IMP-dev

imp-dev@salilab.org

869 discussions

Re: [IMP-dev] [IMP-commits] r1414 - trunk/modules/algebra/include
by Daniel Russel 02 Feb '09

02 Feb '09

Given we already have rotation from_fixed_xyz which is a form of euler angles, this function should probably have a more descriptive name describing which Euler angles it uses. On Feb 2, 2009, at 10:10 PM, Notification of IMP commits wrote: > Author: kerenl(a)SALILAB.ORG > Date: 2009-02-02 22:10:39 -0800 (Mon, 02 Feb 2009) > New Revision: 1414 > > Modified: > trunk/modules/algebra/include/Rotation3D.h > Log: > add rotation_from_euler > > Modified: trunk/modules/algebra/include/Rotation3D.h > =================================================================== > --- trunk/modules/algebra/include/Rotation3D.h 2009-02-03 03:11:05 > UTC (rev 1413) > +++ trunk/modules/algebra/include/Rotation3D.h 2009-02-03 06:10:39 > UTC (rev 1414) > @@ -114,6 +114,24 @@ > return Rotation3D(a,b,c,d); > } > > +//! Initialize a rotation from euler angles > +/** > + \note http://en.wikipedia.org/wiki/ > + Conversion_between_quaternions_and_Euler_angles > + \relates Rotation3D > +*/ > +inline Rotation3D rotation_from_euler(Float phi, Float theta, Float > psi) > +{ > + Float a,b,c,d; > + Float c1,c2,c3,s1,s2,s3; > + c2=cos(theta/2);c1=cos(phi/2);c3=cos(psi/2); > + s2=sin(theta/2);s1=sin(phi/2);s3=sin(psi/2); > + a = c1*c2*c3+s1*s2*s3; > + b = s1*c2*c3-c1*s2*s3; > + c = c1*s2*c3+s1*c2*s3; > + d = c1*c2*s3-s1*s2*c3; > + return Rotation3D(a,b,c,d); > +} > //! Generate a Rotation3D object from a rotation matrix > /** > \throw ValueException if the rotation is not a rotation matrix. > > _______________________________________________ > IMP-commits mailing list > IMP-commits(a)salilab.org > https://salilab.org/mailman/listinfo/imp-commits

1 0

Moving Movers to Containers
by Daniel Russel 02 Feb '09

02 Feb '09

As part of the gradual move to using containers, I would like to move the Mover classes (NormalMover and BallMover) to use the container classes. The visible change would be that they would take a non-optional SingletonContainer as the first argument of the constructor instead of an optional list of particles as the last argument. Any reasons not to?

1 0

how can you retrieve the atom index from AtomDecorator ?
by Keren Lasker 02 Feb '09

02 Feb '09

3 8

problems with IMP.modeller.read_pdb
by Keren Lasker 02 Feb '09

02 Feb '09

hey, Reading either the attached files I get: File "../../../utils/transform_mol.py", line 7, in <module> mp= IMP.modeller.read_pdb(sys.argv[1],m) File "/viola1/home/kerenl/bioinformatics/projects/imp/build/lib/IMP/ modeller/__init__.py", line 460, in read_pdb hcp.add_child(hrp) File "/viola1/home/kerenl/bioinformatics/projects/imp/build/lib/IMP/ core/__init__.py", line 1985, in add_child def add_child(*args): return _IMP_core.MolecularHierarchyDecorator_add_child(*args) RuntimeError: Bad key index File "build/include/IMP/internal/AttributeTable.h", line 245 any clue why ? thanks, Keren.

2 1

can I add HETATM to MolecularHierarchyDecorator::Type ?
by Keren Lasker 02 Feb '09

02 Feb '09

2 1

documentation build
by Daniel Russel 30 Jan '09

30 Jan '09

Upon Javi's suggestion, we should probably have the online IMP docs updated every night whether or not all the builds and tests success so it tracks, as closely as possibly, the svn state. With so many people contributing to IMP, it is likely that something will be broken most nights, so holding it up based on that makes it less useful. It would be good to also have a release version of the documentation once we have releases.

2 2

benchmarks
by Daniel Russel 30 Jan '09

30 Jan '09

After the discussion of connectivityrestraint performance, I suggest we add a benchmarks folder to IMP. This folder will have C++ programs which time some aspect of IMP so that we can see how speed is doing (I can't figure out how to profile code run from python, which is necessary any time we want to get into running times). Eventually it would be nice to have them run periodically on a reference machine and recorded, but that can wait as it sounds annoying to set up :-)

2 1

Cleanup of config/build/install script
by Daniel Russel 30 Jan '09

30 Jan '09

I just committed a cleanup of the config/build/install scripts following comments and patches posted previously. The changes have the following user-visible consequences: new arguments added to scons: - cxxflags, ldflags which allow the addition of arbitrary compile and link flags - pythonpath which allows dirs to be added to the python path linking has been changed as follows: - on a mac the library ids are now set properly, both for when the libraries are in build/lib and when they are in installed. This means that you can link against them without hassle once installed. It also means that you just have to have PYTHONPATH set properly in order to use IMP - search paths orderings should be consistent everywhere - on linux, unusual directories containing libraries (i.e. things passed in the libflags argument) are added to the libraries runtime search path. This makes it less likely that a library is linked against one external library and run against another, as was easily the case before. imppy.sh has been changed as follows: - the local paths are prepended to the search path, not appended, so local copies of IMP will hide ones elsewhere on the system - external LD_LIBRARY_FLAGS and PYTHONPATH and no longer propagated from the environment (but things can be added to them via scons). This means that "scons test" runs tests in the same environment as if you run the tests yourself with ./bin/imppy.sh - on a mac, LD_LIBRARY_FLAGS is not set as it is no longer needed I have tested everything on my linux box and mac, so I am reasonably confident that they should work. But please report any new problems promptly. And hopefully any issues with other platforms should be cleared up by Monday. Unfortunately, it is still not possible to use the diva copy of modeller as the modeller test scripts are kind of strange and I didn't feel like messing with them. Likewise, since EMBed doesn't have a proper installer, it is still a special case.

1 0

FloatKey In a Double precision?
by Seung Joong Kim 30 Jan '09

30 Jan '09

Dear all IMP developers, By default, Float data type is defined as "typedef float Float;" and it can have a corresponding FloatKey. But in the case of "double" data type for higher accuracy, there is no DoubleKey assigned currently. I'm just curious about it. Do you think is it necessary to define DoubleKey as well? or is it safe to use FloatKey instead? Thanks a lot! Seung Joong Kim

4 5

a note on operator>> and operator<<
by Daniel Russel 29 Jan '09

29 Jan '09

The correct way to have operator<< and operator>> work for reading and writing a class is to declare them as functions in the namespace containing the object. Declaring them as member functions does something else entirely (or absolutely nothing useful, depending on how you do it). The problem is that the compiler searches for operators by checking member functions of the left hand object (for i/o it is the stream itself), then the current namespace and the namespaces containing the arguments. It never looks for member functions of the right hand type. I suggest implementing operator>> as: struct Mine { void show(std::ostream &out=std::cout) const; }; IMP_OUTPUT_OPERATOR(Mine)

1 0

← Newer
1
...
50
51
52
53
54
55
56
...
87
Older →

Jump to page:

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

IMP-dev