relax SVN archive

Fix for bug #7241 (https://gna.org/bugs/?7241).

This is fix for the changes made at r2591 which were an attempt to fix the bug. Those changes did
not work though as the Scientific PDB data structure always contains the objects 'peptide_chains'
and 'nucleotide_chains'. Therefore these are now tested to see if they are empty rather than if
they exist.

The algorithm remains the same as in r2591, peptide chains are searched for and used first, then
nucleotide chains are searched for, and if neither are present RelaxNoPdbChainError is raised.

Fix for bug #7226 (https://gna.org/bugs/?7226).

This bug was reported by Alex Hansen. The problem was caused by the model-free results file reading
function, specifically for model m0. The parameter array for this model is the empty array '[]'.
However the test for setting up the model-free models was being bypassed because the empty array
evaluates to False. By removing the test for the parameter array and only checking for the model
name and equation type, the bug is removed.

An attempt at fixing bug #7241 (https://gna.org/bugs/?7241).

The algorithm I've implemented is to first check for and use peptide chains, then check for and use
nucleotide chains. If neither are present a new error RelaxNoPdbChainError is raised.

Added tests for the Dasha, Molmol, Grace, and OpenDX executables.

This is an extension of the ideas introduced to fix bug #7225 (https://gna.org/bugs/?7225). This
change was discussed at https://mail.gna.org/public/relax-devel/2006-09/msg00034.html (Message-id:
<7f080ed10609280917s139d2c2cm9f0edcbd5a9fc04c@...>).

Fix to bug #7225.

This bug was reported by Alex Hansen at https://gna.org/bugs/index.php?7225.

To enable comprehensible error messages when attempting to execute Modelfree4, the function
'self.test_binary()' was added to the file 'io.py'. This was described in the post located at
https://mail.gna.org/public/relax-devel/2006-09/msg00034.html (Message-id:
<7f080ed10609280917s139d2c2cm9f0edcbd5a9fc04c@...>). The implementation is slightly
different to the post though. Firstly the supplied string corresponding to the binary is tested to
see if the full path has been supplied. If so then the string is tested to see if the file exists
and if the file is executable. Otherwise the file is searched for in the PATH environmental
variable. Point 5 of the post at https://mail.gna.org/public/relax-devel/2006-09/msg00034.html,
testing that the binary is in the current directory was not implemented. The reason is because I
was unable to successfully modify the PATH variable.

Three new RelaxErrors have been created to print out comprehensible messages:
RelaxMissingBinaryError, RelaxNonExecError, and RelaxNoInPathError.

Modified the Modelfree4 execution code to test the return status when a PDB file is used.

Fix for bug #7224 (https://gna.org/bugs/?7224).

The problem was simply that the run name was being replaced by None.

Implemented a test for the test suite which catches bug #7224 (https://gna.org/bugs/?7224).

Partial fix for bug #7224 (https://gna.org/bugs/?7224).

The 'param_types' argument of the 'diffusion_tensor.init()' user function was changed to the value
of 3. The rational for this fix is given in the post at
https://mail.gna.org/public/relax-devel/2006-09/msg00042.html (Message-id:
<7f080ed10609300034j2e2487f2p2e252eb941a38e33@...>).

Changed the CSA values in the sample scripts from -160 ppm to -170 ppm.

This idea was suggested at https://mail.gna.org/public/relax-users/2006-09/msg00045.html
(Message-id: <7f080ed10609290808u3d30d31cp8ca6040a239a806d@...>). The parent thread by
Sébastien Morin of the subject of the CSA & bond length is located at
https://mail.gna.org/public/relax-users/2006-09/msg00042.html (Message-id:
<451C1910.6060108@...>).