[Ifeffit] the development branch of Artemis

Bruce Ravel ravel at phys.washington.edu
Mon Aug 30 14:34:24 CDT 2004


As I mentioned in my last email, today's source code release includes
the new development branch of Artemis.  As I said about a month ago, I
am now maintaining a stable branch, which will usually only get bug
fixes, and a development branch, which is where I will be working on
new features.

I put a few notes and a screenshot on my Artemis page:
scroll down to the middle of the page.  There is a box labeled
"Development Branch".

I want to clarify how this is going to work:

1.  The source tarball has both branches and both get built and
    installed.  The development version gets installed as "diana".
    This allows you to try out the development version but still have
    the stable version to fall back on.

2.  I don't intend to build Windows executables of the development
    version until it is almost ready for normal release.  This is in
    part to save myself some work and in part to keep the development
    branch out of wide use until it has been much better tested.

3.  The development version involves some changes to the project file
    format.  It will read projects from the stable version, but some
    information will get lost in the other direction.

4.  If you use the development version, you should anticipate that
    there will be problems and be prepared to submit good, clear,
    helpful bug reports.  If you expect stable behavior, use the
    stable version.  That said, I have been using the development
    version for my own analysis.  It ain't perfect, but I am getting
    work done.

Here are a few notes to get you started with the new features.  You
might want to go take a look at the screenshot on the Artemis web

1.  The window is much larger now.  This takes up more screen real
    estate, but the paths list is much taller.  The blank area in
    the lower right hand corner will get filled with new plotting
    controls in the future.

2.  Fits get added to the Data and Paths List as seen in the screen
    shot.  Old fits are plotted just like anything else -- just select
    it and click a plot button.

3.  Background functions and residuals can still be plotted, but they
    are no longer displayed in the Data and Paths List.  Now to get a
    bkg or res plot, click the check buttons in the plot controls
    panel.  Whenever you plot a fit, the background and residual
    functions from that fit will be plotted along with best fit

4.  When you click on a fit entry, the main panel will display a tool
    for viewing and exploring the log files from the fits you have
    done.  This is essentially the same as the log viewer that Stefano
    was asking about last week.  There is a list of fits on the left
    side of the panel -- double-click and right-click are both
    useful.  The main use of this page is to write little reports
    about parameters so you can compare your various fits.  Fits are
    included into reports by selecting them from the list on the left
    and not by selecting them from the Data and Path List.

5.  In many situations (plotting, saving files, etc) selecting the
    head of the fit branch means to use the most recent fit.
    Selecting both the head of the fit branch and the most recent fit
    is redundant, but should not cause any problems.

6.  The fit entries in the Data and Paths List may behave a little
    surprisingly in the case of a multiple data set fit.  Some things,
    like plotting, depend upon which data set the selected list item
    is under.  Other things, like discarding a fit, will effect every
    data set.

7.  In the upper right corner is a little box which tells you when you
    have modified your project and, therefore, might want to consider
    saving it.  Keep an eye on that box and let me know if there are
    situations where it gets set incorrectly.

8.  I have written some documentation for viewing inside Artemis.
    There are lots of buttons scattered around the program for
    displaying context appropriate documentation (you can see a couple
    in the screenshot).  Feel free to read through this and suggest
    ways to improve upon it or even write some of it yourself!  Lots
    of the documentation is still unwritten.

Things that don't work correctly yet that I know about:

1.  Saving fits to multi-column files certainly doesn't work.  Saving
    individual fits to column data files should work.
2.  Hiding individual or selected fits (see the Fit menu) is flaky.

Things to remember about using the development branch:

1.  You can always go back to the stable branch, they are installed
2.  It's a development branch -- it's going to have lots of bugs.
3.  I want to hear about your problems and your suggestions for new &
    better features.

Have fun!

 Bruce Ravel  ----------------------------------- ravel at phys.washington.edu
 Code 6134, Building 3, Room 405
 Naval Research Laboratory                          phone: (1) 202 767 2268
 Washington DC 20375, USA                             fax: (1) 202 767 4642

 NRL Synchrotron Radiation Consortium (NRL-SRC)
 Beamlines X11a, X11b, X23b
 National Synchrotron Light Source
 Brookhaven National Laboratory, Upton, NY 11973

 My homepage:    http://feff.phys.washington.edu/~ravel 
 EXAFS software: http://feff.phys.washington.edu/~ravel/software/exafs/

More information about the Ifeffit mailing list