Current horae tarball: 067

Bugs

Athena

  1. Sort files when importing (silly Windows!)

  2. Problems reading CL normed from a project

  3. LCF display problem switching from combo. to marked. For each marked group, need to completely initialize the parameters and display. <!>

  4. Point finder find x doesn't work
  5. UI problem fixing edge step. The order of clicking the fix step button and editing the edge step seems to matter
  6. Difference doesn't get plotted in peak fitting
Artemis

  1. Quirky Atoms problem: http://millenia.cars.aps.anl.gov/pipermail/ifeffit/2008-September/003805.html

  2. Palettes frame is messed up, esp. on Windows

  3. New Atoms page -> dialog with "Import atoms.inp" -- button does not expand properly on windows

  4. Many problems have been seen when fits are discarded from a project file. These problems include:
    • there is a scenario in which fits become unplottable without becoming unrecoverable
    • try to plot one of the remaining fits (specifically discard selected with one remaining fit, then try to plot it) and the data cannot be read. Work-around: save and reread the project.
    • discarded fits may still show up in the log viewer list under some situation (I'm having trouble reproducing this one!)
    • a discarded fit might have all its files deleted but its folder not deleted. this leads to problems importing the project file (see project from Victor)
    • older retained fits might not be displayed when the project is imported (see project from Victor)
    • Grab all best fit values behaves poorly before a fit has been done on a project (open project with a history of fits, do it, empty parens. Should use most recent fit.

  5. Make sure all text uses defined fonts

  6. Make sure all the views expand correctly to full height of the window
  7. From Grazia Proietti {{{el problema es que en la nueva version si usas la opcion extended

parameters adonde no se ponga nada el escribe unos 0 en lugar que el empty (" ") y esto le fastidiaba ya el file del apj para siempre.}}}

Windows
Athena document
  1. need a section in 4.1 on CL normalization
Hephaestus
  1. should ion chamber utility use mass energy-absorption coefficients?
  2. incorporate Peter Revesz's flux calculator into Hephaestus' ion chamber utility.

  3. Formula for Ethyl alcohol is incorrect, should be !C2H5OH

Other
  1. Check that document reading is not suffering from the icon bug on Windows

Small suggestions

horae
Athena
  1. Explicitly display energy shift on calibration page
  2. LCF in chi(q)
  3. Ability to save last plot to a column data file

  4. Use BrowseEntry rather than cascaded menus for bkg_z selection (this requires proper validation of the entry when edited directly. That is very difficult to do consistently and reliably)

  5. From Marco Vaccari:

    1. possibility to set the zero-th order cumulant to zero. This is very often the case, for example, of a temperature series where the standard and the unknown system have the same coordination number. The remaining contribution to c_0 is negligible. (Bruce sez: switches for turning off 0, 3, 4)

    2. possibility to set the fourth order cumulant to zero. This is useful especially for low temperatures and for systems characterised by a not too high anharmonicity.

    3. plot the log-ratio of amplitudes and phase differences as a function of k^2 (instead of k). This helps since in that case c_1 would be the intercept and c_3 the slope of the phase differences curve, and c_2 would be the slope of log-ratio amplitudes curves. (Bruce sez: switch for toggling between k and k2)

    4. plot the log-ratio / phase-difference experimental curves using a user-defined q-range of visualization (different from the fitting q-range). This helps in understanding the meaninful fitting range which is, usually, not too extended. (Bruce sez: use plot features q range)

    5. Bruce: Separate buttons for results and log file not needed -- put log info in header

Artemis
  1. Import feff paths using the various criteria (amp, R, nleg, etc)

Long term items

Athena

  1. Normalization options on the horizon: functional normalization, normalization order, Victoreen. Some of the controls exist in the "additional parameter" page of the background parameters section, but they are currently disabled. Ifeffit currently ignores norm_order in spline() and seems to mishandle fnorm as well. At this moment, Matt has not divulged how the Victoreen stuff will work.

  2. Improve peak fitting dialog (implement some of the bells and whistles from the LCF dialog):
    • Implement marked group fitting and a spreadsheet-able report
    • Adjust the data e0
    • toggle plotting of step component with "plot components"
    • pseudo-Voight
    • CL instead of arctan or erf
    • what's the status of the minimize bug in Ifeffit that affected fitting e0's?
  3. Principle Component Analysis on the set of marked groups
  4. Spectral deconvolution
  5. More file type plugins:
    • multi-record Spec files (needs a good OO interface)
    • deadtime corrections
    • SSRL binary data
    • NSLS quick scan data
  6. Data pages -- a way of organizing data in a project by having more than one groups list in a project

  7. Import contents of chi_data folder in an Artemis project into Athena

  8. Integration with some kind of a web-based, standard materials database

  9. R and q records

Artemis

  1. Conversion of PDB (and others) to feff.inp and vice versa. The OpenBabel Project would help.

  2. Convert a data file into a feffNNNN.dat style pseudo-path by Fourier filtering. This may be useful for situations involving hydrogen.

  3. Intrp page: sort by columns
  4. Merge projects
    • page for selecting components (i.e. data and feff calculations) from a project and merging them into the current project

    • all parameters are imported from the merged project and added to GDS list as specially tagged parameters that need resolution, i.e. deletion or definition as some other parameter type

    • on GDS page

      • m: bright red, white background

      • gds2_define and gds2_alter need to allow repeated names for merge type

      • do not allow any summations or fits if merged parameters are not resolved
  5. Use the history mechanism in creative ways:

    1. Parameter mapping, as in, for example, Biochemistry 35 (1996) pp. 9014-9023

    2. Batch processing, i.e. a meachnism to run fits on a series of data files using the same model.

  6. Color coding of reserved words and function names in path parameter math expressions. (This would be a very nice featue, but would require creation of new Tk widget type implementing tags in an entry widget and would require a validation callback to parse the contects of the entry widget. Whew!)

  7. An ability to abort a fit.

  8. Import feffit.inp files correctly, including recursing into include files.

  9. Multiple component (e.g. magnitude+real) plots in R and q, as in Athena
  10. Arrows, text, shapes on plot.
  11. Fix/improve the space group browser

  12. Implement bkg-tweaking, the point of which is to facilitate Shelly's iterative approach to difficult background removal problems. See this PDF file or this post to the mailing list.

Hephaestus
  1. Mirrors
  2. Deal correctly with mass-absorption at high energy
  3. Auger/fluorescence branching rations

Other concepts

  1. Savitsky-Golay filter in smoothing dialog: <main Matlab script> <main Matlab script, filter>

  2. SSRL binary plugin: <Matlab read script> <Matlab write script>

  3. Plugin for NSLS X18b quick scan data: <example file>

  4. Incorporate Peter Revesz's suggestions for getting high energies right in Hephaestus <html and js files from Peter>

BruceRavel/Horae/HoraeToDoList (last edited 2011-06-05 16:25:03 by BruceRavel)