KING Andrew
À :
 FARHI Emmanuel 
 
mercredi 10 juin 2020 12:59
Hi Emmanuel,

  In ruche/share-temp/PSICHE/Python is the tomodata code, and in ruche/share-temp/PSICHE/data is the marta dataset.

  I will try and get a rough documentation finished for next week.  I have various bits already done. 

  The concept is that:

(1) PyHST requires data and a par file.
(2) Tomodata is just a top level function (i.e. doing the job of a GUI), that calls other python functions. 
(3) These functions act on the data or on the pyhst par file.
(4) There are a couple of special functions that convert data from nxs (or tif, or whatever) into the edf format used by pyhst.

A directory structure is assumed:

the workingdirectory is where you call tomodata
inside this, each tomogram has a folder with the name of the scan
inside that folder is the parfile, and a folder with the edf projections
reconstructions are written into the scan directory

workingdirectory/
workingdirectory/scanname/
workingdirectory/scanname/scanname_edfs/

Somewhere there is a spool directory, where the raw data (nxs files + par file) live.

spooldir/scanname/scanname.nxs
spooldir/scanname/scanname.par
spooldir/scanname/pre_ref.nxs
spooldir/scanname/post_ref.nxs
spooldir/scanname/post_dark.nxs

Cheers, Andy