IRC log for #brlcad on 20091126

00:00.57 brlcad down to just 36 zombies remaining, 2964 zombies per cop kills
00:01.37 brlcad oh hnoes.. looks like they actually got him! hah
00:02.05 brlcad ahhhh, never mind.. there he is.. hidden under the google logo
00:08.16 CIA-14 BRL-CAD: 03brlcad * r36673 10/brlcad/trunk/src/libdm/tcl.c: restructure to avoid forward decls, mark unused parameters
00:09.21 CIA-14 BRL-CAD: 03brlcad * r36674 10/brlcad/trunk/src/libdm/tcl.c: ws
00:19.55 CIA-14 BRL-CAD: 03brlcad * r36675 10/brlcad/trunk/src/libtermio/termio.c: ws cleanup, break out TAB3 and XTABS
00:38.40 *** join/#brlcad Ralith (n=ralith@69.90.48.97)
00:59.27 CIA-14 BRL-CAD: 03brlcad * r36676 10/brlcad/trunk/src/libtermio/termio.c:
00:59.27 CIA-14 BRL-CAD: refactor the antiquated tab-expansion logic to work even if there is no tab
00:59.27 CIA-14 BRL-CAD: support (strict c99 mode). try TAB3 if we can, or one of several other variants
00:59.27 CIA-14 BRL-CAD: until we either find something otherwise punt and disable support on our end
00:59.27 CIA-14 BRL-CAD: too.
01:01.20 CIA-14 BRL-CAD: 03brlcad * r36677 10/brlcad/trunk/include/pkg.h: uname shadows a global
02:06.04 CIA-14 BRL-CAD: 03brlcad * r36678 10/brlcad/trunk/src/libtclcad/ (tclcadAutoPath.c tkImgFmtPIX.c): quell warnings
02:23.56 CIA-14 BRL-CAD: 03brlcad * r36679 10/brlcad/trunk/ (include/dm.h src/libdm/axes.c): mark thy constness. rmat can be const.
02:24.26 CIA-14 BRL-CAD: 03brlcad * r36680 10/brlcad/trunk/src/libtclcad/ged_obj.c: quell warnings, make functions use HIDDEN
02:35.45 CIA-14 BRL-CAD: 03brlcad * r36681 10/brlcad/trunk/src/conv/iges/g-iges.c: c99 dictates the use of S_ISDIR() instead of accessing the S_IFDIR mask directly
02:38.39 *** join/#brlcad Don__ (n=Don@c-71-238-51-148.hsd1.mi.comcast.net)
02:40.39 CIA-14 BRL-CAD: 03brlcad * r36682 10/brlcad/trunk/src/nirt/parse_fmt.c: ws indent cleanup
02:43.48 CIA-14 BRL-CAD: 03brlcad * r36683 10/brlcad/trunk/src/nirt/parse_fmt.c: c99 compliance, no popen() available so don't rely on it.
02:46.20 CIA-14 BRL-CAD: 03brlcad * r36684 10/brlcad/trunk/src/sig/dfft.c: ws cleanup
02:48.11 CIA-14 BRL-CAD: 03brlcad * r36685 10/brlcad/trunk/src/sig/dfft.c: use STDIN_FILENO and STDOUT_FILENO to be c99 compliant.
02:52.22 CIA-14 BRL-CAD: 03brlcad * r36686 10/brlcad/trunk/src/sig/syn.c: ws cleanup, add vmath.h for M_PI (in case we're c99 strict), reorder to remove decl.
03:03.31 CIA-14 BRL-CAD: 03brlcad * r36687 10/brlcad/trunk/bench/pixcmp.c: declare the getopt bits, cleanup, remove pixdiffish comment
03:18.39 CIA-14 BRL-CAD: 03brlcad * r36688 10/brlcad/trunk/src/archer/TODO: note a handful of immediate impression issues provoked by feedback from a user (andre anckaert). bad help and missing commands.
03:30.34 CIA-14 BRL-CAD: 03brlcad * r36689 10/brlcad/trunk/ (2 files in 2 dirs): try a manual edit of the sole 'Crossbow' reference in the windows chm file.. couldn't find where BRL-CAD Commands was coming from, though.
03:34.18 starseeker blinks - a chm file? I hadn't noticed that
03:34.30 brlcad burn it!
03:34.33 starseeker no kidding
03:34.50 brlcad only noticed myself after grepping the file tree for Crossbow
03:35.01 starseeker thought it was just a funky html until the only dell printer manual I could find turned out to need a special app to read...
03:35.48 brlcad it is basically
03:35.50 brlcad compiled html
03:35.51 starseeker makes note to pester Bob into getting tkhtml3 working on Windows - then we can bring in the html viewer and nuke that sucker
03:36.33 starseeker however, since I have the viewer I can check the contents...
03:37.23 ``Erik chm? really? does windows even use those anymore?
03:37.52 starseeker snorts - no, I think they noticed it had been decoded and thought up something else...
03:38.02 CIA-14 BRL-CAD: 03brlcad * r36690 10/brlcad/trunk/src/archer/TODO: archer docs should not be in chm format (even for windows)
03:38.08 brlcad there's another change that chm needs.. calls them "BRL-CAD Commands"
03:38.17 starseeker erm... Needed file /Archer/Archer_Introduction.htm is not present in this CHM archive
03:38.28 brlcad lovely
03:38.41 brlcad thought I saw a slew of file refs
03:41.10 *** join/#brlcad ChanServ (ChanServ@services.)
03:41.10 *** join/#brlcad Ralith (n=ralith@69.90.48.97) [NETSPLIT VICTIM]
03:41.10 *** join/#brlcad R0b0t1 (n=Enigma@unaffiliated/r0b0t1) [NETSPLIT VICTIM]
03:41.10 *** join/#brlcad cosurgi (n=cosurgi@atak.bl.pg.gda.pl) [NETSPLIT VICTIM]
03:41.10 *** mode/#brlcad [+o ChanServ] by irc.freenode.net
03:42.08 starseeker extract_chmLib does something with it...
03:42.40 starseeker what the bleep...
03:42.40 starseeker what is a .brs file
03:42.49 starseeker sees some htm and css in there...
03:43.48 starseeker oh, peachy
03:43.56 starseeker a lot of the htm files are zero size
03:44.02 starseeker fines a non-empty one...
03:44.13 brlcad them html2docbook? :)
03:44.22 starseeker actually...
03:44.49 starseeker the only actual content looks like what we're doing in system/man1/en - per-command htm files
03:44.58 starseeker reads...
03:46.09 starseeker what the hell
03:46.23 starseeker the non-zero length items are filled with spaces??
03:47.45 starseeker votes we delete that sucker
03:49.59 ``Erik wonder how legal eliminating trailing whitespace on ALL files would be
03:49.59 starseeker here's everything extrach_chmLib got out of it: http://pastebin.bzflag.bz/d7e84429e
03:50.16 ``Erik :%s/[ \t]*$//
03:50.48 ``Erik find . -type f | xargs sed -e -i.bak 's/[ ^I]+$//'
03:50.48 ``Erik ?
03:51.05 starseeker hmm?
03:51.11 starseeker what, run that on the results?
03:51.32 ``Erik any and all? I dunno, I'm not really reading what you're typing :D
03:51.40 starseeker ah
03:52.07 ``Erik ya said something about a non-zero length item that was just spaces? would terminating trailing whitespace make it a zero length item?
03:52.08 starseeker I mean no tool I have shows any indication of there being ANY useful content in that chm file
03:52.16 starseeker oh
03:52.32 starseeker maybe, but the idea was to find even one file that says something intelligent
03:53.04 starseeker sees no signs of it, and it looks like even what was once there will be supplanted by the man page work
03:53.30 starseeker brlcad: if you're still at work, can a Windows box do anything with it?
03:53.52 brlcad starseeker: i'll test here in a couple min
03:53.59 starseeker WOOT - Lisp in Small Pieces arrived
03:54.09 starseeker somewhat battered, but intact and fully legible
03:54.34 starseeker ``Erik: I'm battling back towards Lisp Respectibility
03:54.56 ``Erik heh
03:55.18 ``Erik y'know, at first, I was really excited, but then after reading and skimming, I'm less impressed
03:55.30 starseeker really?
03:55.43 starseeker not detailed?
03:55.45 ``Erik mebbe it's because I've had it sitting down for so long
03:55.58 ``Erik it had reasonable detail, but nothing staggering to it?
03:56.08 ``Erik I have it and pcl on my coffee table, though
03:56.17 ``Erik gets the girls all worked up, ch'know
03:56.36 starseeker I understood the idea was to document at a very low level what it took to make a lisp implementation
03:57.07 ``Erik yeh, but other texts have that, too
03:57.21 ``Erik it really doesn't take much at all
03:58.21 starseeker raises eyebrow
03:58.27 starseeker glad I didn't blow $100 on it then
03:58.51 starseeker nuts, shoulda waited longer and bought your copy off ya ;-)
04:04.06 ``Erik hehehe, I d'no, it may've been that I bought it to answer some pretty specific questions and wasn't satisfied to those specific answers *shrug* I d'no
04:05.08 brlcad hm, i don't want to log into windows.. leave it for bob later
04:07.44 ``Erik I'm telling him you said tht
04:08.10 ``Erik he'll park his truck ontop of your car in retaliation O.o
04:08.56 brlcad http://www.toplessrobot.com/2009/11/the_great_muppet_bohemian_rhapsody_cover.php
04:09.33 ``Erik I've seen topless robot before O.o
04:12.33 ``Erik hrm, is there a cognative difference between "sci-fi" and science fiction?
04:14.58 ``Erik ohhh, they had one about generating the computer graphics used in the original star wars trilogy, that's right
04:15.11 ``Erik using a projector and vector displays
04:15.19 ``Erik and giving up on using a computer at one point
04:37.26 starseeker ``Erik: isn't it sy-fi now or some such nonsense?
04:40.21 ``Erik I meant "sci-fi" the genre, not "SciFi" the tv network
04:40.30 ``Erik but tyat is "SyFy" now
04:40.43 ``Erik since that STOPPED SHOWING SCIENCE FICTION
04:40.51 ``Erik and just do horror shit and plain old shit now
04:57.41 *** join/#brlcad Aeamus (n=Enigma@unaffiliated/r0b0t1)
05:00.59 CIA-14 BRL-CAD: 03brlcad * r36691 10/brlcad/trunk/src/libtermio/termio.c: ifdef header test, not just if
05:39.29 brlcad woot, we now compile c99 strict (albeit with slews and slews of warnings)
05:41.08 Ralith I doubt there exists any major project that escapes that.
05:43.11 brlcad escapes what?
05:43.18 brlcad warnings?
05:44.55 brlcad even on a large codebase, have to control the entropy and complexity.. does get harder, but all the more reason I think it's important we get things clean, then set it up (with strict flags) so they can stay clean
05:45.32 brlcad pretty cool that about 10 main libs are fully clean now
05:46.39 brlcad next up are probably librt and libged.. then maybe work on finally kicking off a coverity scan
05:58.36 brlcad kicks off another zombie attack, this time to the extreme
05:59.14 brlcad 1 vs 17000
05:59.49 brlcad first few all failed miserably after less than a hundred .. but now this dude made it past the first wave and looks like he just might clean house
05:59.59 brlcad lets it run over the week/weekend
09:35.20 *** join/#brlcad Elrohir (n=kvirc@p5B14EF26.dip.t-dialin.net)
09:44.29 *** join/#brlcad d_rossberg (n=rossberg@BZ.BZFLAG.BZ)
09:47.10 CIA-14 BRL-CAD: 03d_rossberg * r36692 10/brlcad/trunk/include/bio.h: defined S_ISDIR for not-C99 compilers
11:20.05 d_rossberg i tried to test my last change on linux but i had no success:
11:21.13 d_rossberg i got an error in src/libbu/brlcad_path.c:
11:22.04 d_rossberg ../../include/brlcad_version.h:153: Error: »brlcad_ident« defed but not referenced
11:23.10 d_rossberg (it's a warning considered as an error)
11:42.45 *** join/#brlcad jaunty (n=IceChat7@dsl-241-98-87.telkomadsl.co.za)
12:06.58 *** part/#brlcad jaunty (n=IceChat7@dsl-241-98-87.telkomadsl.co.za)
13:47.05 *** join/#brlcad mafm_ (n=mafm@cpc2-bexl3-0-0-cust843.bmly.cable.ntl.com)
14:07.08 brlcad d_rossberg: hm, interesting .. wonder why you get that result
14:09.01 brlcad can you paste the exact compile line and error message?
14:12.15 brlcad suspect it's a "defined but not used" warning?
14:18.19 CIA-14 BRL-CAD: 03brlcad * r36693 10/brlcad/trunk/include/brlcad_version.h: attempt to quell a 'defined but not used' warning reported by daniel. provide a non-static function that calls brlcad_ident() so that it's used (and it calls brlcad_version() so we're good on the second static func).
14:19.05 brlcad see if that helps
14:29.48 d_rossberg now brlcad_path.c goes through
14:30.45 d_rossberg i wonder if the linker will complain about the multiple defined function brlcad_identify() ...
14:31.33 d_rossberg yes, it will!
14:33.43 d_rossberg see http://pastebin.bzflag.bz/m23a44768
14:42.59 brlcad yeah, I wondered about that too.. :)
14:43.22 brlcad hm!
14:43.50 d_rossberg i'm working on a solution: splitting brlcad_version.h into 2 files (?)
14:44.20 brlcad need c++ so could do a static var initializer
14:46.34 CIA-14 BRL-CAD: 03brlcad * r36694 10/brlcad/trunk/include/brlcad_version.h: yeah, bad idea. multiple definitions in the lib since it's not in a compilation unit.
14:46.57 d_rossberg i try one header with the constants and an other header with the static function
14:48.45 brlcad interestingly, there's already a quelling in there to prevent the "this is not used" warning
14:49.09 brlcad gcc apparently got a little smarter and realized it was a trick
14:52.04 d_rossberg this is the "defined but not referenced" message is got: http://pastebin.bzflag.bz/m5b3ec115
14:52.53 d_rossberg gcc (Debian 4.3.4-6) 4.3.4
14:54.38 CIA-14 BRL-CAD: 03brlcad * r36695 10/brlcad/trunk/include/brlcad_version.h:
14:54.38 CIA-14 BRL-CAD: attempt #3: quell the use warning by having brlcad_version() call
14:54.38 CIA-14 BRL-CAD: brlcad_ident(). this creates a recursive cycle that should call both functions
14:54.38 CIA-14 BRL-CAD: the second time through regardless of which is called.. at least that's the
14:54.39 CIA-14 BRL-CAD: theory.
14:54.52 brlcad ahh, heh, in german .. awesome
14:56.06 brlcad bah, my logic isn't sound .. infinite loop
14:59.24 brlcad fixing
15:00.47 d_rossberg my make is still working with the latest version ...
15:02.16 brlcad yeah, it'll build
15:02.19 brlcad it's a run-time inf loop
15:02.40 brlcad flawed double recursion
15:05.29 d_rossberg i have a new one: http://pastebin.bzflag.bz/m1df3620a
15:09.45 CIA-14 BRL-CAD: 03brlcad * r36696 10/brlcad/trunk/include/brlcad_version.h:
15:09.45 CIA-14 BRL-CAD: yeah, previous logic was flawed. would have resulted in a run-time double
15:09.45 CIA-14 BRL-CAD: recursion inf loop. this form should work regardless of which function is
15:09.45 CIA-14 BRL-CAD: called first and has both functions calling the other quelling any non-use
15:09.45 CIA-14 BRL-CAD: warnings.
15:12.31 CIA-14 BRL-CAD: 03brlcad * r36697 10/brlcad/trunk/src/libpkg/pkg.c: quell size_t ->printf-> long warnings
15:23.15 CIA-14 BRL-CAD: 03d_rossberg * r36698 10/brlcad/trunk/src/libpkg/pkg.c: one cast was missing (size_t->long warning)
15:44.00 d_rossberg looks like there are no more errors during make
15:52.45 brlcad cool
15:52.53 brlcad cya daniel
15:56.26 CIA-14 BRL-CAD: 03brlcad * r36699 10/brlcad/trunk/src/libbu/ptbl.c: cast %p format args to void*, quellage
16:00.57 CIA-14 BRL-CAD: 03brlcad * r36700 10/brlcad/trunk/src/libbu/semaphore.c: more warnings to quell...
16:02.41 *** join/#brlcad Elrohir (n=kvirc@91.20.239.38)
16:23.39 ``Erik shudders at the notion of gettext-izing BRL-CAD
16:26.05 CIA-14 BRL-CAD: 03brlcad * r36701 10/brlcad/trunk/src/libbn/ (plane.c sphmap.c tabdata.c wavelet.c): whee... MORE warnings... where do these keep coming from?? this same machine compiled verbose strict a couple days ago.
16:26.23 brlcad yeah, no fun
16:30.35 CIA-14 BRL-CAD: 03brlcad * r36702 10/brlcad/trunk/include/bn.h: cast our args to what we're logging them as
16:31.05 CIA-14 BRL-CAD: 03brlcad * r36703 10/brlcad/trunk/TODO: need to rename bu_ptbl() .. conflicts with c++ compilation
16:34.23 CIA-14 BRL-CAD: 03brlcad * r36704 10/brlcad/trunk/src/nirt/ (8 files): misc cleanup, ws, indent, quellage
17:26.48 *** join/#brlcad Ralith (n=ralith@d142-058-083-015.wireless.sfu.ca)
18:13.41 *** join/#brlcad Ralith_ (n=ralith@142.58.90.101)
19:38.28 *** join/#brlcad Ralith (n=ralith@69.90.48.97)
20:20.37 starseeker ``Erik: is gettext-izing BRL-CAD on the todo?
20:34.36 starseeker oh lovely
20:34.53 starseeker autoconf 2.64 doesn't like BRL-CAD
20:35.09 starseeker pulls a clean checkout to be sure...
20:49.23 starseeker yep
20:50.43 starseeker oh, looks like it might be griping about tkhtml3?
20:50.58 starseeker added AUTHORS, ChangeLog, INSTALL, COPYING and NEWS...
20:51.02 starseeker humph
20:52.54 CIA-14 BRL-CAD: 03starseeker * r36705 10/brlcad/trunk/m4/: Tweak svn:ignore in m4 dir
20:54.05 CIA-14 BRL-CAD: 03starseeker * r36706 10/brlcad/trunk/src/other/step/m4/: Tweak svn:ignore in src/other/step m4 dir
20:59.06 starseeker that's not all of it... six instances of complaining by grep about Makefile.am not found
21:01.45 CIA-14 BRL-CAD: 03starseeker * r36707 10/brlcad/trunk/src/other/tkhtml3/ (AUTHORS COPYING COPYRIGHT ChangeLog INSTALL NEWS): Move COPYRIGHT file to COPYING and add in stubs of other 'standard' files to quiet autoreconf blathering.
21:03.54 starseeker or autogen, more properly...
21:28.05 ``Erik not on the todo, but i10n/l18n ... done pennance on dem
21:48.11 *** join/#brlcad talcite (n=matthew@dhcp-143-147.mcme-students.carleton.ca)
23:17.42 *** join/#brlcad R0b0t1 (n=Enigma@unaffiliated/r0b0t1)

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