IRC log for #brlcad on 20140204

01:42.30 *** join/#brlcad FOSScookie (~brian@107-200-34-31.lightspeed.tulsok.sbcglobal.net)
01:44.28 maths22 brlcad: I feel like I should be doing something for you all, but I'm not sure what
01:44.36 maths22 I just switched computers
02:03.38 *** join/#brlcad konro (~konro@41.205.22.53)
02:25.32 brlcad maths22: getting the new website online would be awesome :)
02:26.42 maths22 can you tell me some steps that need doing?
02:26.51 maths22 Obviously, the header is right now a problem
02:27.07 maths22 Also, I will add the images
02:28.00 brlcad I'd think first up is to review all the GCI-submitted website tasks (the wp themes) and review/merge them together
02:28.17 brlcad anita's was quite impressive iirc
02:28.32 brlcad they all had interesting aspects
02:29.01 maths22 I think I right now am running anita's
02:29.18 brlcad do you have a sense for what the major differences were?
02:29.41 brlcad there were different quality menus
02:30.30 brlcad two(?) were responsive design, but anita's menu overflowed and only had a simple underline
02:30.40 maths22 I can't right now load the melange completed tasks list
02:30.46 maths22 That is my problem with hers
02:31.10 brlcad that's why I think the first step is to create a theme that takes the best from all of them
02:31.44 brlcad or maybe even just shove all three into SVN and create a fourth were features are manually pulled in as cleanly as possible, but starting as simple as possible
02:31.49 maths22 That would make sense
02:32.16 maths22 I personally would use git for that, unless there is a compelling reason to create an svn repo
02:32.37 maths22 Given that I don't have comitt access to the one with a web dir
02:32.42 maths22 (not that I should yet)
02:33.07 brlcad actually you do
02:38.51 brlcad you have the ability to commit to the repo now (I mean c'mon, you've earned it)
02:39.12 brlcad have had it for a while, but haven't been on hand to talk :)
02:39.35 brlcad that said, git vs svn isn't interesting or important to me right now -- getting the site up is :)
02:40.06 brlcad if git makes that easier for you personally, no problem
02:40.22 brlcad git or svn would be fine, but I'd still want it to all push to at least sf.net if not our sf and github project pages eventually
02:40.27 maths22 I guess I did more code work this year than I realized
02:41.34 maths22 SVN is fine for me then
02:41.51 maths22 Should I work in a branch or clean out the version from 2010?
02:41.59 brlcad doesn't matter
02:42.22 maths22 I think I will get rid of the 2010 stuff. It isn't doing anyone any good
02:42.29 maths22 I think
02:42.59 brlcad htdocs is the "web root"
02:43.30 brlcad everything above it was documentation for the checkout, a place to put content not intended to go into the root
02:43.36 brlcad like documentation
02:44.04 brlcad it'd be good to commit a version of everything currently live on the website before deleting it
02:44.12 brlcad just to have a backup
02:44.51 brlcad I could work on doing that if you're not that familiar with svn yet
02:45.55 brlcad or just take up the flag and make it yours - I can do a manual backup instead just as well
02:46.09 brlcad just let me know before files on brlcad.org are actually about to change
02:49.03 maths22 I updated all the doc files
02:49.37 maths22 I was thinking I would comitt those, and then commit the current version (right now the real site and the svn repo have nothing to do with each other)
02:50.06 maths22 could you set up a beta.brlcad.org (or something like that) subdomain?
02:50.48 brlcad yeah, commit early, commit often
02:50.49 brlcad never fear committing frequently ;)
02:51.23 brlcad and yes, subdomains are set up easily -- just let me know what you want
02:52.24 maths22 beta.brlcad.org so I can test without breaking the site or working in a subdir that will mess up links
02:53.09 Notify 03BRL-CAD:maths22 * 59663 (web/trunk/AUTHORS web/trunk/README): Updated AUTHORS to include maths22Updated README to match the current mysql-backed website
02:53.58 maths22 first commit!
02:54.35 maths22 should I comitt the whole current www version
02:55.03 maths22 all files (or should some be excluded from version control besides wp [that is what I will replace it with])
03:09.05 maths22 brlcad: also, I only remember 2 wp theme tasks:
03:09.10 maths22 http://www.google-melange.com/gci/task/view/google/gci2013/4671617942683648
03:09.11 gcibot Create Wordpress theme for BRL-CAD website
03:09.11 gcibot Status: Closed
03:09.11 gcibot Mentor(s): Sean
03:09.15 maths22 http://www.google-melange.com/gci/task/view/google/gci2013/5956421170495488
03:09.16 gcibot Create Wordpress theme for BRL-CAD website #2
03:09.16 gcibot Status: Closed
03:09.16 gcibot Mentor(s): Sean
03:09.27 maths22 Apparently gcibot is still live
03:10.01 brlcad http://beta.brlcad.org/ is in place
03:11.10 brlcad I left the instructions in place for /d /viewvc /buildbot and /xref
03:12.47 brlcad ideally, the current www version is committed, but there is almost certainly content that should not be committed for a variety of reasons (legal and otherwise)
03:13.08 brlcad if you can sort out what's actually in use / active, that would be good
03:13.26 *** join/#brlcad FOSScookie (~brian@107-200-34-31.lightspeed.tulsok.sbcglobal.net)
03:13.53 brlcad maths22: you know what, let's not obsess on the past for web work
03:14.15 brlcad it's not like our C code heritage and it's far more lacking in merit
03:14.54 brlcad if you can leave the current web root alone (on the filesystem), that's good enough for archival
03:15.23 maths22 ok
03:15.27 brlcad new root is a new empty beta.brlcad.org dir -- we can identify and merge in new content as needed
03:15.30 maths22 thanks
03:15.30 brlcad so you can start fresh
03:15.37 maths22 what do you mean by instructions
03:15.45 brlcad and do whatever you like with the svn module
03:16.04 brlcad instructions == the apache directives
03:16.45 maths22 ok
03:17.24 maths22 I will eventually figure out a way for xref to auto-update (cron job)?
03:17.38 maths22 I'm not going to add that to svn (for now at least)
03:18.12 maths22 I'm thinking I'll bring in gallery as is (after I check it for passwords)
03:21.20 maths22 actually, I won't
03:21.27 maths22 I want to do it as cleanly as possible
03:21.28 brlcad not using a wp gallery extension?
03:21.34 maths22 that is what I am realizing
03:21.41 maths22 I should build from the ground up
03:22.14 brlcad heh, no fewer than 40 options in the wp plugins directory
03:24.00 Notify 03BRL-CAD:maths22 * 59664 web/trunk/README: Added missing close parend
03:25.00 Notify 03BRL-CAD:maths22 * 59665 (web/trunk/htdocs/.htaccess =================================================================== and 6 others): Cleaned out htdocs of all old files so we can start new and only reuse oldcontent as desired.
03:26.18 brlcad maths22: the top-level docs aren't sacred
03:26.33 brlcad you're welcome to rework them or eliminate or change as needed
03:26.43 brlcad just need some place to stash files not in the web root
03:26.48 brlcad however you achieve that is fine with me
03:27.05 brlcad if what's there works, great
03:27.21 brlcad otherwise, you should make it work :)
04:10.18 Notify 03BRL-CAD:brlcad * 59666 brlcad/trunk/src/libged/simulate/simrt.c: rewrite to not peek into vls internals
04:15.21 *** join/#brlcad konro (~konro@41.205.22.53)
04:34.08 Notify 03BRL-CAD:brlcad * 59667 brlcad/trunk/include/bu.h: undo r59655 as the 'offset' most definitely should not be signed. if it's ever a negative value, our routines will heartily crash. the msvc warning was almost certainly due to trickery in bu_vls_nibble() which lets one unroll an offset back to zero (i.e., it's a false positive, but a really hard one to detect). need to fix the nibbler. src/libbu
04:34.26 Notify 03BRL-CAD:brlcad * 59668 brlcad/trunk/src/libbu/vls.c: undo r59655 as the 'offset' most definitely should not be signed. if it's ever a negative value, our routines will heartily crash. the msvc warning was almost certainly due to trickery in bu_vls_nibble() which lets one unroll an offset back to zero (i.e., it's a false positive, but a really hard one to detect). need to fix the nibbler.
04:39.50 *** join/#brlcad konro_ (~konro@41.205.22.53)
04:50.47 *** join/#brlcad FOSScookie (~brian@107-200-34-31.lightspeed.tulsok.sbcglobal.net)
05:03.26 Notify 03BRL-CAD:brlcad * 59669 (brlcad/trunk/include/bu.h brlcad/trunk/src/libbu/vls.c): rewrite bu_vls_nibble() so that it only performs unsigned math on the size_t struct elements. this should appease compilers that don't realize we already accounted for negative nibbles that prevent vls_offset from ever underflowing beyond zero. this is an overall readability cleanup anyways with comments added for good measure. we
05:03.28 Notify make the len paramter an off_t to reflect it's offset characteristic (i.e., intentionally signed), even though the vls_offset field is intentionally unsigned (unfortunate pre-c99 name).
05:08.41 *** join/#brlcad konro (~konro@41.205.22.53)
05:20.54 *** join/#brlcad konro_ (~konro@41.205.22.53)
05:27.40 *** join/#brlcad konro (~konro@41.205.22.53)
05:33.03 Notify 03BRL-CAD:brlcad * 59670 brlcad/trunk/src/libged/joint.c: convert the joint error handler that prints a ---^ errror indicator line from knowing anything about vls internals. looks like the logic that was there was actually incorrect in a number of ways (using offset for length, printing vls_str without accounting for offset, etc), so this should actually make it right too.
05:37.12 Notify 03BRL-CAD:brlcad * 59671 (brlcad/trunk/CHANGES brlcad/trunk/include/bu.h and 7 others): remove the duplicitous bu_vls_trunc2() function in favor of bu_vls_trunc(). difference was trunc2 ignoring negative values whereas trunc treated them as offsets from the end (instead of a length). instances are trivially updated to using trunc() so just remove it wholesale as minimally impacting.
05:40.47 Notify 03BRL-CAD:brlcad * 59672 brlcad/trunk/src/libbu/tests/bu_vls.c: meh, just do nothing
05:44.45 Notify 03BRL-CAD:brlcad * 59673 brlcad/trunk/src/conv/step/g-step/g-step.cpp: removed unused vars
06:56.20 *** join/#brlcad konro (~konro@41.205.22.53)
07:19.54 *** join/#brlcad archivist (~archivist@host81-149-189-98.in-addr.btopenworld.com)
07:46.18 *** join/#brlcad luca79 (~luca@net-37-117-179-126.cust.vodafonedsl.it)
08:02.05 *** join/#brlcad konro (~konro@41.205.22.53)
08:51.40 *** join/#brlcad FreezingCold (~FreezingC@135.0.41.14)
09:33.59 *** join/#brlcad konro (~konro@41.205.22.53)
11:26.17 *** join/#brlcad Anaphaxet0n (~george@unaffiliated/anaphaxeton)
12:01.46 *** join/#brlcad FreezingCold (~FreezingC@205.211.50.161)
12:04.15 *** join/#brlcad d_rossberg (~rossberg@66-118-151-70.static.sagonet.net)
12:43.28 *** join/#brlcad FOSScookie (~brian@107-200-34-31.lightspeed.tulsok.sbcglobal.net)
12:46.12 *** join/#brlcad FreezingCold (~FreezingC@205.211.50.163)
13:43.46 *** join/#brlcad luca79 (~luca@net-37-117-179-126.cust.vodafonedsl.it)
14:41.14 *** join/#brlcad luca79 (~luca@net-37-117-179-126.cust.vodafonedsl.it)
17:03.38 *** join/#brlcad gaganjyot (~gagan@125.62.99.0)
19:04.32 *** join/#brlcad Notify (~notify@66-118-151-70.static.sagonet.net)
19:05.31 Notify 03BRL-CAD:n_reed * 59674 brlcad/trunk/src/libbrep/intersect.cpp: add XEventProxy class to make creation of ON_X_EVENTs more readable
19:05.52 Notify 03BRL-CAD:brlcad * 59675 brlcad/trunk/src/other/libregex/regex.h: irix defines _OFF_T
19:06.23 Notify 03BRL-CAD:n_reed * 59676 brlcad/trunk/src/libbrep/intersect.cpp: move overlap boundary curve test to function
19:06.38 Notify 03BRL-CAD:r_weiss * 59677 brlcad/trunk/src/libbu/uce-dirent.h: Change to compensate for the uce-dirent.h assuming windows does not have a direct.h header.
19:06.42 Notify 03BRL-CAD Wiki:WmRobt-USER * 0 /wiki/User:WmRobt-USER:
20:03.23 *** join/#brlcad gaganjyot (~gagan@125.62.99.0)
21:55.07 Notify 03BRL-CAD:starseeker * 59678 brlcad/trunk/src/libbu/tests/CMakeLists.txt: bitv-tests is a CMake file - set the extension accordingly
22:02.45 Notify 03BRL-CAD:starseeker * 59679 brlcad/trunk/src/libbu/tests/test_internals.h: Start adding the DLL foo that Windows MSVC needs...
22:07.41 Notify 03BRL-CAD:starseeker * 59680 brlcad/trunk/src/libbu/tests/test_internals.h: Try to match the CMake generated define.
22:29.14 *** join/#brlcad FreezingCold (~FreezingC@135.0.41.14)
23:29.56 Notify 03BRL-CAD:starseeker * 59681 (brlcad/trunk/src/librt/search.c brlcad/trunk/src/librt/search.h): Checkpoint some debugging work on the -below function in search - it isn't reporting a comb based on the boolean test that I would expect it to report.

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