| 00:14.00 | yukonbob | hates seeing this happen |
| 00:23.04 | starseeker | needs to once in a while - just how it works :/ |
| 00:57.22 | *** join/#brlcad madant (n=madant@117.196.128.23) |
| 01:56.36 | yukonbob | starseeker: agreed, but still not pleasant ;) |
| 01:57.08 | yukonbob | brlcad: congrats and drawing the line that's been a long time coming... |
| 03:43.32 | brlcad | that doesn't merit congratulations, but it has been a long time coming |
| 03:45.48 | brlcad | I expect the best in people |
| 03:46.26 | yukonbob | well... congrats is a weird word for it, but a decision had to made, you made it, so "congrats". It's not easy. |
| 03:46.39 | brlcad | it took a while to convince that he just *won't* be helpful in any way even if it was "nearly" certain when he first showed up |
| 03:47.21 | yukonbob | yup... it's like he tried his hardest to avoid learning and being helpful. But whatever... water under the bridge. |
| 03:47.40 | yukonbob | how's the release coming, or is done? |
| 03:47.41 | brlcad | just took a few more decimal points of precision, 99%, 99.9%, 99.99%, etc... to be sure he wouldn't |
| 03:47.53 | yukonbob | heh |
| 03:47.53 | brlcad | coming along, what I'm doing right now |
| 03:47.59 | yukonbob | nods |
| 03:48.00 | yukonbob | nice |
| 03:48.10 | brlcad | starseeker did most of the hard work |
| 03:48.38 | yukonbob | I had some auto* issues when I went for a complete in-tree build -- no time to pursue before this release though... |
| 03:49.02 | yukonbob | will bring them up in the next few days hopefully, and start working on models and landscapes again :) |
| 03:49.09 | brlcad | k |
| 03:51.22 | brlcad | now that I know what all it took to get the mediawiki extension to work, I'm wanting to make one for drupal too.. too much prioritized against it though |
| 03:51.32 | yukonbob | plays with the Tcl C API atm... is fun |
| 03:51.35 | brlcad | web site needs a new look |
| 03:52.03 | brlcad | Tcl C api is some of the best aspects of Tcl imho |
| 03:54.17 | brlcad | starseeker: any idea how tkhtml3's Makefile.in was clobbered during your merge? makes me dubious of the merge state |
| 03:54.18 | yukonbob | I've not played 'seriously' in C in a while, but I'm digging this after so many years of high level (esp. Tcl)... swimming up/down the 'stack' in my head between C and Tcl, between proc and functions, C vars and Tcl vars is fun and powerful. |
| 03:56.33 | yukonbob | it's neat to take a look at a task and turn it inside-out/outside-in, thinking "is C-driven Tcl, or Tcl-driven C?" |
| 04:32.51 | starseeker | brlcad: no clue how the Makefile.in was clobbered |
| 04:33.07 | starseeker | might be worth a manual diff of the trunk and stable directories |
| 04:36.40 | starseeker | brlcad: Oh, wait |
| 04:37.26 | starseeker | there was some sort of conflict with that file - I THOUGHT I copied in the newest version, but maybe not |
| 04:38.13 | starseeker | ok, I probably messed it up by not checking the conflict more carefully - my bad |
| 04:38.18 | starseeker | sorry :-( |
| 05:45.10 | CIA-40 | BRL-CAD: 03starseeker * r33711 10/brlcad/trunk/regress/mged/ (33 files): |
| 05:45.10 | CIA-40 | BRL-CAD: OK, rework this one more time. Make the mged logic .mged files, and put tcl |
| 05:45.10 | CIA-40 | BRL-CAD: specific stuff all in regression_resources.tcl. Have mged.tcl assemble the |
| 05:45.10 | CIA-40 | BRL-CAD: .mged files and run the test - the way the logic is defined it's easy to run |
| 05:45.12 | CIA-40 | BRL-CAD: individual tests as well. |
| 05:48.27 | *** topic/#brlcad by brlcad -> BRL-CAD Open Source Solid Modeling || http://brlcad.org || http://sf.net/projects/brlcad || GSoC 2008 Highlight: new prototype gui, check it out! || Source Release 7.14.2 is posted (20080207) |
| 06:59.23 | *** join/#brlcad bjorkintosh (n=bjork@ip72-204-40-138.fv.ks.cox.net) |
| 07:31.30 | *** join/#brlcad madant (n=madant@117.196.144.139) |
| 07:37.27 | starseeker | puzzles out how to set LD_LIBRARY_PATH from within tcl |
| 07:46.29 | brlcad | set env(LD_LI...) |
| 07:46.49 | starseeker | thanks :-) |
| 07:47.01 | starseeker | I take it the update to STABLE went OK? |
| 07:47.09 | starseeker | saw topic change |
| 07:47.49 | brlcad | went ahead with the existing tag since it passed testing, patches weren't major |
| 07:47.54 | starseeker | ah |
| 07:48.13 | brlcad | started the merge, but found a handful of other things to manually review |
| 07:48.15 | starseeker | 2008 date and all? ;-) |
| 07:48.35 | brlcad | yep, minor issue |
| 07:48.42 | starseeker | hmm. how many issues cropped up? |
| 07:49.16 | brlcad | about 200 files to review |
| 07:49.32 | brlcad | categorically probably about 50 |
| 07:49.51 | starseeker | jeez |
| 07:50.01 | starseeker | must have messed up big time |
| 07:50.30 | starseeker | blast it, I thought that last merge was pretty clean |
| 07:50.48 | brlcad | well I haven't done the review yet |
| 07:51.51 | starseeker | doesn't really follow how that could happen - isn't subversion supposed to handle merging of this sort correctly? |
| 07:53.14 | brlcad | it handles merging between revision ranges on subtrees |
| 07:53.24 | brlcad | so our revision numbers may have been different |
| 07:53.34 | starseeker | hmm |
| 07:53.46 | brlcad | which by the way, you should include in the comment when you commit any merge |
| 07:54.04 | starseeker | ah, right. |
| 07:56.20 | CIA-40 | BRL-CAD: 03starseeker * r33712 10/brlcad/trunk/regress/mged/ (make_bb.tcl mged.tcl): Whoops, left over tcl test. |
| 07:57.13 | starseeker | pulls a stable tree |
| 07:57.31 | starseeker | I'll try to take a look tomorrow - 3am is reaching my limits :-( |
| 08:20.20 | *** join/#brlcad Ralith (n=ralith@216.162.199.202) |
| 10:13.49 | *** join/#brlcad mafm (n=mafm@65.Red-81-34-125.dynamicIP.rima-tde.net) |
| 10:19.42 | *** join/#brlcad elite01 (n=omg@unaffiliated/elite01) |
| 10:33.28 | *** join/#brlcad madant (n=madant@117.196.139.89) |
| 10:41.37 | *** join/#brlcad mafm_ (n=mafm@65.Red-81-34-125.dynamicIP.rima-tde.net) |
| 10:41.56 | *** join/#brlcad _sushi_ (n=_sushi_@77-58-234-172.dclient.hispeed.ch) |
| 11:25.09 | *** join/#brlcad Ralith (n=ralith@216.162.199.202) |
| 12:18.48 | *** join/#brlcad madant (n=madant@117.196.129.23) |
| 15:36.21 | *** join/#brlcad _sushi__ (n=_sushi_@77-58-245-64.dclient.hispeed.ch) |
| 15:54.42 | *** join/#brlcad madant (n=madant@117.196.128.158) |
| 16:10.01 | *** join/#brlcad samrose (n=samrose@c-24-11-214-181.hsd1.mi.comcast.net) |
| 16:58.15 | *** join/#brlcad madant (n=madant@117.196.141.197) |
| 17:22.38 | starseeker | me growls at sourceforge |
| 17:29.38 | brlcad | starseeker, been looking into it and looks like the merge was clean other than that one Makefile.in |
| 17:29.50 | starseeker | brlcad: phew |
| 17:29.53 | starseeker | thank you! |
| 17:30.12 | starseeker | was sweating it |
| 17:30.31 | starseeker | what were all the diffs about? |
| 17:35.59 | starseeker | starts plotting in a year or so to do a serious upgrade to his home box... |
| 17:36.28 | CIA-40 | BRL-CAD: 03brlcad * r33713 10/brlcad/branches/STABLE/ (40 files in 3 dirs): sync with trunk from 33207 through 33711 |
| 17:39.24 | brlcad | a handful of issues, reapplying merges that had already occurred showing up as a diff, two conflicts (easy to resolve), few diffs that were $Revision:$ identifiers, bunch of ignored files (the generated html file) |
| 17:39.40 | brlcad | *files |
| 17:56.13 | *** join/#brlcad madant (n=madant@117.196.129.21) |
| 18:56.25 | *** join/#brlcad micges (n=michu@146.8.28.78.cable.dyn.petrus.com.pl) |
| 19:01.56 | *** part/#brlcad micges (n=michu@146.8.28.78.cable.dyn.petrus.com.pl) |
| 19:03.03 | *** join/#brlcad micges (n=michu@146.8.28.78.cable.dyn.petrus.com.pl) |
| 19:09.46 | *** part/#brlcad micges (n=michu@146.8.28.78.cable.dyn.petrus.com.pl) |
| 19:18.08 | *** part/#brlcad bjorkintosh (n=bjork@ip72-204-40-138.fv.ks.cox.net) |
| 19:38.36 | CIA-40 | BRL-CAD: 03starseeker * r33714 10/brlcad/trunk/regress/mged/mged.tcl: Add a few notes on potentially useful tools for the problem of finding and running mged |
| 20:35.07 | *** join/#brlcad _sushi_ (n=_sushi_@77-58-239-215.dclient.hispeed.ch) |
| 20:50.30 | starseeker | brlcad: remind me - is there a reason we can't simply always point the regression scripts to ../src/mged/mged as a relative path? |
| 20:57.26 | starseeker | is it that the script deduced that from its own location rather than the current working directory? |
| 21:28.41 | brlcad | starseeker: hm, I don't remember the exact reason, maybe in the logs, but by having them identify the script location they will run from anywhere |
| 21:29.05 | brlcad | including in the $builddir for out-of-dir builds, but also in random 'other' directories like /tmp |
| 21:29.55 | starseeker | brlcad in /tmp you would have to specify both the script location and the mged bin location - neither can be reliably deduced |
| 21:30.26 | starseeker | hmm. |
| 21:31.07 | brlcad | hm? |
| 21:31.17 | brlcad | it finds both |
| 21:31.44 | brlcad | it's still up to the individual scripts for files it generates |
| 21:31.45 | starseeker | how? If you're doing an out of source build, the script location and the bin location may be virtually arbitrary |
| 21:34.01 | brlcad | ah, it has since changed -- it used to be told where builddir was |
| 21:35.03 | brlcad | mind you, it doesn't care about 'bin' location, just builddir and srcdir |
| 21:35.17 | starseeker | nods |
| 21:36.24 | brlcad | ah, iirc another reason was 'make' portability, where a given make may not recurse the current working directory |
| 21:37.13 | starseeker | hmm |
| 21:38.10 | brlcad | not that it really matters for the regress scripts much either way .. just need to be able to run them to verify a build under 'some' set of conditions |
| 21:38.16 | brlcad | they don't need to be arbitrarily flexible |
| 21:39.38 | starseeker | true |
| 21:39.58 | starseeker | is probably overthinking, but I'd like to have some sort of "correct" solution |
| 21:50.57 | brlcad | which per the requirements of a simple, easy-to-grok, and easy to maintain testing environment, that leaves a *whole* lot of room for interpretation |
| 21:51.10 | brlcad | given the project's history, though, I would say "less is more" in this particular area |
| 21:51.50 | starseeker | sigh |
| 21:52.07 | starseeker | well, I'll carry my current train of thought to completion, and see what you think |
| 21:52.29 | starseeker | btw, how do I turn globbing on and off in the mged command line again? |
| 21:52.32 | brlcad | like I said before, sensitivity to change is the biggest evil |
| 21:52.38 | brlcad | set glob_compat_mode 0 |
| 21:52.43 | starseeker | ah, thanks |
| 22:29.32 | CIA-40 | BRL-CAD: 03starseeker * r33715 10/brlcad/trunk/regress/mged/ (18 files): |
| 22:29.32 | CIA-40 | BRL-CAD: One more time with regress/mged - mged.tcl can now be run successfully in an |
| 22:29.32 | CIA-40 | BRL-CAD: external build directory, but now requires it be passed the path of the top |
| 22:29.32 | CIA-40 | BRL-CAD: level source directory (same as mged.sh). This required taking the source |
| 22:29.33 | CIA-40 | BRL-CAD: statements out of the .mged files, so a run_test.tcl script has been added. In |
| 22:29.35 | CIA-40 | BRL-CAD: theory, this is close to the point where it can be integrated into the build |
| 22:29.37 | CIA-40 | BRL-CAD: system successfully, once some sort of diff logic is added. |
| 23:47.06 | *** join/#brlcad ``Erik_ (i=erik@c-76-111-12-116.hsd1.md.comcast.net) |