some functions do not work in Larch 0.9.70

[zhangnaichi2020]

I'm a beginner on how to use larch code , and I want to run larch with python in jupyter, so I import the functions that will be used in a way like
"
import larch
from larch import Group
from larch.builtins import _journal
from larch.io import read_ascii
from larch.xafs import sort_xafs, pre_edge
from larch.math.utils import boxcar, savitzky_golay
from larch.xafs.deconvolve import xas_convolve, xas_deconvolve
"
Here is my questions:

1. after update the larch version from 0.9.68 (maybe) to the latest 0.9.70, it takes much longer to run the pre_edge function;
2. for xas_deconvolve and xas_convolve
   1. xas_convolve runs much much slower than xas_deconvolve
   2. for both function, if i run
      "
      dat = read_ascii('./Cr_standards/CrOH3_Cr_K.xdi')
      xas_deconvolve(dat, esigma=1.0) # or xas_convolve(dat, esigma=1.0)
      "
      it returns "Warning: xas_deconvolve: needs First Argument Group or valid arguments for energy, norm", and it goes well when i run
      "xas_deconvolve(dat.energy, dat.mu, esigma=1.0)"
      it seems does not follow the First Argument Group convention;
   3. when i run
      "
      dat = read_ascii('./Cr_standards/CrOH3_Cr_K.xdi')
      xas_deconvolve(dat, esigma=1.0)
      dat.deconv
      "
      it returns "AttributeError: 'Group' object has no attribute 'deconv'", which is not conform with the samples given in 13.2.7.3.

[newville]

@zhangnaichi2020 I think this because xas_deconvolve expects to work on normalized XAFS data, that is after running pre_edge().

 dat = read_ascii('./Cr_standards/CrOH3_Cr_K.xdi')
pre_edge(dat)
xas_deconvolve(dat, esigma=1.0) 

because otherwise it is just not clear what to do the deconvolution on

And, yes, we did see that pre_edge() could get very slow for some input data (especially, with repeated energy values) with changes in 0.9.70. That should be fixed in the master branch, which you can install with

  python -m pip install https://millenia.cars.aps.anl.gov/xraylarch/downloads/xraylarch-latest-py3-none-any.whl