Merged revisions 9454,9457-9526 via svnmerge from
svn+ssh://bugman@.../svn/relax/1.3
........
r9454 | bugman | 2009-09-03 15:23:45 +0200 (Thu, 03 Sep 2009) | 3 lines
Updated the non-dangerous Python3k update command.
........
r9457 | bugman | 2009-09-03 17:53:41 +0200 (Thu, 03 Sep 2009) | 3 lines
Created a sub-directory for frame order test scripts.
........
r9458 | bugman | 2009-09-03 18:18:51 +0200 (Thu, 03 Sep 2009) | 3 lines
Shifted the rigid frame order system test into the sub-directory.
........
r9459 | bugman | 2009-09-03 18:31:23 +0200 (Thu, 03 Sep 2009) | 3 lines
Added another rigid frame order system test.
........
r9460 | bugman | 2009-09-03 18:32:10 +0200 (Thu, 03 Sep 2009) | 3 lines
Renamed a frame order system test.
........
r9461 | bugman | 2009-09-03 19:05:02 +0200 (Thu, 03 Sep 2009) | 3 lines
Specific architecture and optimisation details are now printed if the frame order system tests fail.
........
r9462 | bugman | 2009-09-03 19:17:08 +0200 (Thu, 03 Sep 2009) | 3 lines
Fix for the new frame order system test - the results file goes to devnull.
........
r9463 | bugman | 2009-09-03 19:25:01 +0200 (Thu, 03 Sep 2009) | 3 lines
The beta angle for the tensor rotations is now folded to be between 0 and pi.
........
r9464 | bugman | 2009-09-04 10:24:52 +0200 (Fri, 04 Sep 2009) | 5 lines
Added a function for converting the frame order matrix to the rotational superoperator.
The function is called daeg_to_rotational_superoperator().
........
r9465 | bugman | 2009-09-04 11:21:28 +0200 (Fri, 04 Sep 2009) | 5 lines
Fix for the new daeg_to_rotational_superoperator() function.
The frame order matrix shape is now restored.
........
r9466 | bugman | 2009-09-04 16:12:31 +0200 (Fri, 04 Sep 2009) | 5 lines
The print_frame_order_2nd_degree() function now accepts the epsilon arg.
This allows zero to be defined differently making the printout nicer.
........
r9467 | bugman | 2009-09-07 19:28:57 +0200 (Mon, 07 Sep 2009) | 6 lines
Improved and expanded the output of the print_frame_order_2nd_degree() function.
The print out can now be in integer form, dots can replace zeros, and commas between elements are
now optional. Matrix forms are also converted to array forms.
........
r9468 | bugman | 2009-09-07 19:31:49 +0200 (Mon, 07 Sep 2009) | 5 lines
Fix for the __update_model() method.
The theta_axis, phi_axis, and theta_cone params were not being added to the parameter array!
........
r9469 | bugman | 2009-09-08 11:14:08 +0200 (Tue, 08 Sep 2009) | 5 lines
Added some unit tests for the R_euler_zyz() function.
This tests some basic axis rotations.
........
r9470 | bugman | 2009-09-08 11:26:37 +0200 (Tue, 08 Sep 2009) | 3 lines
Fixes for the R_euler_zyz() unit tests.
........
r9471 | bugman | 2009-09-08 11:27:23 +0200 (Tue, 08 Sep 2009) | 5 lines
Fix for the R_euler_zyz() function.
The rotations were being performed in the opposite direction?!?
........
r9472 | bugman | 2009-09-08 14:38:05 +0200 (Tue, 08 Sep 2009) | 3 lines
Removed the incorrect part of the R_euler_zyz() docstring.
........
r9473 | bugman | 2009-09-08 14:40:37 +0200 (Tue, 08 Sep 2009) | 5 lines
Bug fix for the R_to_euler_zyz() function.
This matches the important bug fix in the R_euler_zyz() function.
........
r9474 | bugman | 2009-09-08 14:43:47 +0200 (Tue, 08 Sep 2009) | 5 lines
Fixes for the test_func3() and test_func4() unit tests.
The rotations were backwards because of the R_euler_zyz() fix!
........
r9475 | bugman | 2009-09-08 15:11:17 +0200 (Tue, 08 Sep 2009) | 6 lines
Fix to the frame order system tests.
This corresponds to the fixes in R_euler_zyz() where the rotation matrices were in the left-handed
universe and hence the rotation was back to front.
........
r9476 | bugman | 2009-09-08 16:40:16 +0200 (Tue, 08 Sep 2009) | 3 lines
Created 3 unit tests of the maths_fns.rotation_matrix.quaternion_to_R() function.
........
r9477 | bugman | 2009-09-08 17:07:06 +0200 (Tue, 08 Sep 2009) | 5 lines
Created 3 unit tests of the maths_fns.rotation_matrix.R_axis_angle() function.
These are almost identical to those of quaternion_to_R().
........
r9478 | bugman | 2009-09-08 17:42:48 +0200 (Tue, 08 Sep 2009) | 3 lines
Added a unit test for maths_fns.rotation_matrix.R_to_axis_angle() using R = I.
........
r9479 | bugman | 2009-09-08 17:43:36 +0200 (Tue, 08 Sep 2009) | 5 lines
Bug fix for R_to_axis_angle().
All vector elements were NaN when the angle is 0. Now the zero vector is not normalised by zero.
........
r9480 | bugman | 2009-09-08 17:47:59 +0200 (Tue, 08 Sep 2009) | 3 lines
Added another unit test for maths_fns.rotation_matrix.R_axis_angle().
........
r9481 | bugman | 2009-09-08 18:20:08 +0200 (Tue, 08 Sep 2009) | 5 lines
Function renaming and argument order changes in the maths_fns.rotation_matrix module.
The naming scheme is now more logical (as well as the argument order).
........
r9482 | bugman | 2009-09-08 18:27:50 +0200 (Tue, 08 Sep 2009) | 3 lines
Alphabetical reordering of functions.
........
r9483 | bugman | 2009-09-08 18:52:54 +0200 (Tue, 08 Sep 2009) | 3 lines
Huge simplifications to the unit tests for the maths_fns.rotation_matrix package.
........
r9484 | bugman | 2009-09-08 19:17:29 +0200 (Tue, 08 Sep 2009) | 5 lines
Created the maths_fns.rotation_matrix.axis_angle_to_quaternion() function.
This is accompanied by 2 unit tests.
........
r9485 | bugman | 2009-09-08 19:51:14 +0200 (Tue, 08 Sep 2009) | 5 lines
Created the maths_fns.rotation_matrix.quaternion_to_axis_angle() function.
This is accompanied by 2 unit tests.
........
r9486 | bugman | 2009-09-08 20:20:34 +0200 (Tue, 08 Sep 2009) | 5 lines
Created the maths_fns.rotation_matrix.R_to_quaternion() function.
This is accompanied by 2 unit tests.
........
r9487 | bugman | 2009-09-08 21:07:41 +0200 (Tue, 08 Sep 2009) | 13 lines
Added a unit test to bounce around all the conversion functions in maths_fns.rotation_matrix.
This bounces through:
1) euler_zyz_to_R().
2) R_to_axis_angle().
3) axis_angle_to_quaternion().
4) quaternion_to_axis_angle().
5) axis_angle_to_R().
6) R_to_quaternion().
7) quaternion_to_R().
8) R_to_euler_zyz().
........
r9488 | bugman | 2009-09-09 16:04:40 +0200 (Wed, 09 Sep 2009) | 5 lines
Added a unit test to check if all the maths_fns.kronecker_product.transpose_*() functions revert.
Calling the function twice should return the matrix to the original state.
........
r9489 | bugman | 2009-09-10 14:33:33 +0200 (Thu, 10 Sep 2009) | 3 lines
Renamed the data pipes system tests module.
........
r9490 | bugman | 2009-09-10 16:51:39 +0200 (Thu, 10 Sep 2009) | 3 lines
Wrote a system test for the writing and re-reading of data pipes.
........
r9491 | bugman | 2009-09-10 16:53:06 +0200 (Thu, 10 Sep 2009) | 3 lines
Shifted the valid types private function variable into a module variable.
........
r9492 | bugman | 2009-09-10 16:54:55 +0200 (Thu, 10 Sep 2009) | 3 lines
The test_write_read_pipes() system test is now using the VALID_TYPES module variable to get all pipes.
........
r9493 | bugman | 2009-09-10 17:01:39 +0200 (Thu, 10 Sep 2009) | 3 lines
Deleted the __init__() class methods as these were empty, and shifted the docstring to that of the class.
........
r9494 | bugman | 2009-09-10 17:02:22 +0200 (Thu, 10 Sep 2009) | 3 lines
Removed an empty line.
........
r9495 | bugman | 2009-09-10 17:20:21 +0200 (Thu, 10 Sep 2009) | 5 lines
Added a very basic, and completely empty module and class for SRLS analysis.
This is to allow the data pipe to be created.
........
r9496 | bugman | 2009-09-10 17:37:16 +0200 (Thu, 10 Sep 2009) | 3 lines
Updated the specific_fns module __all__ list for the SRLS module.
........
r9497 | bugman | 2009-09-10 17:38:20 +0200 (Thu, 10 Sep 2009) | 5 lines
Shifted the test_write_read_pipes() system test to the 'state' test module.
The program state is now being saved as the results file is for just one data pipe.
........
r9498 | bugman | 2009-09-10 18:07:22 +0200 (Thu, 10 Sep 2009) | 3 lines
A few fixes for the test_write_read_pipes() system test.
........
r9499 | bugman | 2009-09-10 18:32:31 +0200 (Thu, 10 Sep 2009) | 7 lines
Redesign of the relax state loading for adding support for a XML formatted state file.
Half of the load_state() is now in load_pickle(), and a new function load_xml() added.
The determine_format() function will tell if the file is a pickle or XML file, and then
load_pickle() will run the appropriate loading function.
........
r9500 | bugman | 2009-09-10 18:36:40 +0200 (Thu, 10 Sep 2009) | 5 lines
Added the pickle flag to save_state().
This defaults to True to prevent relax failures during the XML file implementation.
........
r9501 | bugman | 2009-09-10 19:01:25 +0200 (Thu, 10 Sep 2009) | 5 lines
The save_state() function now calls the relax data store to_xml() method.
This method is currently only for one data pipe, and is used by the results writing user function.
........
r9502 | bugman | 2009-09-10 19:24:30 +0200 (Thu, 10 Sep 2009) | 6 lines
The relax data store to_xml() method now can add all data pipes to the XML file (or a subset).
The results.write() method now asks to_xml() for just the current pipe to be stored, whereas
state.save() supplies no 'pipes' argument to have all pipes stored.
........
r9503 | bugman | 2009-09-10 19:26:22 +0200 (Thu, 10 Sep 2009) | 5 lines
Shifted the results.write() force arg to the end.
Bool args should be last!
........
r9504 | bugman | 2009-09-10 19:37:33 +0200 (Thu, 10 Sep 2009) | 3 lines
Removed trailing whitespace.
........
r9505 | bugman | 2009-09-10 19:56:32 +0200 (Thu, 10 Sep 2009) | 3 lines
Updates to the load_xml() function. This now does something!
........
r9506 | bugman | 2009-09-10 20:04:05 +0200 (Thu, 10 Sep 2009) | 3 lines
Created the relax data store is_empty() method.
........
r9507 | bugman | 2009-09-11 10:06:05 +0200 (Fri, 11 Sep 2009) | 6 lines
Added the pickle arg to state.save().
This allows the relax state - the contents of the relax data store - to either be saved as XML or as
a pickled object.
........
r9508 | bugman | 2009-09-11 10:07:11 +0200 (Fri, 11 Sep 2009) | 3 lines
Added a warning to the state.save() user function about the pickle format compatibility.
........
r9509 | bugman | 2009-09-11 10:16:04 +0200 (Fri, 11 Sep 2009) | 6 lines
Absorbed the load_xml() function into load_state().
After shifting the data store check to load_state() for all formats, the function only had one
statement.
........
r9510 | bugman | 2009-09-11 10:29:08 +0200 (Fri, 11 Sep 2009) | 3 lines
Docstring clarification of the dir arg.
........
r9511 | bugman | 2009-09-11 11:12:30 +0200 (Fri, 11 Sep 2009) | 3 lines
Updated the relax data store from_xml() method for support of the loading of the XML state file.
........
r9512 | bugman | 2009-09-11 11:31:44 +0200 (Fri, 11 Sep 2009) | 5 lines
Added the force flag to the state.load() user function.
When True, this causes the relax data store to be reset prior to state loading.
........
r9513 | bugman | 2009-09-11 11:36:02 +0200 (Fri, 11 Sep 2009) | 5 lines
Duplicated the test_state() system test into test_state_pickle() and test_state_xml().
This tests both formats.
........
r9514 | bugman | 2009-09-11 11:39:53 +0200 (Fri, 11 Sep 2009) | 6 lines
Fixes for load_pickle().
The ds.__reset__() call has been removed because of redundancy and the new ds.is_empty() has been
blacklisted.
........
r9515 | bugman | 2009-09-11 11:48:17 +0200 (Fri, 11 Sep 2009) | 3 lines
Changed the default save state to XML.
........
r9516 | bugman | 2009-09-11 13:43:34 +0200 (Fri, 11 Sep 2009) | 3 lines
Added a dir arg to a state.save() call in test_state_pickle().
........
r9517 | bugman | 2009-09-11 13:52:36 +0200 (Fri, 11 Sep 2009) | 5 lines
Renamed the state.load() and state.save() directory argument from dir_name to dir.
This is to bring it in line with the rest of relax.
........
r9518 | bugman | 2009-09-11 14:44:41 +0200 (Fri, 11 Sep 2009) | 3 lines
Fixes for print_frame_order_2nd_degree() for when NaN is encountered.
........
r9519 | bugman | 2009-09-11 14:46:04 +0200 (Fri, 11 Sep 2009) | 5 lines
Fix for the print_frame_order_2nd_degree().
The NaN fix was partly incorrect.
........
r9520 | bugman | 2009-09-11 15:04:15 +0200 (Fri, 11 Sep 2009) | 3 lines
Made the integer printout 4 char wide.
........
r9521 | bugman | 2009-09-11 15:31:42 +0200 (Fri, 11 Sep 2009) | 5 lines
Bug fix for the compile_2nd_matrix_iso_cone() function.
The T23 and not T14 transpose should be performed!
........
r9522 | bugman | 2009-09-11 15:59:04 +0200 (Fri, 11 Sep 2009) | 2 lines
Reordered the relax debugging save state so that the format is yyyymmdd_hhmmss rather than yyyyddmm_hhmmss.
........
r9523 | bugman | 2009-09-11 17:55:23 +0200 (Fri, 11 Sep 2009) | 3 lines
Added a check for the setting of the domain the tensor corresponds to.
........
r9524 | bugman | 2009-09-14 13:59:44 +0200 (Mon, 14 Sep 2009) | 3 lines
Renamed the frame order target function module.
........
r9525 | bugman | 2009-09-14 14:07:40 +0200 (Mon, 14 Sep 2009) | 3 lines
Bug fix, the func_iso_cone_elements() method was been replaced with func_iso_cone().
........
r9526 | bugman | 2009-09-14 14:23:10 +0200 (Mon, 14 Sep 2009) | 5 lines
Fix for r9525.
The func_iso_cone() was now not being set up.
........
