IRC log for #brlcad on 20141120

00:09.45 Stragus pokes ``Erik's elfga.com server not answering on port 80
00:25.22 *** join/#brlcad maths22 (~maths22@66-118-151-70.static.sagonet.net)
00:25.23 *** join/#brlcad Ch3ck_ (~Ch3ck@66-118-151-70.static.sagonet.net)
00:25.24 *** join/#brlcad ejno (~ejno@unaffiliated/kazaik)
00:25.44 *** join/#brlcad n_reed (~molto_cre@66-118-151-70.static.sagonet.net)
00:25.56 *** join/#brlcad hsrai (~hsrai@66-118-151-70.static.sagonet.net)
00:25.58 *** join/#brlcad Notify (~notify@66-118-151-70.static.sagonet.net)
00:26.09 *** join/#brlcad starseeker (~starseeke@66-118-151-70.static.sagonet.net)
00:27.06 *** join/#brlcad brlcad (~sean@66-118-151-70.static.sagonet.net)
00:53.12 *** join/#brlcad tofu_ (~sean@66-118-151-70.static.sagonet.net)
00:54.07 *** join/#brlcad ejno_ (~ejno@unaffiliated/kazaik)
01:04.31 Notify 03BRL-CAD:starseeker * 63509 brlcad/trunk/src/libged/attr.c: Needs more testing, but this gets most of the attr sub-commands using regex pattern matching instead of object-name only. Haven't figured out show subcommand yet, but should have the rest in there. Little fuzzy on the distinction between set and append, needs more study.
01:33.07 Notify 03BRL-CAD:starseeker * 63511 brlcad/branches/qtged/src/qbrlcad/QToolPalette.cxx: Make sure we don't 'lose' a button by not having the row count expanded.
01:33.49 Notify 03BRL-CAD:starseeker * 63510 brlcad/trunk/src/libged/attr.c: Add more memory freeing. Probably a goto is in order here...
01:57.03 Notify 03BRL-CAD:starseeker * 63512 brlcad/branches/qtged/src/qbrlcad/cadtreemodel.cxx: Tweak highlighting rules.
02:09.37 ``Erik heh, Stragus ddija not notice the 8 disconnects from that machine not being routed quite right? :D
02:10.58 Stragus Oh. :D
02:14.42 ``Erik was it the comics page you were trying to hit?
02:15.07 Stragus Yup, it's working fine now
02:15.19 ``Erik needs to fix a few comics and probably add something to have the guest switch to a real acct :/
02:16.19 Stragus I noticed quite a few comics are missing that used to be there...
02:16.21 ``Erik yet more crap on the massive todo list *sigh*
02:16.24 ``Erik which ones?
02:17.51 Stragus Hum. Hijings Ensue, the warehouse, Looking for group, Sinfest...
02:18.59 ``Erik I have two methods for updating, "update" and "grind", update is a very basic "just the most recent" method, grind is a more comprehensive method that tries to parse the archives to get a complete history... if the comic is still updating but not being captured, it probably means that it went long enough that I think a 'grind' needs to be made for it, but haven't had time to do it :/
02:19.33 ``Erik so when I do get to fixing it, it'll come as a flood :)
02:19.59 Stragus I see. :)
02:20.06 Stragus Yes, some comics used to have updates and stopped long ago
02:20.28 Stragus And the "guest" account does show some comic from about an year ago
02:20.37 ``Erik fuck, all 4 you mentioned are still live, they just changed html.. :/
02:20.48 Stragus Yes
02:20.57 Stragus There are quite a few like that
02:21.31 ``Erik I have an admin pane that shows me the last update with color coding based on age, foxtrot is my next target...
02:21.48 ``Erik if you feel the urge to do some lisp coding, I can give you access to the repo ;)
02:22.06 Stragus Can I rewrite everything in C? :D
02:23.10 ``Erik I've a feeling that'd be hundreds of thousands of lines... *shrug* can always cffi it :D go for it
02:24.10 ``Erik crud, it's all very wrapped in my web stuff...
02:26.04 ``Erik http://elfga.com/~erik/tmp/comics/ is the src if you want to contemplate it some
02:26.19 Stragus Eheh. I have written quite a bit of web, HTTP 1.1 and HTML parsing all in C already, I'm sure it wouldn't be that much new code... but I don't have much free time either
02:27.41 Stragus Okay, that's less code than just my TCP code, without the entire HTTP(S) 1.1 implementation on top
02:34.19 ``Erik the kloc might be low, but the work done in places is pretty insane :)
02:35.15 Stragus nods to that
02:35.21 ``Erik I'm not sure if the code bits are in that stuff, but it saves error states in a live replayable list, and serializes everything to disk
02:36.06 ``Erik so if something breaks, you can tell me and I can pull up a full stack trace with the ability to fix and restart the code
02:36.34 Stragus I can enumerate all the comics that don't get updated if that helps. :)
02:36.45 ``Erik heh, have that list, in the admin pane :)
02:36.58 ``Erik ordered by 'last update'
02:37.37 Stragus Most of them just seems to be the parsing script choking on the new HTML
02:38.31 ``Erik my list shows most not being updates, several behind enough to need a 'grind' method,
02:38.49 ``Erik a couple not even being resolvable anymore
02:39.12 ``Erik most of what I think needs to be done is in the 'grind' and 'flurb' methods
02:39.22 ``Erik in the grind-comic.lisp file
02:39.45 ``Erik er, grind.lisp
02:40.19 Stragus grind() and flurb() are 2 and 14 lines of code, how hard can it be? :)
02:40.20 ``Erik it's short, but it does the recursive "get all comics, find next comics page in the archive" fu
02:40.37 ``Erik with throttling and all that
02:41.50 ``Erik #'grind is an abstracted method, so there're several implementations...
02:42.59 ``Erik the one that is "(format t "Not implemented~%") '())" doesnt' count ;)
02:43.18 ``Erik woops, (error "not implemented") even, if straight up bombs
02:43.46 Stragus I don't think I could fix/improve that code faster than I could rewrite everything in C, and that's not a good thing. :)
02:45.05 ``Erik aight, so give me the 100 klocs of C and I'll wrap it in cffi to integrate... or step up your lisp game ;)
02:45.24 *** join/#brlcad kintel (~kintel@unaffiliated/kintel)
02:46.21 ``Erik when I get around to fixing the 'grind' methods, things'll probably start happening fast
02:46.57 ``Erik I just haven't had time... (and I'm currently under a lot of pressure to find a consulting client real effin' soon :/ )
02:47.08 Stragus Ouch, I see
04:15.35 *** join/#brlcad kintel (~kintel@unaffiliated/kintel)
04:40.09 Notify 03BRL-CAD:brlcad * 63513 brlcad/trunk/src/libged/typein.c: apply sf patch 276 (ARS Message Error) from Raj Reddy, verfied by Harmanpreet. this corrects a minor wording mistake in an ARS error message during typein.
04:45.38 Notify 03BRL-CAD:brlcad * 63514 brlcad/trunk/AUTHORS: credit raj reddy with special thanks given a minor tweak he made to ARS (sf patch 276)
05:08.58 *** join/#brlcad kintel (~kintel@unaffiliated/kintel)
06:11.09 *** join/#brlcad maths22_ (~maths22@66-118-151-70.static.sagonet.net)
06:24.51 *** join/#brlcad FreezingCold (~FreezingC@135.0.41.14)
06:27.44 *** join/#brlcad FreezingCold (~FreezingC@135.0.41.14)
08:04.41 *** join/#brlcad yiyus (1242712427@je.je.je)
08:23.35 *** join/#brlcad andrei_ (c35a6e7d@gateway/web/freenode/ip.195.90.110.125)
08:40.01 *** join/#brlcad ries (~ries@D979EA84.cm-3-2d.dynamic.ziggo.nl)
09:22.24 Notify 03BRL-CAD Wiki:JacquesBey * 0 /wiki/User:JacquesBey:
10:04.07 *** join/#brlcad teepee- (bc5c2133@gateway/web/freenode/ip.188.92.33.51)
10:05.39 *** join/#brlcad ries (~ries@D979EA84.cm-3-2d.dynamic.ziggo.nl)
11:31.17 *** join/#brlcad teepee- (bc5c2133@gateway/web/freenode/ip.188.92.33.51)
11:41.20 *** join/#brlcad FreezingCold (~FreezingC@135.0.41.14)
13:59.25 *** join/#brlcad teepee- (bc5c2133@gateway/web/freenode/ip.188.92.33.51)
14:05.45 *** join/#brlcad FreezingCold (~FreezingC@135.0.41.14)
14:29.05 *** join/#brlcad teepee- (bc5c2133@gateway/web/freenode/ip.188.92.33.51)
14:41.45 *** join/#brlcad FreezingCold (~FreezingC@135.0.41.14)
15:04.51 *** join/#brlcad kintel (~kintel@unaffiliated/kintel)
15:12.15 *** join/#brlcad mihaineacsu (~textual@92.85.194.225)
15:18.16 Notify 03BRL-CAD:carlmoore * 63515 brlcad/trunk/src/tclscripts/mged/lc.tcl: remove trailing blanks/tabs
15:27.53 *** join/#brlcad clock (~clock@212.203.58.127)
15:28.25 Notify 03BRL-CAD:starseeker * 63516 brlcad/trunk/src/libged/attr.c: Add pattern support to the attr show option. All attr options need testing, but this should allow for easy bulk reporting and manipulation of attributes.
15:59.51 brlcad this is sad and awesome, turn a terrible barbie book into a better one: https://computer-engineer-barbie.herokuapp.com/
16:00.13 Notify 03BRL-CAD:starseeker * 63517 brlcad/trunk/doc/docbook/system/mann/en/attr.xml: Rework the attr man page, make a few updates.
16:00.47 *** join/#brlcad teepee- (bc5c2133@gateway/web/freenode/ip.188.92.33.51)
16:03.40 *** join/#brlcad FreezingCold (~FreezingC@135.0.41.14)
17:04.08 *** join/#brlcad kintel (~kintel@unaffiliated/kintel)
17:04.31 *** join/#brlcad FreezingCold (~FreezingC@135.0.41.14)
17:32.35 Notify 03BRL-CAD Wiki:Yuriyzotov * 0 /wiki/User:Yuriyzotov:
17:51.27 Notify 03BRL-CAD:n_reed * 63518 brlcad/branches/brep-debug/src/libbrep/tests/ppx.cpp: define a coordinate magnitude limit
18:11.01 brlcad maths22: any progress on the task interface?
18:11.56 brlcad will need to get to a stable state probably within a few days if it's going to get used, so we have enough time to define all our tasks
18:29.11 *** join/#brlcad kintel (~kintel@unaffiliated/kintel)
18:42.32 Notify 03BRL-CAD:n_reed * 63519 brlcad/branches/brep-debug/src/libbrep/tests/ppx.cpp: test that identical points intersect using the default tolerance
18:58.43 Notify 03BRL-CAD:starseeker * 63520 (brlcad/trunk/CHANGES brlcad/trunk/include/raytrace.h and 13 others): Deprecate db_regexp_match_all, add pattern matching to db_ls. Need to swap in db_ls instead of the explicit looping in a few places, but at least some of that code looks like it is duplicated in mged and libged... needs more attention. Glob and match don't have man pages, so not sure yet what they're supposed to do...
18:58.45 Notify Checkpoint a compiling state that works.
18:59.36 brlcad maths22: something to check out http://gci.copyleftgames.org/tasks/ reminds me that pretty much all the orgs desire top-level task categorization like that
19:17.54 *** join/#brlcad FreezingCold (~FreezingC@135.0.41.14)
19:25.10 Notify 03BRL-CAD:brlcad * 63521 brlcad/trunk/src/libsysv/CMakeLists.txt: add an initial multiplatform stub for setenv() to ensure that we have a means to set environment variables reliably. this is mostly for Windows where setenv() doesn't exist, but _putenv_s and _putenv (which we wrap in a #define putenv) do exist.
19:25.57 Notify 03BRL-CAD:brlcad * 63522 brlcad/trunk/CMakeLists.txt: need checks for _putenv_s and getenv_s for the setenv() libsysv implementation.
19:27.53 Notify 03BRL-CAD:brlcad * 63523 brlcad/trunk/src/mged/attach.c: we now always have setenv(), provided by libsysv
19:32.17 Notify 03BRL-CAD:brlcad * 63524 brlcad/trunk/src/libsysv/setenv.c: oops, left the myname in there
20:02.27 *** join/#brlcad kintel (~kintel@unaffiliated/kintel)
20:05.50 Notify 03BRL-CAD:carlmoore * 63525 brlcad/trunk/src/util/pix-bw.c: shift the tests which are done after checking the options
20:24.34 Notify 03BRL-CAD:brlcad * 63526 brlcad/trunk/src/libbrep/BBNode.cpp: quell int->bool conversion performance warning
20:45.44 Notify 03BRL-CAD:brlcad * 63527 (brlcad/trunk/include/raytrace.h brlcad/trunk/include/rt/db_fullpath.h): remove duplicate db_path_to_mat() declaration (they didn't match too) and add note why another wasn't moved.
20:47.31 Notify 03BRL-CAD:starseeker * 63528 brlcad/trunk/src/libged/attr.c: Whoops - need the line return after the first item as well.
20:57.12 Notify 03BRL-CAD:brlcad * 63529 brlcad/trunk/src/other/openNURBS/opennurbs_system.h: make the boolean type be a bool since we don't have an SDK compatibility issue to worry about, uses less memory (just one byte) and is better for performance on Windows
21:22.09 Notify 03BRL-CAD:brlcad * 63530 (brlcad/trunk/src/other/openNURBS/opennurbs_system.h brlcad/trunk/src/other/openNURBS/opennurbs_viewport.h): okay, bad idea because this introduces a slew of unimplemented virtual function errors and it's not immediately obvious why
21:27.21 Notify 03BRL-CAD:brlcad * 63531 (brlcad/trunk/src/libbrep/boolean.cpp brlcad/trunk/src/libbrep/intersect.cpp brlcad/trunk/src/libbrep/opennurbs_ext.cpp): quell unused arg warnings
21:43.57 Notify 03BRL-CAD:brlcad * 63532 brlcad/trunk/src/libfb/if_wgl.c: configurewindow signature says return an int. make it match.
21:49.40 Notify 03BRL-CAD:brlcad * 63533 brlcad/trunk/include/ged.h: expand the callback args
21:52.33 Notify 03BRL-CAD:carlmoore * 63534 brlcad/trunk/src/util/pix-bw.c: we can test for > usage immediately if we detect -o followed by argument
22:02.58 Notify 03BRL-CAD:brlcad * 63535 (brlcad/trunk/src/bwish/winMain.c brlcad/trunk/src/fbserv/server.c brlcad/trunk/src/libfb/if_remote.c): bu_log missing
22:04.03 Notify 03BRL-CAD Wiki:Bernardbalak * 0 /wiki/User:Bernardbalak:
22:05.15 Notify 03BRL-CAD:carlmoore * 63536 brlcad/trunk/src/util/pix-ppm.c: as I just did with pix-bw, implement the 'cannot use both -o and >' in case 'o'

Generated by irclog2html.pl Modified by Tim Riker to work with infobot.