[ase-users] Problems reading XCrysDen-files in ASE-gui

[Jakob Blomqvist]

Well it seems like it's working with my file at least! Thank you!

Jakob Blomquist
Associate Professor
Dep. of Material Science
IMP, School of Technology
Malmo University
SWEDEN
+46(0)40 6657751
jakob.blomqvist at mah.se

On 01/16/2013 11:59 AM, Jens Jørgen Mortensen wrote:
> Den 15-01-2013 16:21, Jakob Blomqvist skrev:
>> Dear all,
>>
>> I will take this to the user-list too.
>>
>> I've tried to use ag to read xsf-files converted from Quantum 
>> Espresso output files using QE's own xsf-converter script.
>> There seem to be several problem illustrated with the attached 
>> xsf-file QE_Au.xsf.
>> The first problem is that io/xsf.py doesn't allow for symbols in it 
>> but only atom number. xsf-file format, however, allows for both. 
>> Another problem is that xsf.py requires that the file contains 
>> BEGIN_BLOCK_DATAGRID_3D through the:
>> "assert 'BEGIN_BLOCK_DATAGRID_3D' in line"
>> in line #159 in ase/io/xsf.py
>> but xsf-file format doesn't require it.
>>
>> I suggest to add some alternative section that allows for nothing 
>> after the atomic positions section.
>>
>> Finally, If forces are included as 3 floating-point values directly 
>> after the atomic positions then running ag file.xsf results in error:
>>
>>    File "/home/jakob/ase/ase/gui/ag.py", line 131, in main
>>     run(opt, args)
>>   File "/home/jakob/ase/ase/gui/ag.py", line 91, in run
>>     images.read(args, string2index(opt.image_number))
>>   File "/home/jakob/ase/ase/gui/images.py", line 164, in read
>>     self.initialize(images, names)
>>   File "/home/jakob/ase/ase/gui/images.py", line 71, in initialize
>>     self.F[i] = atoms.get_forces(apply_constraint=False)
>> ValueError: shape mismatch: objects cannot be broadcast to a single 
>> shape
>>
>> which seem to come from problems with lines #152-154 in io/xsf.py
>> ###
>>         if forces is not None:
>>             image.set_calculator(SinglePointCalculator(None, forces, 
>> None,
>>                                                        None, image))
>>         images.append(image)
>> ###
>> As they disappear if I remove these.
>
> Thanks for reporting these bugs.  I think I have fixed them. Could you 
> try again?
>
> Jens Jørgen
>
>> /best
>>
>>
>>
>> _______________________________________________
>> ase-users mailing list
>> ase-users at listserv.fysik.dtu.dk
>> https://listserv.fysik.dtu.dk/mailman/listinfo/ase-users
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.fysik.dtu.dk/pipermail/ase-users/attachments/20130116/442b0c42/attachment.html>