Menu

SVN archive Commit Log


Commit Date  
[r15431] by bugman

Merged revisions 15425-15428 via svnmerge from
svn+ssh://bugman@.../svn/relax/1.3

........
r15425 | bugman | 2012-03-01 14:28:41 +0100 (Thu, 01 Mar 2012) | 3 lines

Epydoc docstring fixes.
........
r15426 | bugman | 2012-03-01 15:06:41 +0100 (Thu, 01 Mar 2012) | 3 lines

Bug fix for old numpy versions missing the float16 object.
........
r15427 | bugman | 2012-03-01 16:15:40 +0100 (Thu, 01 Mar 2012) | 7 lines

The finish button in the new analysis wizard has been renamed to "Start".

This is based on feedback from Hessam Rezaei-Ghaleh.

Now all of the text and icons for the wizards can be programmatically changed.
........
r15428 | bugman | 2012-03-01 16:32:34 +0100 (Thu, 01 Mar 2012) | 3 lines

Expansion and improvements to the multi-processor package and module docstrings.
........

2012-03-02 11:20:12 Tree
[r15430] by bugman

Updated the 2nd pseudo ellipse test data frame_order.py script for the new parameters.

2012-03-02 09:36:34 Tree
[r15429] by bugman

Removed the unused ave_pos_alpha parameter from the Frame_order.test_cam_iso_cone_free_rotor2 sys test.

2012-03-02 09:34:26 Tree
[r15428] by bugman

Expansion and improvements to the multi-processor package and module docstrings.

2012-03-01 15:32:34 Tree
[r15427] by bugman

The finish button in the new analysis wizard has been renamed to "Start".

This is based on feedback from Hessam Rezaei-Ghaleh.

Now all of the text and icons for the wizards can be programmatically changed.

2012-03-01 15:15:40 Tree
[r15426] by bugman

Bug fix for old numpy versions missing the float16 object.

2012-03-01 14:06:41 Tree
[r15425] by bugman

Epydoc docstring fixes.

2012-03-01 13:28:41 Tree
[r15424] by bugman

Merged revisions 15400-15420 via svnmerge from
svn+ssh://bugman@.../svn/relax/1.3

........
r15400 | bugman | 2012-02-29 10:49:46 +0100 (Wed, 29 Feb 2012) | 5 lines

Eliminated all usage of sys.__stdout__ and sys.__stderr__ in the multi-processor package.

This returns full control of IO streams to the parent program.
........
r15401 | bugman | 2012-02-29 10:54:11 +0100 (Wed, 29 Feb 2012) | 6 lines

Clean up of the processor IO module.

The now unused IO_filter and PrependStringIO classes have been eliminated and the imports all
removed.
........
r15402 | bugman | 2012-02-29 11:03:48 +0100 (Wed, 29 Feb 2012) | 5 lines

Fix for the Immediate_result_queue class.

The __init__() method was broken!
........
r15403 | bugman | 2012-02-29 11:06:20 +0100 (Wed, 29 Feb 2012) | 3 lines

Newline cleanup.
........
r15404 | bugman | 2012-02-29 11:13:17 +0100 (Wed, 29 Feb 2012) | 3 lines

Import fix for the recently deleted IO classes.
........
r15405 | bugman | 2012-02-29 13:31:43 +0100 (Wed, 29 Feb 2012) | 6 lines

The load_multiprocessor() function is no longer a static method of the Processor base class.

This function loads the correct Processor class, so doesn't need to be a method of the base class
and operates cleanly and more clearly as a stand alone function.
........
r15406 | bugman | 2012-02-29 13:37:57 +0100 (Wed, 29 Feb 2012) | 8 lines

Shifted the main interface of the multi-processor package into the __init__ module.

The two main interfaces, or the API with the calling program, are the load_multiprocessor() function
and the Processor_box class. These have been shifted from the multi.processor module as this is
mainly a base class. The import_module() function has also been shifted into __init__ and been made
a private function, removing it from the API.
........
r15407 | bugman | 2012-02-29 13:42:09 +0100 (Wed, 29 Feb 2012) | 6 lines

Removed the RelaxError import.

This was supposed to be only temporary. Now the multi-processor package is 100% independent of
relax again.
........
r15408 | bugman | 2012-02-29 13:54:29 +0100 (Wed, 29 Feb 2012) | 7 lines

Shifted the Application_callback class into the multi.__init__ module to shift it into the API.

The __init__ module now hopefully contains the entirety of the public interface or API of the
multi-processor package. This should simplify the interaction of the package with the importing
program.
........
r15409 | bugman | 2012-02-29 14:42:35 +0100 (Wed, 29 Feb 2012) | 17 lines

Another big redesign of the multi-processor package public API.

The processor module has been split into two. It now only contains the Processor base class.
All other functions and classes have been shifted into the api module. The actual API is in the
__init__ module which, via imports from the api module, presents the following to the user:

- Memo
- Result_command
- Slave_command
- load_multiprocessor
- Application_callback
- Processor_box

The API will still require significant simplification and abstraction to increase the usability
of the multi-processor by the parent program.
........
r15410 | bugman | 2012-02-29 14:43:28 +0100 (Wed, 29 Feb 2012) | 3 lines

Import clean up.
........
r15411 | bugman | 2012-02-29 14:51:27 +0100 (Wed, 29 Feb 2012) | 5 lines

Created a new module for holding miscellaneous functions used throughout the package.

This includes import_module() and raise_unimplemented() functions.
........
r15412 | bugman | 2012-02-29 16:16:59 +0100 (Wed, 29 Feb 2012) | 7 lines

Simplification and abstraction of the Slave_command.run() method.

The Processor now calls the _run() method of the Slave_command base class. This is a wrapper for
run() which performs the exception handling. Therefore the program code, such as
MF_minimise_command, is no longer required to handle the multi-processor specific error handling.
........
r15413 | bugman | 2012-02-29 16:23:41 +0100 (Wed, 29 Feb 2012) | 3 lines

Shifted the Memo object into its own module.
........
r15414 | bugman | 2012-02-29 16:32:05 +0100 (Wed, 29 Feb 2012) | 3 lines

Shifted the Capturing_exception class into the misc module.
........
r15415 | bugman | 2012-03-01 10:10:36 +0100 (Thu, 01 Mar 2012) | 3 lines

Updated the Slave_command.run() docstring to state that this must be overridden.
........
r15416 | bugman | 2012-03-01 10:45:50 +0100 (Thu, 01 Mar 2012) | 21 lines

Reverted r15412 as this was triggering a locking condition and no allowing the program to terminate.

The command used was:
svn merge -r15412:r15411 .

.....
r15412 | bugman | 2012-02-29 16:16:59 +0100 (Wed, 29 Feb 2012) | 7 lines
Changed paths:
M /1.3/multi/api.py
M /1.3/multi/multi_processor_base.py
M /1.3/multi/uni_processor.py
M /1.3/specific_fns/model_free/multi_processor_commands.py

Simplification and abstraction of the Slave_command.run() method.

The Processor now calls the _run() method of the Slave_command base class. This is a wrapper for
run() which performs the exception handling. Therefore the program code, such as
MF_minimise_command, is no longer required to handle the multi-processor specific error handling.
.....
........
r15417 | bugman | 2012-03-01 10:47:34 +0100 (Thu, 01 Mar 2012) | 6 lines

Removed the try statement in the model-free Slave_command.run() method.

After testing, this appears to not be needed anyway. Exception handling is correctly performed on
the slave and master.
........
r15418 | bugman | 2012-03-01 10:57:22 +0100 (Thu, 01 Mar 2012) | 7 lines

Future proofed the relax codebase by replacing all ''' with """ in the docstrings.

According to the Python convention PEP 257 (http://www.python.org/dev/peps/pep-0257/), it is stated
that, "[f]or consistency, always use """triple double quotes""" around docstrings". As this is a
convention, it may be enforced in future Python versions.
........
r15419 | bugman | 2012-03-01 13:36:46 +0100 (Thu, 01 Mar 2012) | 11 lines

Created a special Verbosity singleton for controlling the multi-processor package print outs.

This controls the text prepended to the output from the slaves, as well as all print outs from the
master processor. If the verbosity level is set to 0, then nothing but warnings and errors are
printed. The level of 1 prepends text to the slave IO streams and the print out of from the master
of the active slave set.

This will be important for very fine grained parallelisation whereby the multi-processor is used at
the level of the basic function call or lower.
........
r15420 | bugman | 2012-03-01 14:08:02 +0100 (Thu, 01 Mar 2012) | 5 lines

The relax --debug command line option now turns the multi-processor package verbosity to 1.

The default is currently 0, but the debugging flag changes this.
........

2012-03-01 13:15:17 Tree
[r15423] by bugman

Merged revisions 15400-15420 via svnmerge from
svn+ssh://bugman@.../svn/relax/1.3

........
r15400 | bugman | 2012-02-29 10:49:46 +0100 (Wed, 29 Feb 2012) | 5 lines

Eliminated all usage of sys.__stdout__ and sys.__stderr__ in the multi-processor package.

This returns full control of IO streams to the parent program.
........
r15401 | bugman | 2012-02-29 10:54:11 +0100 (Wed, 29 Feb 2012) | 6 lines

Clean up of the processor IO module.

The now unused IO_filter and PrependStringIO classes have been eliminated and the imports all
removed.
........
r15402 | bugman | 2012-02-29 11:03:48 +0100 (Wed, 29 Feb 2012) | 5 lines

Fix for the Immediate_result_queue class.

The __init__() method was broken!
........
r15403 | bugman | 2012-02-29 11:06:20 +0100 (Wed, 29 Feb 2012) | 3 lines

Newline cleanup.
........
r15404 | bugman | 2012-02-29 11:13:17 +0100 (Wed, 29 Feb 2012) | 3 lines

Import fix for the recently deleted IO classes.
........
r15405 | bugman | 2012-02-29 13:31:43 +0100 (Wed, 29 Feb 2012) | 6 lines

The load_multiprocessor() function is no longer a static method of the Processor base class.

This function loads the correct Processor class, so doesn't need to be a method of the base class
and operates cleanly and more clearly as a stand alone function.
........
r15406 | bugman | 2012-02-29 13:37:57 +0100 (Wed, 29 Feb 2012) | 8 lines

Shifted the main interface of the multi-processor package into the __init__ module.

The two main interfaces, or the API with the calling program, are the load_multiprocessor() function
and the Processor_box class. These have been shifted from the multi.processor module as this is
mainly a base class. The import_module() function has also been shifted into __init__ and been made
a private function, removing it from the API.
........
r15407 | bugman | 2012-02-29 13:42:09 +0100 (Wed, 29 Feb 2012) | 6 lines

Removed the RelaxError import.

This was supposed to be only temporary. Now the multi-processor package is 100% independent of
relax again.
........
r15408 | bugman | 2012-02-29 13:54:29 +0100 (Wed, 29 Feb 2012) | 7 lines

Shifted the Application_callback class into the multi.__init__ module to shift it into the API.

The __init__ module now hopefully contains the entirety of the public interface or API of the
multi-processor package. This should simplify the interaction of the package with the importing
program.
........
r15409 | bugman | 2012-02-29 14:42:35 +0100 (Wed, 29 Feb 2012) | 17 lines

Another big redesign of the multi-processor package public API.

The processor module has been split into two. It now only contains the Processor base class.
All other functions and classes have been shifted into the api module. The actual API is in the
__init__ module which, via imports from the api module, presents the following to the user:

- Memo
- Result_command
- Slave_command
- load_multiprocessor
- Application_callback
- Processor_box

The API will still require significant simplification and abstraction to increase the usability
of the multi-processor by the parent program.
........
r15410 | bugman | 2012-02-29 14:43:28 +0100 (Wed, 29 Feb 2012) | 3 lines

Import clean up.
........
r15411 | bugman | 2012-02-29 14:51:27 +0100 (Wed, 29 Feb 2012) | 5 lines

Created a new module for holding miscellaneous functions used throughout the package.

This includes import_module() and raise_unimplemented() functions.
........
r15412 | bugman | 2012-02-29 16:16:59 +0100 (Wed, 29 Feb 2012) | 7 lines

Simplification and abstraction of the Slave_command.run() method.

The Processor now calls the _run() method of the Slave_command base class. This is a wrapper for
run() which performs the exception handling. Therefore the program code, such as
MF_minimise_command, is no longer required to handle the multi-processor specific error handling.
........
r15413 | bugman | 2012-02-29 16:23:41 +0100 (Wed, 29 Feb 2012) | 3 lines

Shifted the Memo object into its own module.
........
r15414 | bugman | 2012-02-29 16:32:05 +0100 (Wed, 29 Feb 2012) | 3 lines

Shifted the Capturing_exception class into the misc module.
........
r15415 | bugman | 2012-03-01 10:10:36 +0100 (Thu, 01 Mar 2012) | 3 lines

Updated the Slave_command.run() docstring to state that this must be overridden.
........
r15416 | bugman | 2012-03-01 10:45:50 +0100 (Thu, 01 Mar 2012) | 21 lines

Reverted r15412 as this was triggering a locking condition and no allowing the program to terminate.

The command used was:
svn merge -r15412:r15411 .

.....
r15412 | bugman | 2012-02-29 16:16:59 +0100 (Wed, 29 Feb 2012) | 7 lines
Changed paths:
M /1.3/multi/api.py
M /1.3/multi/multi_processor_base.py
M /1.3/multi/uni_processor.py
M /1.3/specific_fns/model_free/multi_processor_commands.py

Simplification and abstraction of the Slave_command.run() method.

The Processor now calls the _run() method of the Slave_command base class. This is a wrapper for
run() which performs the exception handling. Therefore the program code, such as
MF_minimise_command, is no longer required to handle the multi-processor specific error handling.
.....
........
r15417 | bugman | 2012-03-01 10:47:34 +0100 (Thu, 01 Mar 2012) | 6 lines

Removed the try statement in the model-free Slave_command.run() method.

After testing, this appears to not be needed anyway. Exception handling is correctly performed on
the slave and master.
........
r15418 | bugman | 2012-03-01 10:57:22 +0100 (Thu, 01 Mar 2012) | 7 lines

Future proofed the relax codebase by replacing all ''' with """ in the docstrings.

According to the Python convention PEP 257 (http://www.python.org/dev/peps/pep-0257/), it is stated
that, "[f]or consistency, always use """triple double quotes""" around docstrings". As this is a
convention, it may be enforced in future Python versions.
........
r15419 | bugman | 2012-03-01 13:36:46 +0100 (Thu, 01 Mar 2012) | 11 lines

Created a special Verbosity singleton for controlling the multi-processor package print outs.

This controls the text prepended to the output from the slaves, as well as all print outs from the
master processor. If the verbosity level is set to 0, then nothing but warnings and errors are
printed. The level of 1 prepends text to the slave IO streams and the print out of from the master
of the active slave set.

This will be important for very fine grained parallelisation whereby the multi-processor is used at
the level of the basic function call or lower.
........
r15420 | bugman | 2012-03-01 14:08:02 +0100 (Thu, 01 Mar 2012) | 5 lines

The relax --debug command line option now turns the multi-processor package verbosity to 1.

The default is currently 0, but the debugging flag changes this.
........

2012-03-01 13:14:35 Tree
[r15422] by bugman

Merged revisions 15400-15420 via svnmerge from
svn+ssh://bugman@.../svn/relax/1.3

........
r15400 | bugman | 2012-02-29 10:49:46 +0100 (Wed, 29 Feb 2012) | 5 lines

Eliminated all usage of sys.__stdout__ and sys.__stderr__ in the multi-processor package.

This returns full control of IO streams to the parent program.
........
r15401 | bugman | 2012-02-29 10:54:11 +0100 (Wed, 29 Feb 2012) | 6 lines

Clean up of the processor IO module.

The now unused IO_filter and PrependStringIO classes have been eliminated and the imports all
removed.
........
r15402 | bugman | 2012-02-29 11:03:48 +0100 (Wed, 29 Feb 2012) | 5 lines

Fix for the Immediate_result_queue class.

The __init__() method was broken!
........
r15403 | bugman | 2012-02-29 11:06:20 +0100 (Wed, 29 Feb 2012) | 3 lines

Newline cleanup.
........
r15404 | bugman | 2012-02-29 11:13:17 +0100 (Wed, 29 Feb 2012) | 3 lines

Import fix for the recently deleted IO classes.
........
r15405 | bugman | 2012-02-29 13:31:43 +0100 (Wed, 29 Feb 2012) | 6 lines

The load_multiprocessor() function is no longer a static method of the Processor base class.

This function loads the correct Processor class, so doesn't need to be a method of the base class
and operates cleanly and more clearly as a stand alone function.
........
r15406 | bugman | 2012-02-29 13:37:57 +0100 (Wed, 29 Feb 2012) | 8 lines

Shifted the main interface of the multi-processor package into the __init__ module.

The two main interfaces, or the API with the calling program, are the load_multiprocessor() function
and the Processor_box class. These have been shifted from the multi.processor module as this is
mainly a base class. The import_module() function has also been shifted into __init__ and been made
a private function, removing it from the API.
........
r15407 | bugman | 2012-02-29 13:42:09 +0100 (Wed, 29 Feb 2012) | 6 lines

Removed the RelaxError import.

This was supposed to be only temporary. Now the multi-processor package is 100% independent of
relax again.
........
r15408 | bugman | 2012-02-29 13:54:29 +0100 (Wed, 29 Feb 2012) | 7 lines

Shifted the Application_callback class into the multi.__init__ module to shift it into the API.

The __init__ module now hopefully contains the entirety of the public interface or API of the
multi-processor package. This should simplify the interaction of the package with the importing
program.
........
r15409 | bugman | 2012-02-29 14:42:35 +0100 (Wed, 29 Feb 2012) | 17 lines

Another big redesign of the multi-processor package public API.

The processor module has been split into two. It now only contains the Processor base class.
All other functions and classes have been shifted into the api module. The actual API is in the
__init__ module which, via imports from the api module, presents the following to the user:

- Memo
- Result_command
- Slave_command
- load_multiprocessor
- Application_callback
- Processor_box

The API will still require significant simplification and abstraction to increase the usability
of the multi-processor by the parent program.
........
r15410 | bugman | 2012-02-29 14:43:28 +0100 (Wed, 29 Feb 2012) | 3 lines

Import clean up.
........
r15411 | bugman | 2012-02-29 14:51:27 +0100 (Wed, 29 Feb 2012) | 5 lines

Created a new module for holding miscellaneous functions used throughout the package.

This includes import_module() and raise_unimplemented() functions.
........
r15412 | bugman | 2012-02-29 16:16:59 +0100 (Wed, 29 Feb 2012) | 7 lines

Simplification and abstraction of the Slave_command.run() method.

The Processor now calls the _run() method of the Slave_command base class. This is a wrapper for
run() which performs the exception handling. Therefore the program code, such as
MF_minimise_command, is no longer required to handle the multi-processor specific error handling.
........
r15413 | bugman | 2012-02-29 16:23:41 +0100 (Wed, 29 Feb 2012) | 3 lines

Shifted the Memo object into its own module.
........
r15414 | bugman | 2012-02-29 16:32:05 +0100 (Wed, 29 Feb 2012) | 3 lines

Shifted the Capturing_exception class into the misc module.
........
r15415 | bugman | 2012-03-01 10:10:36 +0100 (Thu, 01 Mar 2012) | 3 lines

Updated the Slave_command.run() docstring to state that this must be overridden.
........
r15416 | bugman | 2012-03-01 10:45:50 +0100 (Thu, 01 Mar 2012) | 21 lines

Reverted r15412 as this was triggering a locking condition and no allowing the program to terminate.

The command used was:
svn merge -r15412:r15411 .

.....
r15412 | bugman | 2012-02-29 16:16:59 +0100 (Wed, 29 Feb 2012) | 7 lines
Changed paths:
M /1.3/multi/api.py
M /1.3/multi/multi_processor_base.py
M /1.3/multi/uni_processor.py
M /1.3/specific_fns/model_free/multi_processor_commands.py

Simplification and abstraction of the Slave_command.run() method.

The Processor now calls the _run() method of the Slave_command base class. This is a wrapper for
run() which performs the exception handling. Therefore the program code, such as
MF_minimise_command, is no longer required to handle the multi-processor specific error handling.
.....
........
r15417 | bugman | 2012-03-01 10:47:34 +0100 (Thu, 01 Mar 2012) | 6 lines

Removed the try statement in the model-free Slave_command.run() method.

After testing, this appears to not be needed anyway. Exception handling is correctly performed on
the slave and master.
........
r15418 | bugman | 2012-03-01 10:57:22 +0100 (Thu, 01 Mar 2012) | 7 lines

Future proofed the relax codebase by replacing all ''' with """ in the docstrings.

According to the Python convention PEP 257 (http://www.python.org/dev/peps/pep-0257/), it is stated
that, "[f]or consistency, always use """triple double quotes""" around docstrings". As this is a
convention, it may be enforced in future Python versions.
........
r15419 | bugman | 2012-03-01 13:36:46 +0100 (Thu, 01 Mar 2012) | 11 lines

Created a special Verbosity singleton for controlling the multi-processor package print outs.

This controls the text prepended to the output from the slaves, as well as all print outs from the
master processor. If the verbosity level is set to 0, then nothing but warnings and errors are
printed. The level of 1 prepends text to the slave IO streams and the print out of from the master
of the active slave set.

This will be important for very fine grained parallelisation whereby the multi-processor is used at
the level of the basic function call or lower.
........
r15420 | bugman | 2012-03-01 14:08:02 +0100 (Thu, 01 Mar 2012) | 5 lines

The relax --debug command line option now turns the multi-processor package verbosity to 1.

The default is currently 0, but the debugging flag changes this.
........

2012-03-01 13:11:28 Tree
Older >
Want the latest updates on software, tech news, and AI?
Get latest updates about software, tech news, and AI from SourceForge directly in your inbox once a month.