IMP-dev

imp-dev@salilab.org

869 discussions

Decorator array cleanup
by Daniel Russel 13 Jun '08

13 Jun '08

This patch cleans up the code for the attribute arrays in decorators. In terms of API changes it adds a method to get the ith residue from a protein or chain or nucleotide. It also now enforces the create/cast distinction with bonds, bonded and hierarchy nodes (previously any node was a hierarchy node without having create cast on it).

3 5

Particle refines
by Daniel Russel 12 Jun '08

12 Jun '08

This patch adds particle refines to IMP. There is a base class in <IMP/ ParticleRefiner.h> and a couple in <IMP/particle_refiners>. The current particle refines cover an bond with a set of spheres (created on the fly) and move down one level in the hierarchy. I added some new meothods and classes to the python test base class too. SWIG was having problems with figuring out that Particle Refiner was not a virtual class when I had all the methods be virtual, so they are now all non-virtual.

1 0

exception cleanup
by Daniel Russel 11 Jun '08

11 Jun '08

I changed how the check and assert macros work slightly. The former now takes an exception type name as its third argument (instead of an instructor call). The exception is created with the error message passed as the second argument to check (removing the need to specify the message twice). Also, the file and line number are added to the assertion error message. Finally, by default the error message is not printed, just passed to the exception. Python will print the error message from the exception anyway.

1 0

unit cleanup
by Daniel Russel 11 Jun '08

11 Jun '08

The C++ unit classes have been cleaned up and a few new types have been added. Scalar went away as there is now a common base class for all units (so adding operations with floats is easy). The usual caveats about SConscript files not including new headers. (IMP.i shouldn't)

1 0

ParticleRefiner
by Daniel Russel 06 Jun '08

06 Jun '08

I think we should have a new class of object, which I am, for the time being calling a ParticleRefiner to handle hierarchical collision detection and generation of particles on the fly. I am currently using it to generate particles to cover the edges of my unstructured chains in order to better preserve excluded volume. The interface looks more or less like: ParticleRefiner { // return a bunch of particles which refines p Particles refine(Particle p) // clean up the particles which refined p if needed cleanup(Particle p, Particles); }; And the methods must be called as a pair with the result of the first method passed as the second argument to the second method. To use this I currently have a PairScore which I apply to the nonbonded pairs and a ParticleRefine which covers each bond with a set of particles chosen to conserve volume. The PairScore then applies another pair score to all pairs of refined particles. For that particle refined, "refine" creates new particles and "cleanup" destroys them (it probably makes sense to cache particles later). ParticleRefines could also simply move down in a hierarchy, allowing hierarchical collision detection and such. Then cleanup wouldn't do anything at all. Do other people have related use cases they are interested in?

2 2

Bug in GDB 6.6
by Daniel Russel 03 Jun '08

03 Jun '08

The version of GDB installed on the fedora 8 machines has a bug dealing with infinity when running on 64 bit machines. Floats with the value infinity are printed as 0 in the debugger and any functions which take the float by value all called with 0 instead of infinity if you call then the debugger. It can be very confusing :-) It is fixed in gdb 6.8 but there aren't rpms for fedora 8 that I could find. Just so people are aware.

2 2

ParticleRefiner 2
by Daniel Russel 03 Jun '08

03 Jun '08

And cleanup can propagate derivatives to the correct particles.

1 0

Turning on more refcounting
by Daniel Russel 30 May '08

30 May '08

Here is a patch which turns on refcounting for most of the objects (Optimizers and Model being the two exceptions, although, now that I think about, model should have it (but I would have to change ObjectPointer in order to do so, since lots of things have pointers back to model which can't keep the model alive). It shouldn't change anything for python users other than you can add restraints to multiple restraint sets. We should add a check for dependency cycles in restraint sets, but that takes some work. It also adds a trivial restraint ConstantRestraint. Mostly there for testing (since otherwise the ref count check gets buried under the work of creating all the bits needed for a restraint).

1 0

valgrind suppressions for 64 bit machines
by Daniel Russel 30 May '08

30 May '08

I get a bunch in modeller too which I can add. Index: tools/valgrind-python.supp =================================================================== --- tools/valgrind-python.supp (revision 584) +++ tools/valgrind-python.supp (working copy) @@ -13,6 +13,18 @@ } { + ADDRESS_IN_RANGE/Invalid read of size 8 + Memcheck:Addr8 + fun:PyObject_Free +} + +{ + ADDRESS_IN_RANGE/Invalid read of size 8 + Memcheck:Value8 + fun:PyObject_Free +} + +{ ADDRESS_IN_RANGE/Conditional jump or move depends on uninitialised value Memcheck:Cond fun:PyObject_Free @@ -31,6 +43,18 @@ } { + ADDRESS_IN_RANGE/Invalid read of size 8 + Memcheck:Addr8 + fun:PyObject_Realloc +} + +{ + ADDRESS_IN_RANGE/Invalid read of size 8 + Memcheck:Value8 + fun:PyObject_Realloc +} + +{ ADDRESS_IN_RANGE/Conditional jump or move depends on uninitialised value Memcheck:Cond fun:PyObject_Realloc

3 6

Improving the test framework
by Daniel Russel 29 May '08

29 May '08

Is it possible to have unittest run each test in a separate process? That we we don't have to worry about bad memory accesses in one test crashing python and messing up later tests (and hence can leave failed tests in until we can actually fix them). We are kind of abusing unittest (by using it to test C++ code) so it may not be easy to do.

3 3

← Newer
1
...
66
67
68
69
70
71
72
...
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