jnsturgis / hard_discs2 Goto Github PK

Describe the bug
config2eps does not handle non-rectangular regions.
The colors need to be upgraded in the documentation.
The eps file should not be A4 always and in the bottom corner.
Separate C++ code and postscript strings...

Clarify the format of the log files produced.
Separate the format from the middle of the code.
Make sure errors produce helpful messages in the log file.
Make efforts to have a parsable output (by gnuplot for example).

For observed patches - with options of fine-ness

Describe the bug
NVT crashes with test2.config because it is not a periodic rectangle...

Expected behavior
Should run with non-periodic conditions...
Also, test that periodic conditions are compatible with shape (make rectangle if necessary)...
The documentation of NVT also needs to be updated and improved.

Tidy up the command line handling and error reporting associated with it.

Describe the bug
NVT run with repulsive potential and complex molecules does not keep all of every molecule inside the box in the y direction.

To Reproduce
NVT -t topo3eps.dat -f AqpZ_ff_repulsion.ff -o AqpZ_repulsefaible1000000.conf -c AqpZ_repulsefaible.conf -n 20000 -s AqpZ_repulsefaible.gz 1000000 20000 1 1

Expected behavior
Nothing escaping from top of box....

Screenshots

AqpZ_repulsefaible1000000.pdf

Desktop (please complete the following information):

• OS: [e.g. iOS]
• Browser [e.g. chrome, safari]
• Version [e.g. 22]

Smartphone (please complete the following information):

• Device: [e.g. iPhone6]
• OS: [e.g. iOS8.1]
• Browser [e.g. stock browser, safari]
• Version [e.g. 22]

Additional context
Add any other context about the problem here.

Requires robust definition of the configuration file (including non-rectangular border, and boundary conditions) but backward compatible. Requires robust output datafile (suitable for gnuplot or pyplot style graphing routines).

Would be an improvement if:

• For the force field interaction potentials, the length scale could be interaction dependent.
• It is not clear that the colors belong in the force field.
• The cut-off does not really belong in the force-field.

Go over the documentation in refman.pdf (and the HTML version)...

Add images (decide what) but at least the color swatch for predefined colors. See
http://www.doxygen.nl/manual/commands.html#cmdimage
for how it should be done.

Describe the bug
Written forcefield is not rereadable

Expected behavior
Forcefield written to log or elsewhere should be readable by read routine (i.e. same syntax details).

Topology destructor does not completely free memory of atoms... especially when created with float the atom is never destroyed...
Needs unit test and check with valgrind...

g(6)r for example. Can this be done in such a way as to add different patches together?

Currently, the postscript generation programs have independent copies of the preamble, dictionaries and ending... this is suboptimal as there are specific functions added to the different dictionaries for example.

There should only be one version of these constant strings - perhaps a postscript output stream class is the answer.

Describe the bug
Currently, atom sizes are defined in the topology file (in part to ensure that placement and drawing do not depend on interactions) and in the force field file (because it is part of the interaction potential) this is schizophrenic and I should decide on one option.

Expected behavior
A single place where atom and object sizes are defined

Describe the bug
Colinear points on the convex hull can produce nan in polygon coordinate output, and generally cause problems with the algorithms.

To Reproduce
analysis/wrap -v -o hex2.config <test/hex.config

Expected behavior
Configuration without nan in the polygon saved to file, either colinear points still present or removed as in test/hex2.config

Allow the configuration file to contain blank lines (which are ignored) and comments on lines starting with '#'

Describe the solution you'd like
To be able to run multiple MCMC trajectories in parallel with different temperatures (saving individual trajectories) and swapping between temperatures. This should greatly increase the efficiency of solution finding etc.

Allow the topology file to contain blank lines (which are ignored) and comments on lines starting with '#'

Is your feature request related to a problem? Please describe.
Calculation of order parameters for discs... without intrinsic directionality

Describe the solution you'd like
Maybe can do this by pre-treatment of configuration (rotate all of a certain type of object to point to nearest neighbour)