Menu
▾
▴
Re: [PyMOL] I just built pymol, comes up but hangs on any attempt to view a "test" pdb.
|
From: Mike M. <mar...@ho...> - 2012-10-05 14:49:14
|
---------------------------------------- > From: jas...@sc... > Date: Fri, 5 Oct 2012 09:36:29 -0500 > Subject: Re: [PyMOL] I just built pymol, comes up but hangs on any attempt to view a "test" pdb. > To: mar...@ho... > CC: li...@co...; pym...@li... > > Hi Mike, > > >> I suggest downloading the latest code, v1.5.0.4, from the open-source > >> project page, if you want to build by hand. > >> > > see comments in my last post, I think that is what I did. Although > > autoreconf may not have helped anything. > > I see the confusion now. We'll update SourceForge on our next release > so that it says the latest is v1.5.0.5, not v1.5.0.1. To get the > latest code, please do this: > > svn co https://pymol.svn.sourceforge.net/svnroot/pymol/root/pymol/trunk/pymol > pymol > > > >> To build, you need Python 2.7 or later and the process is very simple: > >> > >> sudo python setup.py install > >> sudo python setup2.py install > >> > > that was what created the first version I ran, it seemed fine until > > I tried to display a test pdb then it hung in 100 cpu on one core. > > What kind of computer is this? What video card do you have? > Dell Laitittude, but this is probably due to other issues. > > > I don't really recall what the first version said, I think it was hypothetical > > as in "may need license" etc. The version from "./configure" had a whole > > bunch of problems- simple syntax errors like missing comma to no > > link to the ccealignmodule but once I kluged around the CGO problems > > I think I was on my own :) > > The "make; make install" setup was deprecated a long time ago which is > why this didn't work. We use Python distutils to build, as you've > found out. > I never use python and I probably never would have gotten .1 to work if I had to deal with that :) I'm not sure of the rationale here but while you were phasing out a number of things I thought that "make" was just lagging, intended for developers. > > > Anyway it seems fine now, although attempting to use "fit" on my > > 2 molecules gave a non-intuitive result I think the one was so mangled that > > it had problems aligning them. > > Glad it's working. Fit is a mathematically optimal superposition of > vector sets. Are your atom pairings correct? If you want to specify > them manually, check out Wizard > Pair Fitting. If you send me the > molecules I'm happy to take a look for you, too. > Thanks, I have no idea. Naive overlap was a bit confusing but I'm dealing with simple things right now. I have 2 beta-carotene xtz files created from scripts, one from the mmcif files on pdb.org and another derived from jdftx after 300 position iterations using very simple C and H models. In any case, I think part of the problem is just using .53 for atomic/angstrom conversions but it looks like "fit" results were bizarre: the 2 ends of the molecule were in about the same place but the chains extended in opposite directions from there ( it wasn't immediately obvious if I created some wierd linking errors based on this behaviour LOL). > > Cheers, > > -- Jason > > > > >> > >> > >> On Thu, Oct 4, 2012 at 9:18 PM, Mike Marchywka <mar...@ho...> wrote: > >> > > >> > > >> > > >> > ---------------------------------------- > >> >> CC: pym...@li... > >> >> From: li...@co... > >> >> Subject: Re: [PyMOL] I just built pymol, comes up but hangs on any attempt to view a "test" pdb. > >> >> Date: Thu, 4 Oct 2012 22:03:59 -0400 > >> >> To: mar...@ho... > >> >> > >> >> You don't need a license for the open source version. > >> >> > >> >> If you provided details on what OS/distribution you are building on, people may try to replicate and understand why the python version hung. In my experience, using setup.py generally worked out of the box. > >> > > >> > Well, I thought there may be a dev list or something for the most of this stuff but in any case I'm on Ubuntu > >> > but that probably was not a big factor for some of the issues I ran into I just wanted to display > >> > 2 molecules side by side and could not get jmol or gdis to work right away and someone suggested > >> > this. ( fwiw, I have my own molecule viewer that IIRC I wrote to learn glut on cygwin but > >> > I imagine that would be an even bigger chore to port LOL). > >> > > >> > If I can reconstruct, I apparently started with pymol-v1.5.0.1.tar and I think at some point I ran > >> > autoreconf -l thinking I could go the "configure" route but then did python setup.py build and > >> > install in separeate steps., This seemed to work but attempts to display the test pdb > >> > files never displayed anything, I could not get any more menu's and one core was running > >> > at 100 percent until I killed it. > >> > > >> > Going the confugre route, I think the first problem was with a bunch of undefined CGO things. > >> > but you can see the changes below. The EXT thing was spurious probably as there was a missing -lGLEW > >> > and I had to relink. This is obviously a raw dump of things that I tried along with stuff that ultimately > >> > worked but fwiw. > >> > > >> > > >> > marchywka@marchywka-Latitude-E6510:~/d/pymol/pymol$ find -mtime -2 -name "*.h" > >> > ./layer0/os_gl.h > >> > ./contrib/uiuc/plugins/include/molfile_plugin.h > >> > ./generated/include/ShaderText.h > >> > ./config.h > >> > marchywka@marchywka-Latitude-E6510:~/d/pymol/pymol$ find -mtime -2 -name "*.c" > >> > ./layer0/ShaderText.c > >> > ./layer0/ShaderMgr.c > >> > ./layer3/Executive.c > >> > ./generated/src/ShaderText.c > >> > ./layer2/ObjectVolume.c > >> > ./layer1/CGO.c > >> > ./layer1/Scene.c > >> > marchywka@marchywka-Latitude-E6510:~/d/pymol/pymol$ > >> > > >> > > >> > > >> > marchywka@marchywka-Latitude-E6510:~/d/pymol/pymol$ grep -n -A 2 mjm `find -mtime -2 -name "*.c" ` > >> > ./layer0/ShaderMgr.c:97: FeedbackAdd(G, " mjm forcing glew ok Shaders available.\n"); > >> > ./layer0/ShaderMgr.c:98:// this should work but turn off for now mjm > >> > ./layer0/ShaderMgr.c-99-// if (GLEW_VERSION_2_0) > >> > ./layer0/ShaderMgr.c-100- if (1) > >> > -- > >> > ./layer3/Executive.c:76:// mjm > >> > ./layer3/Executive.c-77-#include"modules/cealign/src/ce_types.h" > >> > ./layer3/Executive.c-78-/* Externals for CEAlign */ > >> > -- > >> > ./layer1/Scene.c:2910://mjm > >> > ./layer1/Scene.c-2911- 1.0F, 1.0F, 0.1F, 1.f > >> > ./layer1/Scene.c-2912- }; > >> > -- > >> > ./layer1/Scene.c:10000: //mjm > >> > ./layer1/Scene.c-10001-//glBlitFramebufferEXT (0, 0, I->offscreen_width / 2, I->offscreen_height, > >> > ./layer1/Scene.c-10002- glBlitFramebuffer (0, 0, I->offscreen_width / 2, I->offscreen_height, > >> > -- > >> > ./layer1/Scene.c:10007://mjm > >> > ./layer1/Scene.c-10008- glBlitFramebuffer (I->offscreen_width / 2, 0, I->offscreen_width, I->offscreen_height, > >> > ./layer1/Scene.c-10009- //glBlitFramebufferEXT (I->offscreen_width / 2, 0, I->offscreen_width, I->offscreen_height, > >> > -- > >> > ./layer1/Scene.c:10016: //mjm > >> > ./layer1/Scene.c-10017- //glBlitFramebufferEXT (0, 0, I->offscreen_width, I->offscreen_height, > >> > ./layer1/Scene.c-10018- glBlitFramebuffer (0, 0, I->offscreen_width, I->offscreen_height, > >> > -- > >> > ./layer1/Scene.c:10024: //mjm > >> > ./layer1/Scene.c-10025-// glBlitFramebufferEXT (0, 0, I->offscreen_width, I->offscreen_height, > >> > ./layer1/Scene.c-10026- glBlitFramebuffer (0, 0, I->offscreen_width, I->offscreen_height, > >> > marchywka@marchywka-Latitude-E6510:~/d/pymol/pymol$ > >> > > >> > > >> > marchywka@marchywka-Latitude-E6510:~/d/pymol/pymol$ grep -n -A 2 mjm `find -mtime -2 -name "*.h" ` > >> > ./layer0/os_gl.h:6:// mjm > >> > ./layer0/os_gl.h-7-//#if defined(_PYMOL_IOS) || defined(ANDROID) > >> > ./layer0/os_gl.h-8-#if 1 > >> > ./layer0/os_gl.h:9:// mjm > >> > ./layer0/os_gl.h-10-// #define _PYMOL_PURE_OPENGL_ES > >> > ./layer0/os_gl.h-11-#define _PYMOL_CGO_DRAWARRAYS > >> > -- > >> > ./contrib/uiuc/plugins/include/molfile_plugin.h:249:// mjm - these are in the games lass LOL > >> > ./contrib/uiuc/plugins/include/molfile_plugin.h-250-int have_esp; > >> > ./contrib/uiuc/plugins/include/molfile_plugin.h-251-int have_npa; > >> > -- > >> > ./contrib/uiuc/plugins/include/molfile_plugin.h:312:// mjm > >> > ./contrib/uiuc/plugins/include/molfile_plugin.h-313-int multiplicity; > >> > ./contrib/uiuc/plugins/include/molfile_plugin.h-314-int * esp_charges; > >> > marchywka@marchywka-Latitude-E6510:~/d/pymol/pymol$ > >> > > >> > > >> > > >> > marchywka@marchywka-Latitude-E6510:~/d/pymol/pymol$ cat relink > >> > gcc --verbose -shared -fPIC -DPIC -Wl,--no-as-needed -Wl,--whole-archive contrib/uiuc/plugins/molfile_plugin/src/.libs/libmolfile_plugin0.a modules/cealign/src/ccealignmodule.o ov/src/.libs/libov.a layer0/.libs/liblayer0.a layer1/.libs/liblayer1.a layer2/.libs/liblayer2.a layer3/.libs/liblayer3.a layer4/.libs/liblayer4.a layer5/.libs/liblayer5.a -Wl,--no-whole-archive -L/usr/lib/x86_64-linux-gnu -Wl,--whole-archive -lGLEW -lGL -lGLU /usr/lib/x86_64-linux-gnu/libfreetype.so -lz -lpng -lglut -Wl,--no-whole-archive -ldl -O2 -Wl,-soname -Wl,libpymol.so.0 -o .libs/libpymol.so.0.0.0 > >> > #gcc -shared -fPIC -DPIC -Wl,--whole-archive contrib/uiuc/plugins/molfile_plugin/src/.libs/libmolfile_plugin0.a ov/src/.libs/libov.a layer0/.libs/liblayer0.a layer1/.libs/liblayer1.a layer2/.libs/liblayer2.a layer3/.libs/liblayer3.a layer4/.libs/liblayer4.a layer5/.libs/liblayer5.a -Wl,--no-whole-archive -L/usr/lib/x86_64-linux-gnu -lGL -lGLU /usr/lib/x86_64-linux-gnu/libfreetype.so -lz -lpng -lglut -ldl -O2 -Wl,-soname -Wl,libpymol.so.0 -o .libs/libpymol.so.0.0.0 > >> > marchywka@marchywka-Latitude-E6510:~/d/pymol/pymol$ > >> > > >> > > >> > > >> > > >> > > >> >> > >> >> -David > >> >> > >> >> On Oct 4, 2012, at 9:22 PM, Mike Marchywka <mar...@ho...> wrote: > >> >> > >> >> > > >> >> > > >> >> > > >> >> > fwiw, I never figured out why the out-of-the-box build hung using the python approach > >> >> > but I went through the "./cogifure, make make install route" > >> >> > and did get the test molecules to display. However, I ran into lots of problems with the ShaderMgr and > >> >> > ccealignmodule and then finally the link was missing -lGLEW. > >> >> > > >> >> > In any case it looks like I am all set but I thought the download would come closer to working right > >> >> > away. > >> >> > > >> >> > And the xyz files I wanted to display do appear to come up just great. > >> >> > > >> >> > So do I need a license for the open source version? > >> >> > Thanks. > >> >> > > >> >> > > >> >> > > >> >> > note new address > >> >> > Mike Marchywka 2295 Collinworth Drive Marietta GA 30062. > >> >> > formerly 487 Salem Woods Drive Marietta GA 30067 404-788-1216 (C)<- leave message 989-348-4796 (P)<- emergency > >> >> > > >> >> > > >> >> > > >> >> > ---------------------------------------- > >> >> >> From: mar...@ho... > >> >> >> To: pym...@li... > >> >> >> Date: Thu, 4 Oct 2012 15:43:53 -0400 > >> >> >> Subject: [PyMOL] I just built pymol, comes up but hangs on any attempt to view a "test" pdb. > >> >> >> > >> >> >> > >> >> >> > >> >> >> > >> >> >> Is there some common reason for this? > >> >> >> CPU on one core stays at 100 percent and no more menu's come up. > >> >> >> > >> >> >> The larger window says at the bottom that the version is supposed to be licensed > >> >> >> but I just pulled it down from sourceforge and not sure that this version needs a key > >> >> >> to work. > >> >> >> > >> >> >> Thanks. > >> >> >> > >> >> >> > >> >> >> note new address > >> >> >> Mike Marchywka 2295 Collinworth Drive Marietta GA 30062. > >> >> >> formerly 487 Salem Woods Drive Marietta GA 30067 404-788-1216 (C)<- leave message 989-348-4796 (P)<- emergency > >> >> >> > >> >> >> > >> >> >> ------------------------------------------------------------------------------ > >> >> >> Don't let slow site performance ruin your business. Deploy New Relic APM > >> >> >> Deploy New Relic app performance management and know exactly > >> >> >> what is happening inside your Ruby, Python, PHP, Java, and .NET app > >> >> >> Try New Relic at no cost today and get our sweet Data Nerd shirt too! > >> >> >> http://p.sf.net/sfu/newrelic-dev2dev > >> >> >> _______________________________________________ > >> >> >> PyMOL-users mailing list (PyM...@li...) > >> >> >> Info Page: https://lists.sourceforge.net/lists/listinfo/pymol-users > >> >> >> Archives: http://www.mail-archive.com/pym...@li... > >> >> > > >> >> > ------------------------------------------------------------------------------ > >> >> > Don't let slow site performance ruin your business. Deploy New Relic APM > >> >> > Deploy New Relic app performance management and know exactly > >> >> > what is happening inside your Ruby, Python, PHP, Java, and .NET app > >> >> > Try New Relic at no cost today and get our sweet Data Nerd shirt too! > >> >> > http://p.sf.net/sfu/newrelic-dev2dev > >> >> > _______________________________________________ > >> >> > PyMOL-users mailing list (PyM...@li...) > >> >> > Info Page: https://lists.sourceforge.net/lists/listinfo/pymol-users > >> >> > Archives: http://www.mail-archive.com/pym...@li... > >> > > >> > ------------------------------------------------------------------------------ > >> > Don't let slow site performance ruin your business. Deploy New Relic APM > >> > Deploy New Relic app performance management and know exactly > >> > what is happening inside your Ruby, Python, PHP, Java, and .NET app > >> > Try New Relic at no cost today and get our sweet Data Nerd shirt too! > >> > http://p.sf.net/sfu/newrelic-dev2dev > >> > _______________________________________________ > >> > PyMOL-users mailing list (PyM...@li...) > >> > Info Page: https://lists.sourceforge.net/lists/listinfo/pymol-users > >> > Archives: http://www.mail-archive.com/pym...@li... > >> > >> > >> > >> -- > >> Jason Vertrees, PhD > >> PyMOL Product Manager > >> Schrödinger, Inc. > >> > >> (e) Jas...@sc... > >> (o) +1 (603) 374-7120 > > > > > > -- > Jason Vertrees, PhD > PyMOL Product Manager > Schrödinger, Inc. > > (e) Jas...@sc... > (o) +1 (603) 374-7120 |
