Grassmanizing and/or groups, continued

[funderburkjim]

In #64, @Andhrabharati requests python code to generate

• from csl-orig/v02/mw/mw.txt (the base cdsl form of MW digitization)
• to an 'AB3' form which may facilitate comparisons between his versions and cdsl.

This issue addresses that request. Time will tell whether the solution provided is useful. It may take me a day or two to get a first approximation.

[funderburkjim]

First try:

Files are in issue181 directory. To get an 'AB' version (which I'll call ab2a for since one program is called convert_ab2a.py) from a cdsl version, here's what to do:

• download the program components in some local directory
  • convert_ab1.py, convert_ab2a.py, digentry.py
• get a copy of cdsl version of mw.txt (such as from csl-orig/v02/mw/mw.txt)
  • following steps assume this is called temp_mw_1.txt and is in same directory, but you can change file names as convenient
• python convert_ab1.py cdsl,ab temp_mw_1.txt temp_mw_1_ab1.txt
  • temp_mw_1_ab1.txt is an intermediate form
• optionally check the previous step is invertible.
  • python convert_ab1.py ab,cdsl temp_mw_1_ab1.txt temp_mw_1_chk.txt
  • diff temp_mw_1.txt temp_mw_1_chk.txt | wc -l
    • should output '0' if inversion works
  • rm temp_mw_1_chk.txt # no further need of this file
• python convert_ab2a.py cdsl,ab temp_mw_1_ab1.txt temp_mw_1_ab2a.txt
  • temp_mw_1_ab2a.txt is the result - the pseudo-AB version

[Andhrabharati]

Thanks Jim, for giving this python script (I didn't notice that you had made this available earlier at issue 176 itself).

It sure would help me correlate quickly with current CDSL file (I've been using the August 2024 file so far for correlation; and as there have been some major changes in the cdsl file, I thought I should again "match" the two versions).

Now I have the current cdsl file, duly converted to iast.