| --- Day changed Tue Mar 10 2009 | ||
| xp_prg | I am going to read them right now with the phd guy | 00:00 |
|---|---|---|
| xp_prg | I will tell him you gave me them | 00:00 |
| kanzure | meh, pull some bullshit up on google scholar about hox | 00:00 |
| gene_ | look up xenopus morpholino | 00:01 |
| gene_ | I can't see anything | 00:02 |
| gene_ | crashed out | 00:02 |
| xp_prg | kanzure what was that link you gave me about how to get dna into xenopus? | 00:17 |
| kanzure | what? | 00:35 |
| -!- gene_ [n=chatzill@wireless-128-62-91-173.public.utexas.edu] has quit [Read error: 110 (Connection timed out)] | 00:38 | |
| -!- wrldpc [n=worldpea@c-98-217-178-69.hsd1.ma.comcast.net] has joined #hplusroadmap | 00:41 | |
| -!- gene_ [n=chatzill@wireless-128-62-91-173.public.utexas.edu] has joined #hplusroadmap | 00:41 | |
| -!- any78568049 [n=someone@75-120-44-19.dyn.centurytel.net] has joined #hplusroadmap | 01:30 | |
| -!- katsmeow-afk [n=someone@75-120-44-19.dyn.centurytel.net] has quit [Nick collision from services.] | 01:30 | |
| -!- any78568049 is now known as katsmeow | 01:30 | |
| -!- gene_ [n=chatzill@wireless-128-62-91-173.public.utexas.edu] has quit ["ChatZilla 0.9.84 [Firefox 3.0.7/2009021910]"] | 01:33 | |
| -!- wrldpc [n=worldpea@c-98-217-178-69.hsd1.ma.comcast.net] has quit [] | 01:39 | |
| -!- katsmeow is now known as katsmeow-afk | 02:01 | |
| -!- any32300344 [n=someone@75-120-44-19.dyn.centurytel.net] has joined #hplusroadmap | 04:15 | |
| -!- katsmeow-afk [n=someone@75-120-44-19.dyn.centurytel.net] has quit [Read error: 110 (Connection timed out)] | 04:33 | |
| -!- splicer__ [n=patrik@h72n3c1o261.bredband.skanova.com] has joined #hplusroadmap | 05:03 | |
| -!- splicer_ [n=patrik@h80n3c1o261.bredband.skanova.com] has quit [Read error: 110 (Connection timed out)] | 05:05 | |
| -!- elias` [n=me@unaffiliated/elias/x-342423] has joined #hplusroadmap | 06:26 | |
| kanzure | http://www.transforum.net/m.cgi?num=2919 | 09:07 |
| kanzure | http://www.spatialrobots.com/archives/category/reconfigurable-robots | 10:53 |
| kanzure | had to make pythonOCC/build as a dir, then chmod it appropriately, then I forgot to (re?)install python2.5 and python2.5-minimal, now trying scons again (it was last complaining about some Python.h headers that weren't existing) | 11:48 |
| -!- kanzure-- [i=bbishop@66.112.232.182] has quit [Read error: 104 (Connection reset by peer)] | 11:48 | |
| kanzure | wasn't that me? | 11:48 |
| kanzure | requires swig to be run with the -classic -nomodern -nomodernargs options. blah. | 11:48 |
| bkero | What's goin' on #hplusroadmap? | 11:48 |
| fenn | nuffn | 11:51 |
| -!- xp_prg [n=xp_prg3@c-24-130-14-113.hsd1.ca.comcast.net] has quit ["This computer has gone to sleep"] | 11:55 | |
| kanzure | fenn: line 284 of SConstruct, I think, is where swig flags go. | 12:09 |
| kanzure | but then they don't show up in the swig line that is generated when 'scons' is executed | 12:10 |
| kanzure | ah, the env line a few lines down is wrong. has a commented out thingy that is kinda important. | 12:10 |
| kanzure | hahah | 12:10 |
| fenn | oo nasty scons wont let me kill it with Ctrl-C | 12:10 |
| kanzure | "swig error : Unrecognized option -classic -nomoddern -nomodernargs" | 12:11 |
| kanzure | fuck, typo | 12:11 |
| kanzure | swig error : Unrecognized option -classic -nomodern -nomodernargs -python -modern -fcompact -c++ -DHAVE_LIMITS_H -DHAVE_CONFIG_H -DCSFDB -DOCC_CONVERT_SIGNALS -DLIN -DLININTEL -D_GNU_SOURCE=1 -outdir /home/bryan/projects/gears/pythonOCC/src/../build/OCC | 12:11 |
| kanzure | wtf, why is it "-nomodern" | 12:11 |
| kanzure | and then "-modern" | 12:11 |
| kanzure | it specifically asked for -nomodern | 12:11 |
| kanzure | SWIG_src_modular_linux_darwin/AIS_wrap.cc:3294:3: error: #error "This python version requires swig to be run with the '-nomodern' option" | 12:12 |
| fenn | SWIG_OPTS = ['-python','-modern' ... | 12:12 |
| fenn | in two places even | 12:13 |
| * kanzure trots off to class. | 12:13 | |
| fenn | enjoy | 12:13 |
| fenn | i dont suppose you installed pyplusplus | 12:14 |
| kanzure | no? | 12:15 |
| * kanzure is really gone now | 12:15 | |
| fenn | jeebus! pygccxml is 20MB? | 12:19 |
| fenn | (compressed) | 12:19 |
| -!- xp_prg [n=xp_prg3@99.2.31.217] has joined #hplusroadmap | 12:20 | |
| fenn | ah it's just excessive amounts of unit test data | 12:25 |
| xp_prg | https://www.noisebridge.net/wiki/User:Xp_prg | 12:49 |
| kanzure-_ | fenn: get anywhere? | 13:58 |
| fenn | commenting out missing modules.. | 14:12 |
| fenn | some of them certainly shouldn't be missing, like BRep | 14:14 |
| -!- wrldpc [n=worldpea@c-24-60-81-53.hsd1.ma.comcast.net] has joined #hplusroadmap | 14:18 | |
| wrldpc | Yo does anyone have the pdf for MIT's "A Robotics Primer" http://books.google.com/books?id=WWJPjgz-jgEC | 14:18 |
| fenn | nope | 14:21 |
| wrldpc | darn | 14:26 |
| -!- procto [n=boo@logarchy.org] has quit [Remote closed the connection] | 14:46 | |
| fenn | you'd think scons would remember what it had already compiled | 14:55 |
| fenn | instead of making me wait a half hour before throwing a new error | 14:55 |
| kanzure-_ | would it be simpler at this point to just configure our own automake script? | 15:00 |
| fenn | simpler for who? too many separate build systems will drive the devs crazy (and make it unmaintainable == more pain later) | 15:14 |
| fenn | i dont see any pattern in the modules that dont work | 15:15 |
| -!- wrldpc_ [n=worldpea@c-24-60-81-53.hsd1.ma.comcast.net] has joined #hplusroadmap | 15:35 | |
| -!- wrldpc_ [n=worldpea@c-24-60-81-53.hsd1.ma.comcast.net] has quit [] | 15:46 | |
| -!- wrldpc [n=worldpea@c-24-60-81-53.hsd1.ma.comcast.net] has quit [Read error: 113 (No route to host)] | 16:06 | |
| kanzure-_ | fenn: woah, it's as awesome as it seems | 16:11 |
| -!- any32300344 is now known as katsmeow | 16:11 | |
| kanzure-_ | it works well on windows. | 16:11 |
| kanzure-_ | although "from OCC import *" does not work, and it's annoying to try to remember the weird opencascade API, but other than that, it works. | 16:12 |
| kanzure-_ | :) | 16:12 |
| fenn | is there even an OCC module? i thought there were like 20 or 100 modules | 16:12 |
| ybit | http://fluid.media.mit.edu/ | 16:19 |
| -!- kanzure-- [i=bbishop@66.112.232.182] has joined #hplusroadmap | 16:23 | |
| -!- katsmeow is now known as katsmeow-afk | 16:37 | |
| fenn | http://blog.ponoko.com/2008/05/05/micro-origami-puts-miniature-paper-crane-folders-to-shame/ | 16:46 |
| fenn | apparently it uses magnetic material to actually do the folding motion | 16:47 |
| fenn | not really origami either from the looks of it | 16:48 |
| kanzure-- | http://heybryan.org/mediawiki/index.php/PythonOCC <- installation instructions for Windows whores. | 16:48 |
| kanzure-- | http://blog.ponoko.com/2008/05/05/micro-origami-puts-miniature-paper-crane-folders-to-shame/ | 16:54 |
| kanzure-- | blargh. copy+paste is broken on all systems. | 16:55 |
| kanzure-- | huh, they closed the structures with the use of capillary force | 16:59 |
| fenn | i thought it was magnets | 17:00 |
| kanzure-- | http://heybryan.org/~bbishop/docs/Voxels%20-%20volume-enclosing%20microstructures.pdf | 17:01 |
| kanzure-- | magnets are also involved | 17:01 |
| kanzure-- | hm, so I guess they claim that there is some collision imparted due to the capillary force of the liquid medium in the electroplating bath | 17:06 |
| kanzure-- | and that this closes the sides via the degrees of freedom offered by the hinges | 17:06 |
| fenn | collision? | 17:07 |
| kanzure-- | but what's to guarantee that they close? maybe a locking mechanism, I guess, but otherwise fluid flowing in the opposite direction should open it | 17:07 |
| kanzure-- | collision of water molecules with some velocity and some mass = some momentum. | 17:07 |
| kanzure-- | right? | 17:07 |
| fenn | eh, they're not that small | 17:07 |
| kanzure-- | or is there some other mechanism involved here? | 17:07 |
| kanzure-- | hm | 17:07 |
| fenn | its just surface tension | 17:07 |
| kanzure-- | "directional wetting" heh | 17:07 |
| fenn | i wonder if it has to be radially symmetric | 17:08 |
| fenn | they didnt complete the dodecahedron | 17:08 |
| kanzure-- | check out page 4 second to last paragraph | 17:08 |
| kanzure-- | so maybe I am not understanding how the folding is actually taking place her? | 17:09 |
| kanzure-- | *here | 17:09 |
| kanzure-- | suppose there was a two-link hinge thingy, and it had two flaps, and the surface tension of the water on the flap would force it to move over its degrees of freedom? | 17:10 |
| kanzure-- | does surface tension work when an object is submerged in a large volume of fluid? | 17:10 |
| fenn | yes | 17:10 |
| fenn | if it's just floating around then all the forces cancel out | 17:11 |
| kanzure-- | so theoretically, a perfectly hydrophobic sphere released instantaneously into the center of the ocean should .. ? | 17:11 |
| fenn | if it's asymmetric it'll fold up into a minimal energy configuration | 17:11 |
| fenn | likewise when up against a surface | 17:11 |
| kanzure-- | right. | 17:12 |
| kanzure-- | hrm. | 17:12 |
| kanzure-- | pg 5 "Velev and others that when two hydrophilic structures are immersed in a liquid there is a net attractive force between those structures that increases inversely with the distance between those objects [8, 9]." | 17:12 |
| kanzure-- | so it's interesting that evaporation is a mechanism of actuation as a way to get the objects into a minimal free energy configuration. but their fabrication technique sucks immensely. | 17:15 |
| fenn | what's wrong with the technique? | 17:16 |
| kanzure-- | http://www.memscap.com/memsrus/docs/polymumps.dr.v10.pdf | 17:21 |
| kanzure-- | mostly that the link is broken I guess :) | 17:21 |
| kanzure-- | but I thought they were using some conventional photolithography method involving a billion dollar fab that is completely inaccessible, | 17:22 |
| kanzure-- | but that was only because I assumed | 17:22 |
| kanzure-- | http://web.mit.edu/tinytech/Nanostructures/Spring2003/ABelcher/Science.pdf Ordering of quantum dots using genetically engineered viruses | 17:22 |
| -!- splicer__ [n=patrik@h72n3c1o261.bredband.skanova.com] has quit [Read error: 104 (Connection reset by peer)] | 17:23 | |
| fenn | cool its like aptamers | 17:25 |
| kanzure-- | http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.125.49&rep=rep1&type=pdf Capillary pumps for autonomous capillary systems (basically, obstacle courses) | 17:30 |
| kanzure-- | http://fluids.snu.ac.kr/publication/cbv.pdf How the capillary burst microvalve works | 17:32 |
| kanzure-- | while an interesting paper, note the snu.ac.kr .. | 17:32 |
| fenn | heh you could make some kind of cellular automata out of those capillary pumps | 17:33 |
| fenn | if they had movable elements | 17:34 |
| fenn | hinges that would get stuck one way or the other | 17:34 |
| kanzure-- | let's just make it easier on me: http://scholar.google.com/scholar?hl=en&lr=&client=firefox-a&q=site%3Asnu.ac.kr&btnG=Search | 17:37 |
| kanzure-- | http://fluids.snu.ac.kr/ | 17:39 |
| kanzure-- | revised search: http://scholar.google.com/scholar?q=site%3Asnu.ac.kr+seoul&btnG=Search&hl=en&lr=&client=firefox-a | 17:40 |
| kanzure-- | http://mims.snu.ac.kr/introduction/ | 17:44 |
| kanzure-- | they have web-to-micro-cnc? why doesn't ut have that? :( | 17:44 |
| fenn | i gotta say these are quite pretty http://farm4.static.flickr.com/3195/2900951549_1a079b1550.jpg | 17:54 |
| kanzure | fenn, is that their web micromachining service? | 18:18 |
| fenn | that's reprap in metalab | 18:19 |
| kanzure | metalab? | 18:20 |
| fenn | http://metalab.at/wiki/RepRap | 18:21 |
| fenn | a hacker space in vienna | 18:21 |
| kanzure | so, I've discovered that the air conditioning control panel ("thermostat", if you can call it that) has been off this whole time | 18:21 |
| kanzure | I have now rediscovered AC | 18:21 |
| kanzure | and it's lovely. | 18:22 |
| kanzure | I wonder what a wrapper to opencascade might look/feel like. the API is way too hairy/messy, IMHO. | 18:34 |
| fenn | a "pythonic" wrapper? | 18:37 |
| kanzure | I don't think that answers my question :) | 18:37 |
| fenn | it wasn't an answer really | 18:37 |
| kanzure | but yes. | 18:37 |
| kanzure | I wonder if this is worth installing everything through wine. | 18:38 |
| fenn | i guess thomas has some ideas | 18:38 |
| fenn | http://www.pythonocc.org/GSoc2009/ | 18:38 |
| fenn | there isnt much content though | 18:39 |
| kanzure | hah, I should get the Automated Design Lab to "sponsor" this to make it look like academic research (which I guess it could be) | 18:40 |
| fenn | it should be | 18:40 |
| kanzure | "Testing whether a parameter is trimmed or not is certainly not a trivial task, and the above works well, even the code is even fairly readable. The big issue though is finding which classes / methods to call. Figuring out which of the thousands of available classes to use is the hard part. Though when condensed to a class method it is a perfectly encapsulated and relevant piece of API knowledge." | 18:40 |
| fenn | with so many classes i think it'll always be difficult just to figure out what's there | 18:41 |
| fenn | would be neat to have a tool that you could use to see what methods accept your object | 18:42 |
| fenn | so like, i have a nurbs surface, what can i do with it | 18:42 |
| fenn | and you get a list of operators | 18:42 |
| kanzure | that would be an easy hack to the python system actually, | 18:42 |
| fenn | yes | 18:42 |
| kanzure | I was once talking with so-and-so about function signature searching | 18:42 |
| kanzure | who started python? G* something | 18:42 |
| fenn | guido | 18:43 |
| kanzure | yes, so guido slapped up some python code for functional signature searching. | 18:43 |
| kanzure | one moment. | 18:43 |
| kanzure | http://svn.python.org/view/sandbox/trunk/overload/overloading.py?rev=43727&view=auto | 18:43 |
| kanzure | (that's for function overloading, of course) | 18:44 |
| kanzure | so, I thought ducktyping wouldn't easily allow this? | 18:44 |
| fenn | maybe, but it would be slow | 18:44 |
| fenn | you'd want to build some kind of lookup table first | 18:44 |
| kanzure | that's not too painful. | 18:44 |
| kanzure | what I sent to guido: " | 18:45 |
| kanzure | what I sent to guido: " | 18:45 |
| kanzure | I was wondering if you had any thoughts on what I guess I'd call | 18:45 |
| kanzure | "significantly more dynamic overloading", and this is a result of me | 18:45 |
| kanzure | yak shaving for a very long time. To make it short, I'd like to *not* | 18:45 |
| kanzure | have to import very large packages that might implement hundreds or | 18:45 |
| kanzure | thousands of different variations on a single function. " | 18:45 |
| kanzure | gah. I even specifically deleted the newlines from that paste. | 18:45 |
| fenn | will have to play around with pythonocc before i really understand what OCC can do | 18:47 |
| fenn | bbiab getting my tools from shop for excursion to san antonio tomorrow (looking at steve langford's mess) | 18:47 |
| kanzure | heekscad is based off of occ. | 18:49 |
| kanzure | (I thought?) | 18:49 |
| kanzure | http://code.google.com/p/heekscad/ | 18:49 |
| kanzure | The solid modelling is provided by Open CASCADE. http://www.opencascade.org The graphical user interface is made using wxWidgets. http://www.wxwidgets.org | 18:49 |
| kanzure | hm, both pythonocc and heekscad uses wxWidgets | 18:49 |
| kanzure | so a heekscad-like interface for pythonocc is possible- there's already a partial interface, but it sucks, i.e. only has a few viewing features, no grid, etc. | 18:50 |
| -!- wrldpc [n=worldpea@c-98-217-178-69.hsd1.ma.comcast.net] has joined #hplusroadmap | 18:56 | |
| kanzure | had to install python2.5-dev | 19:08 |
| kanzure | hm, I think thomas might be forgetting to link to python? | 19:10 |
| kanzure | ah, OCC_LIB = '/usr/local/lib' whereas I think the default installation path in debian is /usr/lib/ | 19:15 |
| kanzure | (line 116 of SConstruct) | 19:16 |
| kanzure | okay, fixed that particular problem I guess. | 19:18 |
| kanzure | --cache-force is my savior | 19:30 |
| kanzure | http://www.scons.org/doc/0.92/HTML/scons-user/x1674.html | 19:30 |
| kanzure | http://www.scons.org/doc/1.2.0.d20090223/HTML/scons-user/c4135.html | 19:31 |
| -!- wrldpc [n=worldpea@c-98-217-178-69.hsd1.ma.comcast.net] has quit [] | 19:36 | |
| kanzure | coalition of programmers against swig without caching. | 20:20 |
| -!- Netsplit verne.freenode.net <-> irc.freenode.net quits: kanzure--, elias` | 20:23 | |
| -!- Netsplit over, joins: elias`, kanzure-- | 20:23 | |
| kanzure | scons: done building targets. | 20:31 |
| kanzure | hm, running python setup.py install -> runs gcc and swig (again). blah. | 20:34 |
| kanzure | it's possible though that I just didn't look hard enough for the compiled python files from scons | 20:35 |
| -!- elias` [n=me@unaffiliated/elias/x-342423] has quit [Read error: 145 (Connection timed out)] | 20:53 | |
| fenn | wait a minute.. you got the scons build to work | 21:04 |
| fenn | ? | 21:04 |
| ybit | haven't read the log yet, but i don't see a problem with scons | 21:07 |
| ybit | maybe i should read the log though :) | 21:08 |
| fenn | the problem is it doesnt fucking compile | 21:08 |
| ybit | haha, damn. you guys have been working on it all day | 21:09 |
| fenn | multiple days | 21:09 |
| ybit | i see | 21:10 |
| ybit | i'll take a stab at it after installing gentoo on this machine | 21:12 |
| ybit | in the middle of compiling the kernel currently, so it shouldn't be too much longer | 21:12 |
| -!- xp_prg [n=xp_prg3@99.2.31.217] has quit ["This computer has gone to sleep"] | 21:12 | |
| kanzure | fenn: yes, I did | 21:27 |
| kanzure | fenn: thanks for taking care of Nathan there (wtf though?) | 21:27 |
| fenn | any special pointers? | 21:27 |
| fenn | i think he's trying too hard to be "fair" | 21:28 |
| kanzure | ever get the feeling that he might be a pussy? | 21:28 |
| fenn | naw :) | 21:29 |
| kanzure | http://heybryan.org/~bbishop/docs/scons_notes.txt <- what I did to make it work | 21:30 |
| fenn | ok how do i get scons to actually use the damn cache | 21:30 |
| kanzure | erm | 21:30 |
| kanzure | http://heybryan.org/~bbishop/docs/pythonocc/scons_notes.txt <- what I did to make it work | 21:30 |
| kanzure | also, use: scons --cache-force | 21:30 |
| kanzure | also, #scons recommends upgrading to scons 1.2 (debian's repo has 1.0.0), but I haven't tried that | 21:31 |
| fenn | ok i'm missing a lot more modules than those three | 21:31 |
| fenn | and it takes like 12 minutes per compile so i can see what the next one is | 21:31 |
| fenn | which is why this is taking all day | 21:32 |
| kanzure | oh? what errors are you getting? | 21:33 |
| kanzure | I only started taking notes after a while :/ | 21:33 |
| fenn | missing dependencies.i file usually | 21:33 |
| kanzure | also, finds of the day (only because I was in a lab and not able to easily upload properly): | 21:33 |
| kanzure | http://heybryan.org/~bbishop/docs/2009-03-10/ | 21:33 |
| kanzure | fenn: there should only be 3 or 4 of those | 21:33 |
| kanzure | which I list in the .txt, so go fix. | 21:33 |
| fenn | i guess after a while i just stopped looking at the error message and blindly commented them out | 21:34 |
| fenn | anyway I have: AlienImage BRep BRepFeat Image ShapeAnalysis ShapeBuild ShapeConstruct ShapeFix ShapeProcess ShapeProcessAPI ShapeUpgrade | 21:34 |
| fenn | those are the modules i've commented out so far in Modules.py | 21:34 |
| kanzure | gah. | 21:35 |
| kanzure | why are you commenting things out in Modules.py ? | 21:35 |
| kanzure | I mean, what led you to do that? | 21:35 |
| fenn | "Something you can try is to comment out the modules that halt your compilation process." | 21:35 |
| kanzure | where is the std namespace? | 21:36 |
| kanzure | I thought it'd be #include <stdlib> but that's apparently not it | 21:36 |
| kanzure | ah, cstdlib | 21:37 |
| kanzure | (I'm compiling via "python build install" at the moment (after the scons bullshit)) | 21:38 |
| kanzure | (so I've included cstdlib in src/Visualization/Visualization.h) | 21:38 |
| fenn | which OCC packages do you have installed? | 21:39 |
| fenn | or are you using the source from opencascade.org? | 21:39 |
| kanzure | that's a good question. um. | 21:40 |
| kanzure | I have 6.3.0 headers it seems | 21:40 |
| fenn | please just dpkg -la | grep opencascade | 21:40 |
| kanzure | I commented last night about which one I was using, so it's at least in the logs | 21:40 |
| kanzure | yeah, I have some 6.3.0 stuff installed, but also 6.2-7 for libopencascade6.2 and opencascade-examples and opencascade-tools | 21:41 |
| kanzure | error: ‘WDF_NOERASEBKGRND’ was not declared in this scope <--- this would seem to suggest that I don't have the OCC headers? hrm. | 21:41 |
| fenn | you wouldnt have gotten this far without the headers | 21:42 |
| fenn | sigh | 21:42 |
| fenn | IOError: [Errno 28] No space left on device: | 21:42 |
| fenn | i wish opencascade werent so huge | 21:43 |
| kanzure | why is it that this was all so much easier on Windows? even though I had to download a bajillion packages | 21:43 |
| fenn | because it was actually supposed to work? | 21:44 |
| fenn | arent there even binaries for windows? | 21:44 |
| kanzure | yes. | 21:44 |
| kanzure | there were binaries for python, wxpython, pycom or whatever, opencascade, etc. etc. | 21:45 |
| fenn | alright i want to reproduce whatever it is you've done | 21:46 |
| fenn | so could you tar up your src/ directory for me? | 21:46 |
| kanzure | so how hah | 21:46 |
| kanzure | grumble grumble, unedited line | 21:46 |
| kanzure | *hah | 21:46 |
| * fenn squints | 21:46 | |
| fenn | is that really 400MB? | 21:46 |
| fenn | i dont recall downloading 400MB of source | 21:46 |
| kanzure | it generates lots of bullshit | 21:47 |
| fenn | 2076 pythonOCC-md0.1.tar.gz | 21:47 |
| fenn | why is it fucking up the src directory? | 21:47 |
| kanzure | should I just zip up the changes I made? | 21:47 |
| fenn | yeh i guess | 21:47 |
| kanzure | isn't there a way to compare from a .tar.gz what differences exist in a directory based off of md5 hashsum magic? | 21:48 |
| kanzure | it's like an option. :/ | 21:48 |
| fenn | like diff -r? | 21:48 |
| kanzure | well, yes, but explicitly for tar and listing out which files are different | 21:48 |
| fenn | it wouldnt work because there isnt 400MB of source in that .tgz | 21:48 |
| * kanzure compresses | 21:48 | |
| fenn | that's why i edited Modules.py | 21:48 |
| fenn | because it's a source file | 21:49 |
| fenn | i fucking hate generated code | 21:49 |
| fenn | ALWAYS EDIT THE SOURCE | 21:49 |
| kanzure | 235 MB zip file | 21:49 |
| fenn | ok so now for a dilemma | 21:49 |
| fenn | do i delete what i have so i can download yours? | 21:49 |
| fenn | rawr | 21:49 |
| kanzure | hrm. | 21:50 |
| kanzure | how much disk space do you have? | 21:50 |
| fenn | its a 40GB disk | 21:50 |
| fenn | packed full of precious treasures | 21:50 |
| kanzure | you've been living off of 40 GB? wtf mate | 21:50 |
| kanzure | well. hm. | 21:51 |
| fenn | the sad thing is 40GB ought to be plenty | 21:51 |
| fenn | but everything is so stupid and bloated | 21:51 |
| kanzure | I'm up to something like 600 GB of usage on the server. | 21:51 |
| fenn | data expands to fill available capacity | 21:52 |
| kanzure | oh, do you have the ability to wire up a SATA drive? | 21:52 |
| fenn | i suppose so | 21:52 |
| kanzure | I have two spare 500 GB SATA disks, and I'd be willing to let you use/have one | 21:52 |
| fenn | well i can't turn that down | 21:52 |
| kanzure | I bought four, but there were only two ports on my motherboard because I was an idiot | 21:53 |
| kanzure | anyway, for the record- http://heybryan.org/books/Manufacturing/pythoncc/ has the zip file. | 21:54 |
| kanzure | gah, 321 GB of used disk on this box? | 21:54 |
| kanzure | 489 GB used on the server. | 21:55 |
| kanzure | okay, just commented out the lines that were bitching to me. (iiiiiiiiiiiit was just a flag for the window anyway, not like it can do anything terrible) | 22:01 |
| kanzure | "import wx.au" no module named wx.aui | 22:02 |
| kanzure | guess that's wxpython | 22:03 |
| fenn | i thought BRep might be important :x | 22:03 |
| kanzure | maybe :) | 22:03 |
| kanzure | python-wxgtk2.4, 2.6, or 2.8? hrm. | 22:04 |
| fenn | i wish the stupid cache feature would work | 22:04 |
| fenn | it generates 400MB of cache and takes the exact same amount of time to compile | 22:04 |
| kanzure | swig is the culprit. it works for gcc. | 22:04 |
| fenn | ccache? | 22:04 |
| fenn | (i am using ccache btw) | 22:04 |
| kanzure | I mean, when --cache-force is thrown at it, gcc or g++ doesn't seem to compile in between swig calls | 22:04 |
| kanzure | there apparently is a ccache system for swig, but I didn't try to get it to work | 22:05 |
| kanzure | because I'd have to edit the SConstruct file or something | 22:05 |
| fenn | just add a "use cache" directive pretty much | 22:05 |
| kanzure | or if you do the python setup.py install method, there's another way you'd have to edit the swig call obviously. (but then how do you pass the cache force param?) | 22:05 |
| fenn | oh wait swig, not scons | 22:05 |
| fenn | hum | 22:05 |
| fenn | i'm not familiar with any of this stuff | 22:05 |
| kanzure | apt-get install pythoncard <- didn't do what I needed. | 22:05 |
| kanzure | easily googleable though. | 22:06 |
| fenn | well it sounds cool | 22:06 |
| fenn | hey this src zip is taking too long | 22:07 |
| kanzure | okay, trying wxgtk 2.8 | 22:07 |
| kanzure | hrm, would you like something else? | 22:07 |
| fenn | could you cd SWIG*linux*/ and rm *.cc *.os | 22:07 |
| kanzure | what files do you want? the *.i's ? | 22:08 |
| fenn | those are what you modified right? | 22:08 |
| kanzure | I only modified four, you can go do it manually :) | 22:08 |
| fenn | dammit | 22:08 |
| kanzure | AlienImage.i, BRepFeat.i, Image.i, and | 22:08 |
| fenn | i dont believe it | 22:08 |
| kanzure | urhg, one that begins with an x | 22:09 |
| * fenn starts over with a fresh checkout | 22:10 | |
| fenn | grumble grumble | 22:10 |
| fenn | dont they teach you the value of replicability in school? | 22:10 |
| fenn | you know, scientific method and all that | 22:10 |
| kanzure | well, I must have been so overjoyed that it worked, that I forgot to record which file I edited :/ | 22:11 |
| kanzure | and no, I'm really not learning much of anything | 22:11 |
| fenn | hey at least you're busy getting the Fab Lab going right? :P | 22:11 |
| kanzure | except maybe how to be raped | 22:11 |
| kanzure | haha, didn't Paul say that in a recent email? | 22:12 |
| kanzure | I'm so glad I'm able to be in two places at once | 22:12 |
| fenn | so i was talking to les today | 22:13 |
| kanzure | but really, I hope you understand why I'm not there | 22:13 |
| fenn | not really | 22:13 |
| kanzure | oh? | 22:13 |
| kanzure | what do you suggest? | 22:13 |
| fenn | you have to take tests or something? | 22:13 |
| kanzure | classes? | 22:13 |
| fenn | well there isnt anything going on there anyway | 22:13 |
| fenn | part of why i'm grumpy | 22:13 |
| kanzure | what did you and les talk about today? | 22:14 |
| fenn | about idealism and infrastructure and "what to do" | 22:14 |
| kanzure | sounds like a good start | 22:14 |
| kanzure | anything in particular? | 22:15 |
| fenn | he wouldn't believe me when i said that if you start a for profit company and try to grow as big as possible while excluding all competitors you'll just turn into another Dell or GE | 22:15 |
| fenn | regardless whether your intentions were good to start out with | 22:15 |
| fenn | so i'm just wondering if he's playing devil's advocate or what | 22:15 |
| kanzure | I'm missing context | 22:16 |
| fenn | but anyway he said he wouldn't give people the plans for the "open source washing machine" | 22:16 |
| fenn | so WTF is that | 22:16 |
| kanzure | wtf indeed? | 22:16 |
| fenn | some more context was that he wanted to sell picture frames to artists, all good and fine right? | 22:17 |
| fenn | and to grow that business he'd invest in special purpose machines to make the job easier and faster etc | 22:17 |
| fenn | but here's the kicker | 22:17 |
| fenn | it has to be so expensive and hard to get started that none of his employees would start their own business | 22:18 |
| fenn | otherwise they'd "divide the market" | 22:18 |
| fenn | so i was trying to show him that that was evil | 22:18 |
| kanzure | error error market definition | 22:18 |
| * kanzure installs wx from the wxpython repos. http://wiki.wxpython.org/InstallingOnUbuntuOrDebian | 22:19 | |
| kanzure | hurray | 22:21 |
| kanzure | import OCC.NIS | 22:22 |
| kanzure | ImportError: No module named NIS | 22:22 |
| kanzure | *** glibc detected *** python: double free or corruption (!prev): 0x091f54c0 *** | 22:22 |
| kanzure | blah | 22:22 |
| kanzure | commented out that line. | 22:23 |
| kanzure | well, it's running. sort of. | 22:23 |
| kanzure | the window immediately freezes. | 22:24 |
| fenn | cant you just apt-get install python-wxgtk2.6 | 22:24 |
| kanzure | no. | 22:25 |
| fenn | it has to be 2.8? | 22:25 |
| kanzure | I tried that, and it failed. latest version is 2.8.blah. | 22:25 |
| fenn | weak | 22:25 |
| kanzure | yes, but debian does have 2.8 packaged, but it seems to fail? or something? it installs successfully, but it didn't help | 22:25 |
| fenn | -dev? | 22:25 |
| kanzure | anyway, earlier today I had to obviously download the latest wx version for the Windows install, and that worked, so.. | 22:25 |
| kanzure | uh, didn't see a -dev for python-wxgtk2.8 | 22:25 |
| fenn | hm i guess it wouldnt | 22:26 |
| kanzure | ImportError: No module named NIS | 22:26 |
| kanzure | blargh. | 22:26 |
| fenn | what is NIS? | 22:26 |
| kanzure | OCC.NIS; I suspect that it's the visualization thingy, because when I comment out that import line in /usr/lib/python2.5/site-packages/OCC/Display/OCCViewer.py , the wx gui boots up properly, but then nothing is clickable, and there's no visualization widget like I saw on my Windows version | 22:27 |
| fenn | i should have done a fresh checkout after i got the rest of the OCC libraries (WOK and draw) | 22:28 |
| kanzure | w0k? | 22:28 |
| fenn | dunno what it is | 22:28 |
| fenn | WOK (Workshop Organization Kit) is Open CASCADE development environment, which allows a large numbers of developers to work on a variety of products | 22:29 |
| kanzure | http://www.opencascade.org/org/forum/thread_14292/ OCC NIS | 22:29 |
| fenn | hmm scons still doesn't see TKWOK or TKDraw | 22:29 |
| kanzure | I never had it complain about those two packages, or even mention them | 22:30 |
| fenn | it's not an error | 22:31 |
| fenn | in the list of "Checking for..." at the beginning | 22:31 |
| kanzure | hah. | 22:31 |
| kanzure | yeah, I had some issues with tk/tcl a while back for some package I was installing and must have done something special to get it to work back then? | 22:32 |
| fenn | almost all the opencascade libs start with TK | 22:32 |
| fenn | *shrug* | 22:32 |
| kanzure | any thoughts on where I should check for NIC? maybe I missed it somewhere in my steps? which is more like random spagetti steps. | 22:33 |
| fenn | weird now i am getting a bunch of "NOTHING!!!" | 22:33 |
| fenn | if you don't know what you did, I certainly dont know what you did | 22:33 |
| fenn | or why you're trying to import it | 22:34 |
| kanzure | it's from pythonOCC/Tools/InteractiveViewer/InteractiveViewer.py | 22:34 |
| -!- fenn_ [n=fenn@cpe-72-177-52-180.austin.res.rr.com] has joined #hplusroadmap | 22:36 | |
| kanzure | it's from pythonOCC/Tools/InteractiveViewer/InteractiveViewer.py | 22:36 |
| -!- fenn [n=fenn@cpe-72-177-52-180.austin.res.rr.com] has quit [Nick collision from services.] | 22:36 | |
| -!- fenn_ is now known as fenn | 22:37 | |
| fenn | hmm now it's not even whining about AlienImage | 23:17 |
| fenn | i wonder where it even gets this name "BRepTest" | 23:18 |
| kanzure | hah, Nathan officially uber-sucks. | 23:45 |
Generated by irclog2html.py 2.15.0.dev0 by Marius Gedminas - find it at mg.pov.lt!
