[Pw_forum] exctracting data from evc.dat

Andrea Ferretti ferretti at mit.edu
Sat Jun 25 14:09:20 CEST 2011



Hi Aris,

once you've managed to get the files evc.dat, I think the easiest way to read 
them (together with more or less all the data produced and dumped by QE) is to 
use the QEXML library.
This lib (you can find in PP/qexml.f90 ) is meant to be used also by third 
party codes and can be compiled easily also out of QE and in connection with 
other codes...
Note you need to link the iotk lib, but that is compiled by espresso
(have a look at make.sys to see how to link it)

you may want to have a look at the routines:

qexml_init,  qexml_openfile, qexml_closefile

qexml_read_planewaves
qexml_read_gk
qexml_read_wfc

(feel free to have alook inside the library)

The idea is that these routines are written to first read the dimensions of the 
data you need, then you can allocate the needed workspace, and then you can 
call the routine again to get the massive data...

For some examples about how to use them, you can have a look at

PP/qexml_example.f90

or inside the sources of WanT (downloadable at qe-forge), just grep "qexml_" in
want/wannier/*.f90


hope it helps
andrea


>
>   I'm trying to read the values of the kohn-sham wavefunction even for points
> different from the gamma one (otherwise I used pp.x) . I managed to obtain a
> file "evc.dat" that should contain the information, but it is unformatted and 
> I
> don't know it's structure (in that case I could use a read statement from
> fortran).  How can I extract and understand my data?
> 
> Sorry for the questions of a beginner, but I got stuck.
>
>                      Regards
> 
> Marcolongo Aris
> 
> Scuola Internazionale Superiore di Studi Avanzati, Trieste
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>

-- 
Andrea Ferretti
S3 Center, Istituto Nanoscienze, CNR
via Campi 213/A, 41100, Modena, Italy
Tel: +39 059 2055300;  Skype: andrea_ferretti
URL: http://www.nano.cnr.it



More information about the Pw_forum mailing list