Hi Michael,

To use the functional code, you may need to jump back a few revisions.
 In your local copy, just type:

$ svn up -r10525

This will get you to a stable spot, before I broke everything spinning
the Rx code out of relax_gui.  You'll also see with this last commit
the full resolution relax logo, which may come in handy later on.

Regards,

Edward


On 28 January 2010 16:37, Edward d'Auvergne <[email protected]> wrote:
> Hi Michael,
>
> After all the huge infrastructural changes in tightly integrating the
> GUI into relax (most of which are not visible), I finally have the
> automatic model-free protocol running!  As this been a massive set of
> changes, I'll recap these as a summary:
>
> * Splitting of 'relax_gui' into many modules.
> * A clean system for image and icon usage in the program with the
> 'path' module (and the addition of the oxygen icons).
> * Shifted the building of the main GUI elements (main window, menu
> bar, automatic model-free analysis frame) into self.build_*
> * Spun out all of the automatic model-free code into its own module.
> * GUI package file layout improvements.
> * Alphabetical ordering of all functions, methods, and classes.
> * Whitespace rule violations have been fixed.
> * Class and function naming rule violations have been fixed.
> * Shifted to using a single code base across relax (GUI and
> full_analysis.py) for the automatic model-free protocol by using the
> new automatic_analysis.dauvergne_protocol module.
> * Conversion of module, class, function, and method comments into docstrings.
> * Isolated many GUI elements into their own methods (aggregating
> creation, property setting, layout, and event binding all in one
> place).  This allows for most variables to be local and hidden from
> the rest of the GUI, and for easy large scale GUI rearrangements.
> * Elimination of most global variables.  All have to go!
> * Creation of the ds.relax_gui object for storing all GUI information
> within the relax data store.
> * Creation of the ds.relax_gui.analyses list object for storing all
> automatic analysis specific data.
> * The relaxGUI save file has been replaced by the relax save state -
> all data is now in the relax data store.
> * A set of sync_ds() methods have been added to the main class and the
> analysis classes to synchronise the data between the GUI and relax
> data store.
> * Added epydoc documentation to the docstrings.
> * The file dialogs change the current working directory (cwd) using
> chdir, so that the next dialog will open in the same place (not
> complete).
> * Object (variable, function, class, etc.) renaming to better reflect
> what they do.
> * self is never passed into functions.
> * The self.analysis_frames list has been added to the main class to
> allow dynamic creation and deletion of these GUI elements (as well as
> easy access to the data and methods of the class).
> * Variable name simplification and shifting into the local method name
> space of that GUI element.
> * Made the controller window fully independent (can be opened and
> closed at any time, and the GUI does not freeze during execution).
>
>
> I hope you were able to keep up ;)  The rest of the code base must
> follow these concepts.  The following still needs to be done:
>
> * Repair all code where FIXME is written.
> * Spin out the R1, R2, and NOE code (I will work on this to give you
> the infrastructure to advance the code).
> * Eliminate the choice between AIC and BIC in the automatic model-free
> protocol - BIC can never be used for this specific protocol.
> * Variable name simplifications.
> * Further isolation of each GUI element into methods or classes (and
> the elimination of all __set_properties() and __do_layout() methods).
> * Controller window improvements (too much memory used by the STDOUT
> messages, thread issues, IO redirection issues, etc.).
>
>
> I'll try to spin out the R1, R2, and NOE code from relax_gui.  I'll
> leave all other code unmodified so that you can start to dig into it.
>
> Regards,
>
> Edward
>

_______________________________________________
relax (http://nmr-relax.com)

This is the relax-devel mailing list
[email protected]

To unsubscribe from this list, get a password
reminder, or change your subscription options,
visit the list information page at
https://mail.gna.org/listinfo/relax-devel

Reply via email to