47.05 | programmerq | they probably don't want employees pasting internal stuff there. |
| 20:47.40 | Migs | as if there's not hundreds of other paste sites |
| 20:48.55 | programmerq | I wouldn't be surprised if the effect is making this thought process happen more often than not. "I have to share $thing with my coworker. Oh no, pastebin doesn't work! Guess I'll just email it to them instead." |
| 20:59.53 | jlp_znc | levi: well, Perl-Tk (native) works, but on the Mac, it wants you to start up X. Tcl::pTk has a fairly compatible model, and it uses the native tcl library to do it's thing and hence is integrated with cocoa (or whatever that mac UI is called) |
| 21:01.52 | jlp_znc | try https://paste.ee/p/mKMfz instead |
| 21:02.25 | levi | Quartz is the low-level graphics API, but yeah. Odd that perl didn't adopt the tk version that uses the native API. |
| 21:02.50 | jlp_znc | so I converted my app over to Tcl::pTk, and it's working great on the Mac, but on Linux it now throws this weird error and calls abort() |
| 21:03.09 | jlp_znc | I can hide the error, but I have been unsuccessful in trying to catch SIGABRT. |
| 21:04.33 | jlp_znc | and I haven't found a way to call Tcl_SetPanicProc yet (admitedly, I haven't looked too hard) to override the default panic action. |
| 21:11.28 | levi | I dunno; maybe grab the perl module source and see what's in the Tcl.xs file for XS_Tcl__Finalize()? Might be helpful to see exactly what table is being used-after-deleted. |
| 21:12.02 | levi | Your guess is probably better than mine, since I use neither perl nor tcl. |
| 21:16.13 | jlp_znc | Tcl_FindHashEntry |
| 21:19.12 | jlp_znc | Looks like it iterates over the hvInterps hash (global has of live interps), calling Tcl_DeleteInterp() on each one, then getting hte next with hv_iternext() |
| 21:20.27 | jlp_znc | hmmmm, maybe I can just call POSIX::_exit() from my exit handler and avoid the deconstruction entirely. ugly, but it might work... |
| 21:21.27 | jlp_znc | yep, that gets rid of the error message and the abort() |
| 21:21.48 | jlp_znc | thanks for the idea to look at Tcl.xs, levi |
| 21:22.19 | jlp_znc | checks to make sure it doesn't break the Mac |
| 21:23.26 | jlp_znc | looks good. okay, not good, as such, but it works. |
| 21:27.08 | jlp_znc | now I just have to get MouseWheel scrolling working (apparently, in X, it generates Button 4 and Button 5 events instead of MouseWheel events) |
| 21:28.18 | jlp_znc | I don't know if that's really happening or if it's just because my linux box is in a VirtualBox VM |
| 21:37.57 | *** join/#utah RuShan (~RuShan@68.69.169.116) |
| 21:40.52 | jlp_znc | guess I'll try it on a real linux machine tonight. |
| 21:50.53 | levi | Is there such thing as MouseWheel events in X? |
| 21:55.28 | levi | I mean, at some point I stopped really having to care about what events my mouse generated in X, like I stopped having to care about the exact modeline timings of my monitor, and they may have changed some of the plumbing after that point. |
| 21:56.42 | levi | Also, I finally understand what all the modeline numbers mean, because I implemented a VGA display via microcontroller bit-banging. |
| 21:58.26 | levi | I also have working PS/2 keyboard input! |
| 22:01.33 | levi | PS/2 keyboard scancodes are ridiculous. |
| 22:04.37 | *** join/#utah frostyfrog (~frostyfro@unaffiliated/frostyfrog) |
| 22:04.42 | frostyfrog | waves. |
| 22:04.53 | frostyfrog | Yes, that worked! :D |
| 22:25.57 | *** join/#utah RobertReynolds (~RobertRey@128.187.116.7) |
| 22:36.32 | jlp_znc | In X, the mouse wheel generates button 4 and button 5 events. I don't think it has a concept of a mouse wheel. |
| 23:02.36 | levi | jlp_znc: Apparently it has some config options relating to mouse wheels, but it may be that the event model still uses button 4 and 5. |
| 23:04.29 | jlp_znc | I assume it was easier to code since the X protocol supported up to 10 mouse buttons. |
| 23:17.12 | jlp_znc | so, the white house e-mailed me asking me to share my ObamaCare horror story. I sent them the story about how ObamaCare made it possible for my son's wife to receive coverage from his new employer even though she was pregnant at the time he left the military. I'm not entirely sure that's the kind of horror story they were looking for, but hey, it's what I've got. |
| 23:30.43 | levi | You do what you can. |
| 23:33.08 | *** join/#utah RobertReynolds (~RobertRey@2605:a601:53b:f800:f986:aeb:c6d2:6e06) |
| 23:35.03 | jlp_znc | btw, I'm a grandpa now! |
| 23:35.26 | levi | Congrats! |
| 23:35.30 | romanovic | congratulations! |
| 23:43.50 | spangborn | Congrats! |
| 23:53.15 | jlp_znc | thanks! |