| 00:01.16 | ``Erik | mmmm, pastrami sandwich and a salad |
| 00:01.30 | CIA-73 | BRL-CAD: 03starseeker * r38615 10/brlcad/trunk/src/tclscripts/archer/ (Archer.tcl LoadArcherLibs.tcl): Add some (commented out) code exploring the use of TkTreeCtrl in the html help viewer. This widget offers more flexibility than ttk::treeview, but it remains to be seen if it's worth using it. |
| 00:13.35 | *** join/#brlcad talcite (~matthew@bas4-toronto21-1176312364.dsl.bell.ca) |
| 00:26.50 | ``Erik | so when megan fox was auditioning for her part in transformers (as a nobody), bay made her, get this, wash his ferrari. while he taped. and then he lost the tape. O.o crazy, ainnit? |
| 00:27.20 | starseeker | is that legal? |
| 00:27.29 | ``Erik | probably not... that's why it was in the news |
| 00:27.37 | ``Erik | http://www.nypost.com/p/pagesix/item_HCXmPrelfpVtpJnd9lshgP;jsessionid=358FACAEAE8AE72BBE54191D9BDC2AB2 |
| 00:27.58 | starseeker | shakes head |
| 00:29.33 | ``Erik | even if it's legal, still seems a bit... scummy :) |
| 00:29.44 | ``Erik | especially the part about that video not being made public *cough* O:-) |
| 00:30.46 | ``Erik | only knows about it after seeing this webcomic: http://hijinksensue.com/2009/07/13/a-fetish-revealed/ |
| 00:41.15 | brlcad | considering one of the first scenes of her in the movie is of her washing a car, seems perfectly reasonable to me |
| 00:41.48 | starseeker | ah. (haven't seen the movie) |
| 00:45.00 | ``Erik | from autoblog.com: Since the role of Mikaela Banes required Fox to act as sexy as she already looks, we can see why Bay chose for her to wash a Ferrari as part of the audition, but the fact that the 'audition' took place at his home and the tape has gone missing does seem a bit "casting couch" to us. |
| 00:52.12 | brlcad | he probably wishes, heh |
| 01:06.00 | *** join/#brlcad Nohla (~jesica@201.255.230.147) |
| 01:45.27 | CIA-73 | BRL-CAD: 03starseeker * r38616 10/brlcad/trunk/ (4 files in 2 dirs): |
| 01:45.27 | CIA-73 | BRL-CAD: Less than ideal, but looks like it may function - use an html list of html |
| 01:45.27 | CIA-73 | BRL-CAD: articles with links as the left hand side of the help browser, and activate the |
| 01:45.27 | CIA-73 | BRL-CAD: hyperlinks to display the resulting page in the right html viewer. List is not |
| 01:45.27 | CIA-73 | BRL-CAD: complete yet and need a 'cover sheet' introduction page, but may serve to get |
| 01:45.28 | CIA-73 | BRL-CAD: things working in an initial cut. |
| 01:46.48 | starseeker | notes he has some fixing up to do of the image links... |
| 01:46.59 | brlcad | cool |
| 01:47.10 | starseeker | making screen shot, one sec... |
| 01:50.13 | starseeker | http://bzflag.bz/~starseeker/html_help_viewer.png |
| 01:51.18 | starseeker | doesn't have the glitz and glamour of a tree widget, but it does function |
| 01:54.48 | brlcad | looking pretty snazzy |
| 01:56.02 | brlcad | that looks plenty nice actually |
| 01:56.37 | starseeker | cool, thanks - guess that means tktreectrl won't be needed after all |
| 01:59.42 | brlcad | is the left side all html? |
| 02:00.21 | starseeker | yes |
| 02:00.33 | starseeker | (well, 'cept for the scrollbars |
| 02:00.54 | CIA-73 | BRL-CAD: 03starseeker * r38617 10/brlcad/trunk/src/tclscripts/archer/Archer.tcl: Oh yeah, this dialog should be modality none |
| 02:01.06 | brlcad | cool, that's possibly even better then .. more flexible |
| 02:01.23 | starseeker | kinda hard to collapse a particular "subtree" though |
| 02:01.45 | brlcad | could implement a far more complicated interface if needed with some simple css/javascript |
| 02:02.15 | brlcad | or even plan old-school hyperlinks, ul's, tables, etc |
| 02:02.24 | starseeker | winces - true in theory, but we've never put tkhtml's css support to the test, and javascript would involve importing a lot more of hv3 and it's requirements |
| 02:03.18 | brlcad | could also fake it with a simple cgi/generator backend that controls the menu |
| 02:03.26 | starseeker | nods |
| 02:03.36 | starseeker | if it's needed - I suppose for a first cut it really isn't |
| 02:03.42 | brlcad | yep |
| 02:03.50 | brlcad | it's loads better already |
| 02:04.10 | brlcad | the only feature that's really missing/needed is some search capability |
| 02:04.38 | starseeker | nods - that one will probably involve the most "grunt work" |
| 02:05.31 | starseeker | hv3 might have some relevant code we could snarf, I suppose... don't know if they have any search functionality |
| 02:06.17 | brlcad | actually, I bet you could make a really good search mechanism with tcl pretty simply -- hash all of the html pages into a tcl list, search the lists and incrementally display results as a filtered menu |
| 02:07.56 | brlcad | yeah, something like this: http://wiki.tcl.tk/3751 |
| 02:07.57 | ``Erik | baysian scoring, pheer O.o |
| 02:09.33 | brlcad | straight up keyword would be sufficient |
| 02:10.32 | brlcad | show the resulting article matches, similar to searching a pdf with preview or acrobat |
| 02:10.33 | ``Erik | for now, but having "scoring" like mac help or winders help would be nifty some day :D |
| 02:10.38 | starseeker | notes that switching from one html page to another is a tad slow... wonder if they should be "pre-cached" |
| 02:11.16 | ``Erik | if you go from a->b, then b->a, is the b->a fast? |
| 02:11.59 | starseeker | not really - I wouldn't expect it to be |
| 02:12.00 | brlcad | starseeker: how slow? |
| 02:12.22 | starseeker | maybe a second or so |
| 02:12.34 | brlcad | do you know what part is slow? |
| 02:12.41 | brlcad | rendering? file i/o? |
| 02:12.48 | starseeker | not sure |
| 02:12.49 | brlcad | searching some list? |
| 02:12.59 | brlcad | redraw of the menu |
| 02:13.05 | brlcad | could be something simple |
| 02:13.17 | starseeker | suspects it might be loading large images |
| 02:13.21 | ``Erik | (the a->b->a test shoulda been fast on b->a if it was i/o, which is how pre-loading the cache would help) |
| 02:13.37 | starseeker | ``Erik: cept I'm not caching anything anywhere |
| 02:13.41 | starseeker | full reload each time |
| 02:13.53 | ``Erik | your OS should do some caching for you... :D |
| 02:14.06 | starseeker | I don't think it's a show stopper at this point |
| 02:14.23 | starseeker | more important to get the image links straightened out |
| 02:14.33 | starseeker | (pops up a nice tcl error window if it can't find one...) |
| 02:14.34 | brlcad | yeah, that's not too horrible at a sec .. especially if there's any feedback |
| 02:14.51 | brlcad | could display a busy/loading icon/text/whatever |
| 02:15.03 | starseeker | the images are way too big anyhow - I gummed that up the first time around |
| 02:15.05 | brlcad | or progress bar if you it's parcelable |
| 02:15.22 | starseeker | has to re-extract them from the word docs again |
| 02:15.28 | starseeker | do it right this time |
| 02:15.38 | brlcad | or from the original source |
| 02:15.48 | starseeker | if we have it... |
| 02:15.52 | brlcad | we have it |
| 02:16.04 | starseeker | cool - that'd be best, of course |
| 02:16.33 | brlcad | finding it might take a little digging, but we definitely have it |
| 02:16.54 | starseeker | iirc, the "save image as" from Word was doing a resolution down sampling (why???) and I tried to work around it by doing big screenshots |
| 02:17.16 | starseeker | I stumbled onto another way later - maybe using Preview? - that worked much better |
| 02:17.42 | starseeker | but the originals would avoid that mess altogether |
| 02:18.08 | brlcad | vol II ? |
| 02:18.34 | starseeker | II and III |
| 02:19.06 | starseeker | III is more important - most of the ones in II I can re-create if need be |
| 02:19.22 | starseeker | (might be nice to have an MGED version > 5 in the screenshots...) |
| 02:19.44 | brlcad | uploads |
| 02:23.22 | starseeker | This kinda puts the spotlight on the "MGED manual" in the html directory and the distinctions between it and the Vol II material - it's gonna look pretty redundant |
| 02:28.29 | brlcad | so good news |
| 02:28.31 | brlcad | is I found the images |
| 02:28.35 | brlcad | bad news |
| 02:28.47 | brlcad | is they're mixed together with images that weren't used too |
| 02:28.54 | starseeker | urk |
| 02:29.09 | brlcad | and there are lots of them |
| 02:29.11 | CIA-73 | BRL-CAD: 03starseeker * r38618 10/brlcad/trunk/doc/docbook/articles/en/build_pattern.xml: Er, oops - correcting the title... |
| 02:29.14 | brlcad | hundreds :) |
| 02:29.26 | starseeker | brlcad: heh. OK, tomorrow for that then |
| 02:29.37 | brlcad | hm, including tons that are PNG files without a png extension.. lovely |
| 02:30.20 | starseeker | still has to recreate some figures with inkscape and graphviz |
| 02:30.43 | starseeker | iirc, there were a few "text figures" that didn't lend themselves to html conversion very well |
| 02:30.57 | brlcad | http://brlcad.org/design/archer/images/vol1 |
| 02:32.36 | starseeker | I've actually got to head home pretty quick... |
| 02:33.09 | starseeker | did anyone do Vol I as docbook? |
| 02:33.12 | starseeker | looks... |
| 02:33.47 | starseeker | hmm |
| 02:34.04 | starseeker | well, that should be easy to (re)write in Docbook |
| 02:34.21 | brlcad | it's already html -- pretty simple |
| 02:34.30 | starseeker | ah |
| 02:34.31 | brlcad | also have translations for it |
| 02:34.51 | starseeker | http://brlcad.org/wiki/Overview ? |
| 02:35.01 | brlcad | that's basically the core that survived |
| 02:36.09 | starseeker | we may have to update that CSG focus a bit with NURBS coming online :-) |
| 02:38.00 | starseeker | packs it in while he's still able to drive |
| 02:40.47 | brlcad | vol2 uploading now |
| 02:43.11 | brlcad | fixed all the png files that no extension |
| 02:43.41 | brlcad | done |
| 02:43.44 | brlcad | http://brlcad.org/design/archer/images/vol2 |
| 02:44.56 | brlcad | some of the images that come right to mind that were eventually cut or just glancingly considered were the dog diagrams and paper airplane images |
| 02:45.13 | brlcad | have fun sorting it all out |
| 03:01.56 | brlcad | vol3 mostly up |
| 03:47.24 | starseeker | O.o |
| 03:54.36 | brlcad | there's some more in other formats (e.g., ppt, doc), but won't upload without individually reviewing so let me know if something is missing and needed |
| 03:54.47 | starseeker | k, thanks :-) |
| 03:55.07 | starseeker | that should be plenty to start |
| 04:55.34 | *** join/#brlcad Ralith (~ralith@69.90.48.97) |
| 06:33.46 | *** join/#brlcad ``Erik (~erik@c-69-140-109-104.hsd1.md.comcast.net) |
| 07:52.15 | *** join/#brlcad Ralith (~ralith@69.90.48.97) |
| 10:09.58 | *** join/#brlcad mafm (~mafm@81.35.69.130) |
| 10:53.25 | *** join/#brlcad Nohla (~jesica@201.255.230.147) |
| 12:44.32 | d-lo | Mernin all |
| 13:02.54 | ``Erik | yargh |
| 13:04.48 | d-lo | /usr/include/sys/types.h:198: error: conflicting declaration ?typedef long int int64_t? |
| 13:04.51 | d-lo | /home/dloman/include/brlcad/pstdint.h:456: error: ?int64_t? has a previous declaration as ?typedef long long int int64_t? |
| 13:04.54 | d-lo | /usr/include/unistd.h:238: error: conflicting declaration ?typedef __intptr_t intptr_t? |
| 13:04.58 | d-lo | /home/dloman/include/brlcad/pstdint.h:715: error: ?intptr_t? has a previous declaration as ?typedef int64_t intptr_t? |
| 13:05.01 | d-lo | /usr/include/unistd.h:238: error: conflicting declaration ?typedef __intptr_t intptr_t? |
| 13:05.04 | d-lo | /home/dloman/include/brlcad/pstdint.h:715: error: ?intptr_t? has a previous declaration as ?typedef int64_t intptr_t? |
| 13:05.07 | d-lo | issue :( |
| 13:05.20 | ``Erik | os? |
| 13:05.34 | d-lo | RHEL |
| 13:06.28 | ``Erik | sucks to be you |
| 13:06.28 | ``Erik | :D |
| 13:07.00 | ``Erik | pstdint.h is coming after types,h and unistd.h, right? are the defines being set correctly? |
| 13:08.48 | ``Erik | <-- starts a fresh build on rhel5/x86_64 |
| 13:12.39 | ``Erik | compiled fine here :/ |
| 13:15.02 | CIA-73 | BRL-CAD: 03erikgreenwald * r38619 10/brlcad/trunk/src/librt/primitives/metaball/metaball.c: fix uninitialized variable warning |
| 13:35.45 | d-lo | fresh build of brlcad? |
| 13:36.38 | d-lo | I am building rt3, and these errors came out of the blue. Dunno what caused it. |
| 13:46.05 | ``Erik | oh, I did BRL-CAD, yes... hrm, the RHEL machine lacks cmake |
| 13:46.38 | d-lo | trying an uninstall/recompile/install of BRLCAD. Mebbe I am missing something new :/ |
| 13:47.25 | ``Erik | it might be that pstdint.h is expecting defines from the configure.ac that cmake is not providing? |
| 13:48.58 | ``Erik | hm, how do I add include and library paths to cmake? |
| 13:49.05 | ``Erik | it's not finding my X |
| 13:58.36 | ``Erik | is there a way to do an "out of dir" build with cmake? I make a machine specific dir in tmp, cd -, cmake $HOME/src/rt^3/CMakeLists.txt, but it generated the makefiles with the source instead of where I wanted :/ |
| 13:58.53 | d-lo | ``Erik: in the top level CMakelists.txt, there are INCLUDE_SEARCH_PATHS and LIB_SEARCH_PATHS vars you can manipulate |
| 13:59.04 | d-lo | yes there is, but I haven't setup anything for it yet |
| 13:59.50 | ``Erik | hm, those vars have my X dirs in them (/usr/local/include and /usr/local/lib), but it doesn't find them when compiling. :/ |
| 14:00.10 | d-lo | what part is fialing to compile? |
| 14:00.15 | d-lo | lol, failing even |
| 14:00.43 | ``Erik | ohhhh, OIS, that'd be the uh, ogre crap |
| 14:00.47 | CIA-73 | BRL-CAD: 03starseeker * r38620 10/brlcad/trunk/doc/html/toc.html: Rearrange a bit, add the Books (need to test) |
| 14:01.01 | d-lo | yeah, I haven't bothered trying to wire up all that jazz |
| 14:01.15 | d-lo | well, some of /other is wired in but not all. |
| 14:01.16 | ``Erik | ok, compiling your stuff for dummies, chapter 1... go |
| 14:01.54 | d-lo | should be as simple as: |
| 14:02.10 | d-lo | 1) cd to $whatever/rt3 |
| 14:02.14 | d-lo | 3) cmake . |
| 14:02.31 | d-lo | nice, I skipped 2 |
| 14:02.36 | d-lo | 4) make |
| 14:02.52 | ``Erik | will just assume 2) svn up |
| 14:03.03 | ``Erik | that hits the X11 issue in OIS |
| 14:03.25 | ``Erik | or, it did from my last checkout, I'll try again before I say anything else O:-) |
| 14:03.50 | d-lo | I havent' touched OIS cmake stuff in a while, probably still exists |
| 14:04.56 | ``Erik | hm, in your cmake, it searches for libtkimg, that's gone now... tkpng replaced it |
| 14:05.10 | d-lo | kk |
| 14:06.50 | ``Erik | only failures on fbsd8/x86_32 seem to be in src/other/ois/ |
| 14:06.55 | ``Erik | :/ |
| 14:07.44 | d-lo | I think I see it. |
| 14:08.25 | d-lo | I am pretty syre the X11 paths are put into RT3_INCLUDE_DIR at the top level CMakelists.txt but that var isn't included with the OIS target. |
| 14:08.32 | d-lo | lemme push in a 'fix' and see if it works |
| 14:09.05 | ``Erik | compiles cmake on a rhel5 box |
| 14:10.28 | CIA-73 | BRL-CAD: 03davidloman * r38621 10/rt^3/trunk/cmake/FindBRLCAD.cmake: dropped the search for libtkimg and replaced it with libtkpng |
| 14:12.25 | CIA-73 | BRL-CAD: 03davidloman * r38622 10/rt^3/trunk/src/other/ois/CMakeLists.txt: Add rt3 includes into OIS build. |
| 14:12.34 | CIA-73 | BRL-CAD: 03erikgreenwald * r38623 10/brlcad/trunk/src/tclscripts/hv3/Makefile.am: add the missing backslash so EXTRA_DIST actually has some info to it |
| 14:13.03 | d-lo | ``Erik: There, see if tha tmakes OIS happy :/ |
| 14:19.09 | CIA-73 | BRL-CAD: 03davidloman * r38624 10/rt^3/trunk/src/utility/CMakeLists.txt: Add policy to enforce new link style directories. |
| 14:22.18 | ``Erik | same issue :/ *shrug* |
| 14:23.17 | d-lo | wel thanks for trying anyways :/ |
| 14:23.27 | d-lo | I'll add that to a list of 'thingados' |
| 14:24.44 | CIA-73 | BRL-CAD: 03starseeker * r38625 10/brlcad/trunk/doc/docbook/books/en/ (2 files in 2 dirs): That table is a little tricky to get as Docbook - punt for now and make it an image. |
| 14:26.45 | CIA-73 | BRL-CAD: 03starseeker * r38626 10/brlcad/trunk/doc/docbook/books/en/images/tutorial_series_volIII_table_image_1.png: Confound it, let's at least get the image right. |
| 14:50.04 | CIA-73 | BRL-CAD: 03starseeker * r38627 10/brlcad/trunk/doc/docbook/books/en/ (23 files in 2 dirs): Get VolIII image sizes closer to reasonable. May still need to up the dpi - pdf will be the real test - but getting closer. |
| 15:01.37 | starseeker | majority of the VolIII images appear to actually be Word drawings + pics in final form |
| 15:01.45 | CIA-73 | BRL-CAD: 03indianlarry * r38628 10/brlcad/trunk/src/libbu/units.c: Added NULL conv_table element to end of unit_lists[]. Looping constructs were setup to expect this as final value and would cause a segmentation fault when unit not found. |
| 15:01.48 | starseeker | blegh |
| 15:31.50 | CIA-73 | BRL-CAD: 03starseeker * r38629 10/brlcad/trunk/doc/docbook/ (19 files in 4 dirs): More Docbook image tweakage |
| 15:47.37 | brlcad | I have several other files for vol3 in other formats |
| 15:51.55 | brlcad | there, bunch more uploaded |
| 15:52.28 | brlcad | also have more ppt and doc files if there is a specific image you need, probably pulled from doc to doc |
| 16:00.34 | CIA-73 | BRL-CAD: 03brlcad * r38630 10/brlcad/trunk/TODO: the rt* tools need some documentation refactoring love. |
| 16:19.42 | d-lo | ``Erik: Hrm, just checked the CMakeLists.txt file in rt^3/src/other/ and OIS SHouldn't be building... commented out. |
| 16:27.20 | CIA-73 | BRL-CAD: 03brlcad * r38631 10/brlcad/trunk/doc/deprecation.txt: deprecate ALL of the plot tools from using pl in the command name, instead anticipatingly using plot3. this is per patch 2989897 from tom browder to avoid confusion and conflict with perl files. |
| 16:35.37 | CIA-73 | BRL-CAD: 03brlcad * r38632 10/brlcad/trunk/doc/deprecation.txt: mention .pl suffix changing to .plot3 as a general catch-all |
| 17:13.25 | CIA-73 | BRL-CAD: 03brlcad * r38633 10/brlcad/trunk/ (include/raytrace.h src/librt/comb/db_comb.c): make db_mkgift_tree() take a long instead of an int, but make db_mkbool_tree() take a size_t. the prior can't take a size_t as it uses looping that waits until it hits negative to terminate. |
| 17:18.12 | CIA-73 | BRL-CAD: 03brlcad * r38634 10/brlcad/trunk/src/librt/comb/db_comb.c: db_mkgift_tree() now takes a long, so cast from size_t |
| 17:18.29 | CIA-73 | BRL-CAD: 03brlcad * r38635 10/brlcad/trunk/src/librt/db5_io.c: tweak message. don't know for certain that it's a 32-bit. |
| 17:18.32 | CIA-73 | BRL-CAD: 03starseeker * r38636 10/brlcad/trunk/doc/docbook/lessons/en/ (4 files in 2 dirs): Fix lesson 6 image links. |
| 17:19.15 | CIA-73 | BRL-CAD: 03brlcad * r38637 10/brlcad/trunk/src/librt/db5_io.c: ws |
| 17:21.08 | CIA-73 | BRL-CAD: 03brlcad * r38638 10/brlcad/trunk/src/librt/db5_io.c: fix non-standard extension of declaring functions with file scope (msvc quellage) |
| 17:22.45 | CIA-73 | BRL-CAD: 03brlcad * r38639 10/brlcad/trunk/src/librt/db5_scan.c: cast to the appropriate function pointer type |
| 17:23.21 | CIA-73 | BRL-CAD: 03brlcad * r38640 10/brlcad/trunk/src/librt/db5_scan.c: consistency cleanup |
| 17:25.10 | CIA-73 | BRL-CAD: 03brlcad * r38641 10/brlcad/trunk/src/librt/db_alloc.c: quellage |
| 17:28.19 | CIA-73 | BRL-CAD: 03brlcad * r38642 10/brlcad/trunk/src/librt/db_alloc.c: style ws consistency cleanup |
| 17:30.28 | CIA-73 | BRL-CAD: 03brlcad * r38643 10/brlcad/trunk/src/librt/ (db_alloc.c db_inmem.c): more -1 casting quellage |
| 17:31.18 | CIA-73 | BRL-CAD: 03brlcad * r38644 10/brlcad/trunk/src/librt/db_inmem.c: minor ws cleanup |
| 17:34.03 | CIA-73 | BRL-CAD: 03brlcad * r38645 10/brlcad/trunk/src/librt/db_io.c: size_t casting on error result (begging for define) |
| 17:35.46 | CIA-73 | BRL-CAD: 03starseeker * r38646 10/brlcad/trunk/doc/docbook/lessons/en/ (6 files in 2 dirs): Add in some images for missing contents in lesson 16 - need to be redone, but that should be the last of the 'missing image' errors |
| 17:40.44 | CIA-73 | BRL-CAD: 03davidloman * r38647 10/rt^3/trunk/tests/GE/GeometryEngineTest.cxx: Stub in GeometryEngineTest. Precursor to Build system cleanup. |
| 17:42.52 | CIA-73 | BRL-CAD: 03davidloman * r38648 10/rt^3/trunk/cmake/ProjectPrinter.cmake: Extract common cmake project printing lines into a single include. Precursor to Build system cleanup. |
| 17:44.52 | ``Erik | asonofabitch, more files added? |
| 17:45.13 | starseeker | sorry |
| 17:46.11 | ``Erik | <-- almost tempted to write a make target or script to generate the manifest |
| 17:47.27 | CIA-73 | BRL-CAD: 03erikgreenwald * r38649 10/brlcad/trunk/src/librt/db_io.c: size_t instead of sizet. |
| 17:48.11 | brlcad | whoops |
| 17:48.28 | brlcad | lotta commits en route |
| 17:48.38 | ``Erik | and now a conflict, mwahahaha |
| 18:06.25 | *** join/#brlcad parigaudi (~quassel@pd95b7f5e.dip0.t-ipconnect.de) |
| 18:12.16 | CIA-73 | BRL-CAD: 03brlcad * r38650 10/brlcad/trunk/ (5 files in 3 dirs): |
| 18:12.16 | CIA-73 | BRL-CAD: substantial overhaul of offset management for mem_map objects, |
| 18:12.16 | CIA-73 | BRL-CAD: db_scan()/db5_sca() handlers and dbi's dbi_eof position (from scan). make them |
| 18:12.16 | CIA-73 | BRL-CAD: off_t's instead of size_t's as they do represent file and memory offsets, not |
| 18:12.16 | CIA-73 | BRL-CAD: sizes. it's also a signed type so the -1 marker we use all over the place won't |
| 18:12.16 | CIA-73 | BRL-CAD: get converted through to a potentially unsigned type. |
| 18:17.31 | CIA-73 | BRL-CAD: 03indianlarry * r38651 10/brlcad/trunk/src/rt/opt.c: |
| 18:17.31 | CIA-73 | BRL-CAD: Added 'model' keyword to the '-u' units option parsing. When the 'model' keyword |
| 18:17.31 | CIA-73 | BRL-CAD: is passed as the units option a 'model_units' flag is set signaling programs |
| 18:17.31 | CIA-73 | BRL-CAD: like 'rtarea' to produce results in the current model working units. |
| 18:18.13 | CIA-73 | BRL-CAD: 03indianlarry * r38652 10/brlcad/trunk/src/rt/viewarea.c: |
| 18:18.13 | CIA-73 | BRL-CAD: Added external reference to 'model_units' flag. Added warning when using current |
| 18:18.13 | CIA-73 | BRL-CAD: model units as result units. Added some limited logic to show a common larger |
| 18:18.13 | CIA-73 | BRL-CAD: unit in the parenthesized portion of the result: mm^2,cm^2, dm^2 -> m^2, m^2 -> |
| 18:18.13 | CIA-73 | BRL-CAD: km^2, in^2 -> ft^2, ft^2 -> yd^2 otherwise defaults to mm^2. |
| 18:20.06 | CIA-73 | BRL-CAD: 03indianlarry * r38653 10/brlcad/trunk/src/libged/rt.c: |
| 18:20.06 | CIA-73 | BRL-CAD: Added logic to ged_rt(...) to append "-u model" to 'rtarea' command when the |
| 18:20.06 | CIA-73 | BRL-CAD: user does not specify the output units explicitly. This means that 'rtarea' when |
| 18:20.06 | CIA-73 | BRL-CAD: run from within 'mged' will produce results in the current working model units |
| 18:20.06 | CIA-73 | BRL-CAD: unless overridden by the user. Note: For backward compatibility with existing |
| 18:20.06 | CIA-73 | BRL-CAD: scripts running 'rtarea' from the system command line(not from within mged) will |
| 18:20.07 | CIA-73 | BRL-CAD: still produce output in mm^2 unless otherwise specified with the '-u' option. |
| 18:24.10 | CIA-73 | BRL-CAD: 03brlcad * r38654 10/brlcad/trunk/src/librt/ (db_inmem.c db_io.c db_open.c): off_t cleanup |
| 18:26.16 | ``Erik | changing dbi_eof to off_t is causing issues, it's compared against (size_t)-1 in places |
| 18:26.34 | ``Erik | RT_DIR_PHONY_ADDR for example |
| 18:27.30 | brlcad | yeah, not unexpected, just odd I can't reproduce them -- may be off_t's match size_t's on my config |
| 18:27.58 | brlcad | directory pointer addresses is probably another off_t |
| 18:29.07 | ``Erik | um, my mac and leenewx builds are 64b, wonder if that's related |
| 18:29.57 | CIA-73 | BRL-CAD: 03davidloman * r38655 10/rt^3/trunk/ (29 files in 22 dirs): Cleanup of CMake build system. Standardized include path and library logic. Fixed multiple issues with QT4 configuration and use. |
| 18:30.26 | CIA-73 | BRL-CAD: 03brlcad * r38656 10/brlcad/trunk/include/raytrace.h: d_addr's (i.e., d_un.file_offset's) in a directory structure are file offsets so use off_t instead of size_t. this probably has a cascade of fallout that will need cleanup. |
| 18:39.46 | CIA-73 | BRL-CAD: 03bob1961 * r38657 10/brlcad/trunk/ (6 files in 4 dirs): Added a new bot_split command (the old one didn't do anything). The new one splits out disconnected pieces within a bot into separate bots. |
| 18:41.11 | CIA-73 | BRL-CAD: 03brlcad * r38658 10/brlcad/trunk/ (include/raytrace.h src/librt/db_io.c): db_write()/db_get()/db_put() also use offsets. make it so. off_t. |
| 18:43.26 | CIA-73 | BRL-CAD: 03davidloman * r38659 10/rt^3/trunk/src/ (CMakeLists.txt other/CMakeLists.txt): Restore coreinterface compilation. Accidentally commented it out. |
| 18:45.30 | CIA-73 | BRL-CAD: 03erikgreenwald * r38660 10/brlcad/trunk/src/librt/ (db5_scan.c db_io.c memalloc.c): off_t/size_t casting |
| 18:47.02 | ``Erik | wonders if brlcad has turned off strict compile flags? O.o |
| 18:47.34 | brlcad | nah, they're all listed |
| 18:47.42 | CIA-73 | BRL-CAD: 03brlcad * r38661 10/brlcad/trunk/src/librt/db_io.c: fread() returns a count less than what we requested, not -1. print the error regardless if it's less. |
| 18:48.53 | brlcad | casting RT_DIR_PHONY_ADDR to size_t is probably an indication of faulty logic, something that should be testing for a diff value or that needs to be off_t itself |
| 18:50.23 | ``Erik | probably, I'm scrambling to get commits thrown before conflicts happen heh |
| 18:50.38 | ``Erik | (of course, now I'm getting asc2g failures on one of my test platforms) |
| 18:59.36 | CIA-73 | BRL-CAD: 03brlcad * r38662 10/brlcad/trunk/ (include/raytrace.h src/librt/memalloc.c): rt_memget()'s place param looks to be an offset |
| 19:12.54 | *** join/#brlcad R0b0t1 (~Enigma@unaffiliated/r0b0t1) |
| 19:15.27 | d-lo | interesting |
| 19:15.56 | d-lo | so if i config brlcad to install to my home dir (/home/dloman) and then ' |
| 19:16.00 | d-lo | 'make install' |
| 19:16.29 | d-lo | running brlcad-config --includedir still outputs '/usr/brlcad/include' |
| 19:30.00 | CIA-73 | BRL-CAD: 03starseeker * r38663 10/brlcad/trunk/doc/docbook/books/en/ (9 files in 2 dirs): Add some of the Volume I content to serve as a 'cover page' for the help browser. |
| 19:33.27 | CIA-73 | BRL-CAD: 03starseeker * r38664 10/brlcad/trunk/src/tclscripts/archer/Archer.tcl: Switch html help viewer to using hv3 megawidget, point it to Vol1 for intro page, conditionalize drawing html windows on presence of the Vol1 html file. |
| 19:34.51 | CIA-73 | BRL-CAD: 03starseeker * r38665 10/brlcad/trunk/doc/html/toc.html: Add vol1 to the toc list. |
| 19:38.57 | CIA-73 | BRL-CAD: 03davidloman * r38666 10/rt^3/trunk/cmake/FindBRLCAD.cmake: Fixed brlcad lib finding logic. FOUND and NOTFOUND were reversed. |
| 19:41.57 | CIA-73 | BRL-CAD: 03davidloman * r38667 10/rt^3/trunk/CMakeLists.txt: Removed 'common install' paths from master CMakeLists.txt file. Now fully relies on PATH env variable. |
| 19:59.05 | CIA-73 | BRL-CAD: 03davidloman * r38668 10/rt^3/trunk/ (cmake/FindBRLCAD.cmake src/coreInterface/CMakeLists.txt): Cleaned up brlcad-config output by removing the newlines. Logic rolled down into coreinterface. |
| 20:09.59 | CIA-73 | BRL-CAD: 03davidloman * r38669 10/brlcad/trunk/ (18 files in 18 dirs): Found a whole slew of build byproducts not on the svn:ignore list. |
| 20:42.47 | starseeker | cannot escape the feeling the whole nirt/mged hookup is waaay more complicated than it needs to be... |
| 20:48.57 | brlcad | probably is, but how so? |
| 20:50.22 | brlcad | d-lo: sounds like a previous config |
| 20:51.05 | starseeker | just seems like a lot of stuff going on - would have thought it would be a simple call to shootray with the handlers set up, no need for pipe... |
| 20:51.15 | starseeker | I suppose that might be the script handling |
| 20:52.39 | brlcad | a lot of it is script handling so that it's intentionally going through nirt's application front-end |
| 20:52.55 | brlcad | since nirt isn't _just_ a plain wrapper on rt_shootray() |
| 20:54.32 | brlcad | shot setup, various options to modify behavior and reporting, the script interface, etc |
| 20:56.04 | starseeker | methinks nirt's frontend needs to turn into some libanalyze functions |
| 20:56.48 | brlcad | and it's formatted output into libbu functions |
| 20:57.38 | starseeker | keeps having to fight the urge to dig into that... must get Archer alpha ready Soon |
| 20:59.47 | brlcad | July 1st is a good alpha ship date goal |
| 20:59.56 | brlcad | potentially three releases before then |
| 21:00.15 | starseeker | nods - what are your target "must have this in there" features? |
| 21:01.15 | brlcad | you've already hit up several of them |
| 21:02.08 | starseeker | see the two major remaining ones as comb editing in the right panel and the remaining primitive editing functionality (pipe, sketch, bot) |
| 21:02.21 | brlcad | in addition to mged command parity.. |
| 21:03.03 | starseeker | oh, and figure out why we can't raytrace in-frame anymore... |
| 21:05.48 | brlcad | integrated help, hierarchical browsing, object list view, edit panels for all objects, and migration of most mged "tools" (overlaps, adc, snap-to-grid, rendering, query, and patterns) |
| 21:09.02 | brlcad | more pedantic: checking out the input bindings for consistency/availability (shift grips and view hot keys at a minimum); making sure there's a graphical means for editing attributes, prims, and combs; and faceplate |
| 21:09.19 | brlcad | that's probably all that can be achieved, if even that |
| 21:09.44 | starseeker | nods |
| 21:11.16 | starseeker | I'm eyeing tackling the pattern tool - it's either a quick and dirty snarf of patterns.tcl and pattern_gui.tcl, or a scrubdown of clone.c and porting patterns.tcl into the clone c code |
| 21:15.13 | starseeker | must consult the modelers |
| 21:17.03 | brlcad | I'd say scrub down clone.c and port patterns |
| 21:17.15 | brlcad | if it's worth doing, it's worth doing well |
| 21:17.25 | brlcad | and cloning/patterns is a CAD fundamental |
| 21:17.51 | starseeker | nods - yeah, guess that makes sense. |
| 21:18.22 | brlcad | should be integrated into the gui and available on the fly, so you could band select some objects and replicate them in a pattern with just a couple clicks and a drag |
| 21:18.36 | starseeker | I do need to ask 'em about that GUI though - if they would prefer to work exclusively with clone on the command line (maybe with better illustrative docs) that might be a better place to put effort than a GUI redo |
| 21:18.41 | starseeker | nods |
| 21:19.11 | brlcad | the command line should not be required for base functionality |
| 21:19.14 | starseeker | integrating anything like that into our display manager seems to be a real trick though |
| 21:19.29 | brlcad | it should be required to get more advanced options, more fine-grained control |
| 21:19.51 | brlcad | but they shouldn't have to touch the command line if we do things right |
| 21:20.19 | starseeker | would dearly love to have graphical control of thinks like keypoint in-display, but is not even sure how to get started building that into the current architecture |
| 21:20.23 | brlcad | so while the gui doesn't need to have ALL of the options that clone has, there should be a way to replicate in various pattern shapes via the GUI (deep or shallow) |
| 21:21.48 | starseeker | (the nirt mockup has an in-display illustration of the ray origin point direction - I had intended to ask Bob if that was possible) |
| 21:28.04 | ``Erik | yeesh http://www.youtube.com/watch?v=o4MwTvtyrUQ&feature=player_embedded |
| 21:30.38 | *** join/#brlcad akafubu (~akafubu@unaffiliated/akafubu) |
| 21:31.21 | brlcad | <PROTECTED> |
| 21:35.02 | ``Erik | something with all this off_t noise seems to have broken something somewhere O.o asc2g bombs on BU_ASSERT_SIZE_T (when running asc2g) on 32b builds (but not 64) |
| 21:35.16 | ``Erik | starts a binary svn revision search |
| 21:35.23 | brlcad | k, I fix |
| 21:35.36 | ``Erik | length is being passed in as 0 at some point |
| 21:37.44 | *** join/#brlcad akafubu (~akafubu@unaffiliated/akafubu) |
| 21:41.28 | brlcad | length&7 or just length? |
| 21:41.34 | brlcad | which assert? |
| 21:43.01 | *** join/#brlcad talcite (~matthew@bas4-toronto21-2925505118.dsl.bell.ca) |
| 21:43.31 | talcite | brlcad: ping? |
| 21:43.36 | brlcad | pong? |
| 21:44.16 | talcite | brlcad: did we end up doing the upstream takeover of the bundled external projects? |
| 21:44.34 | talcite | a senior fedora dev is asking for a status update |
| 21:44.44 | brlcad | we have tkhtml taken over, lot of work to prepare an updated release |
| 21:45.53 | brlcad | no response from utah so we're good to go forward with URT as well, next step being to decide on project name to establish a new home |
| 21:46.35 | talcite | brlcad: ok. I'll report that back to him. Anything I can do to help speed things up? |
| 21:47.30 | brlcad | such as? |
| 21:48.19 | brlcad | you could do the fossil setup for tkhtml, clean up the files, update the docs, import the trackers, ... there's a long laundry list |
| 21:49.04 | talcite | brlcad: ok. Is this written down anywhere? I can try to tackle them one at a time |
| 21:50.16 | brlcad | someone probably could write up a checklist on the wiki, i'll look into that |
| 21:50.51 | ``Erik | neat: http://i.imgur.com/flMMU.gif (animated gif of the iceland volcano cloud) |
| 21:51.48 | talcite | thanks. I'll update bugzilla in the mean time |
| 21:58.14 | brlcad | ``Erik: what was the full assert fail line? |
| 21:58.28 | brlcad | should have a file , line in it |
| 22:00.56 | ``Erik | BU_ASSERT_SIZE_T(length>=8) failed, lhs=34359738368, rhs=2835355298232, file z¸PÕ, line 671672320 |
| 22:01.09 | ``Erik | mah po' stack |
| 22:01.25 | ``Erik | db5_io.c ~665ish, based on printf debugging... |
| 22:01.35 | brlcad | wow |
| 22:01.57 | ``Erik | 38655 works, 38658 fails, still honing in |
| 22:02.57 | CIA-73 | BRL-CAD: 03brlcad * r38670 10/brlcad/trunk/ (doc/deprecation.txt include/bu.h): genptr_t/GENPTR_NULL are hereby deprecated. their use is no longer relevant with ansi compliance as a requirement. |
| 22:03.45 | ``Erik | but I'm about to head home *shrug* I can look into it more tomorrow |
| 22:04.59 | ``Erik | 38656 fails |
| 22:05.39 | ``Erik | (damnit, now I wanna stay and kill it) |
| 22:06.07 | brlcad | hrmm |
| 22:06.07 | ``Erik | changing file_offset to off_t ... huh |
| 22:06.10 | ``Erik | that's all that patch is |
| 22:06.25 | brlcad | yep: http://brlcad.svn.sourceforge.net/viewvc/brlcad/brlcad/trunk/include/raytrace.h?r1=38656&r2=38655&pathrev=38656 |
| 22:06.40 | brlcad | and RT_DIR_PHONY_ADDR getting cast to off_t |
| 22:07.39 | ``Erik | hm |
| 22:07.52 | ``Erik | probably a test to see if file_offset < 0 somewhere |
| 22:09.18 | brlcad | that's kinda scary that it worked with size_t as that would have just been a super large offset value |
| 22:10.02 | ``Erik | aHHH |
| 22:10.05 | ``Erik | get this |
| 22:10.14 | ``Erik | off_t is __int64_t, size_t is __uint32_t |
| 22:10.34 | brlcad | sounds reasonable |
| 22:10.44 | ``Erik | so -1 to one is not -1 to the other |
| 22:11.01 | brlcad | that's why it's scary that size_t worked at all |
| 22:11.06 | ``Erik | ayup |
| 22:11.17 | brlcad | unsigned it would hhave just been a big num MAX_UINT-1 |
| 22:11.26 | brlcad | or MAX_UINT or whatever |
| 22:11.46 | brlcad | looks like db5_write_free() is where it's coming from |
| 22:12.17 | brlcad | which is db5_realloc() |
| 22:12.54 | ``Erik | yeah, I can give you a bt if you want... |
| 22:12.55 | brlcad | ahh, maybe that's related |
| 22:13.00 | brlcad | sure |
| 22:13.15 | brlcad | I think I found it |
| 22:13.24 | ``Erik | ah, then I'll let ya roll |
| 22:13.30 | ``Erik | I've been here too long, I have cats to annoy O.o |
| 22:13.33 | ``Erik | wanders off |
| 22:13.35 | brlcad | db5_alloc.c:155 |
| 22:13.52 | brlcad | d_addr getting set to a size_t |
| 22:16.16 | talcite | brlcad: what was the verdict about STEP? are we doing a takeover as well? |
| 22:18.03 | brlcad | that basically already happened before this all started -- but project infrastructure still isn't in place (a home) |
| 22:22.57 | talcite | oh ok. I'll add that to the list of things to do |
| 22:26.27 | CIA-73 | BRL-CAD: 03brlcad * r38671 10/brlcad/trunk/src/librt/db_io.c: returns more than -1, update comment. |
| 22:26.52 | CIA-73 | BRL-CAD: 03brlcad * r38672 10/brlcad/trunk/src/librt/db_scan.c: cast to off_t for clarity on the type |
| 22:28.43 | CIA-73 | BRL-CAD: 03brlcad * r38673 10/brlcad/trunk/src/librt/ (db5_alloc.c db_alloc.c): BAD juju. no donut for using size_t vars to store negative values. causing a whole world of db i/o hurt since -1 cast to unsigned cast to off_t (bigger signed) certainly won't match correctly. |
| 22:34.24 | talcite | brlcad: do we have existing options for project infrastructure, or do we need to find one? Fedora offers trackers and version control for projects that are released under compatible software licenses (which isn't a problem for us) |
| 22:35.16 | talcite | it would be a big head start, and we wouldn't have to worry about maintenance |
| 22:35.45 | brlcad | for ease of management, I'm partial to sf.net and google code with preference for sf.net most of the time unless there's a strong motivating factor for something else |
| 22:36.20 | talcite | brlcad: ahh ok. I can set us up with something there if you want |
| 22:37.17 | brlcad | the issue then still comes down to finalizing on a short and long name for the project, as they are permanent |
| 22:37.48 | brlcad | and UtahRasterToolkit doesn't exactly roll off the tounge |
| 22:38.04 | talcite | heh |
| 22:38.17 | brlcad | aside from having no utah involvement any more |
| 22:38.41 | brlcad | librle is at the core of the kit, so maybe something based on that |
| 22:39.25 | brlcad | step is a bit easier |
| 22:40.12 | brlcad | I have a good name in mind there but want to let it simmer and check on conflicts |
| 22:40.37 | talcite | the good name is for STEP, or URT? |
| 22:40.41 | brlcad | step |
| 22:40.56 | brlcad | need brainstorming on urt |
| 22:41.08 | brlcad | got to run, back in a few |
| 22:41.16 | talcite | is this something that's suitable for the mailing list? |
| 22:41.18 | talcite | k |
| 22:41.36 | brlcad | could, but brainstorming here would suffice too |
| 22:41.55 | brlcad | I just haven't put too much critical thought into it lately |
| 22:41.59 | brlcad | urt that is |
