relax SVN archive

svn+ssh://bugman@.../svn/relax/1.3

........
r15471 | bugman | 2012-03-12 15:26:57 +0100 (Mon, 12 Mar 2012) | 5 lines

Added some code to detect the bit version of MS Windows in the information print out.

This should help better determine if 32 or 64-bit Windows is being used.
........
r15473 | bugman | 2012-03-13 16:56:38 +0100 (Tue, 13 Mar 2012) | 3 lines

Added the ability to skip scripts in the package __all__ list unit test checks.
........
r15474 | bugman | 2012-03-13 16:57:14 +0100 (Tue, 13 Mar 2012) | 3 lines

The multi-processor package 'test_implementation.py' script is now blacklisted from the __all__ check.
........
r15479 | bugman | 2012-03-15 16:48:13 +0100 (Thu, 15 Mar 2012) | 5 lines

The extern and minfx.scipy_subset relax packages are now excluded form the API documentation.

This will be updated at http://www.nmr-relax.com/api/ after the next relax release.
........
r15480 | bugman | 2012-03-15 16:58:15 +0100 (Thu, 15 Mar 2012) | 3 lines

Epydoc docstring fix.
........
r15481 | bugman | 2012-03-15 17:05:18 +0100 (Thu, 15 Mar 2012) | 3 lines

The setup.py script can now be imported for the epydoc API documentation system.
........
r15482 | bugman | 2012-03-15 17:06:12 +0100 (Thu, 15 Mar 2012) | 5 lines

The multi-processor test implementation scripts are now excluded from the API documentation.

The import of the scripts was failing, and they are not part of the relax API.
........
r15483 | bugman | 2012-03-15 17:21:14 +0100 (Thu, 15 Mar 2012) | 3 lines

Formatting of the epydoc exclusion list.
........
r15487 | bugman | 2012-03-15 20:05:40 +0100 (Thu, 15 Mar 2012) | 13 lines

Fix for bug #19528 (https://gna.org/bugs/?19528).

This bug was reported by Hugh Dannatt (h dott dannatt att sheffield dot ac dot uk).

The problem was in the multi-processor code for the model-free calculations, specifically for the
Monte Carlo simulations. Rather than processing the queue of slave processor commands for the
model-free calculation for each spin system at the end of each simulation, the processing of the
queue was happening at the end of all Monte Carlo simulations. This caused memory usage to bloat!
A full description of the problem can be found at
https://mail.gna.org/public/relax-users/2012-03/msg00037.html (Message-id:
<CAED9pY_Z1wGEg3gZo3WCkhvYTBdqs-Feex5Og4vJFovgJkUOyw@...>).
........