[Q-e-developers] Error parsing output from pwscf

[David C. Lonie]

From: Paolo Giannozzi <giannozz at democritos.it>
Subject: Re: [Q-e-developers] Error parsing output from pwscf
Date: Sun, 15 Aug 2010 15:44:51 +0200

> Hi
> 
> sorry for the very long delay in answering: your message was waiting
> for
> approval, unbeknownst to everybody, due to problems with this mailing
> list. I just modified the output format (from f10.6 to f11.6) so that
> in the
> unfortunate case of long axis with negative sign, there is still a
> space
> between the components.

Thanks!
 
> I am happy to hear that somebody is maintaining a PWscf output reader.
> Note however that the format of output files is subject to quite
> frequent
> changes, often depending on the mood of some developer. In the future,
> hopefully by the end of the year, it is planned to have a XML output
> file,
> with stable and documented syntax, in addition to the human-readable
> one. This will be much easier and safer to read.

That sounds great -- I look forward to seeing this addition.

Thanks again,

Dave

> Best regards
> 
> Paolo
> 
>> I'm the maintainer for the PWSCF reader for OpenBabel, and I've run
>> across a problem in certain pwscf outputs.
>>
>> When parsing the crystal axes block, it is possible for the values to
>> run into each other, e.g.
>>
>>      crystal axes: (cart. coord. in units of a_0)
>>                a(1) = ( 12.349300  0.000000  0.000000 )
>>                a(2) = ( -1.547240  3.258450  0.000000 )
>>                a(3) = (-10.238900-20.488400 16.770400 )
>>
>> The a(3) vector's first two elements are read as a single token during
>> parsing. Of course, this is fine for human-readability, but makes
>> parsing much more difficult. Are there any objections to changing this
>> in future releases?
> 
> ---
> Paolo Giannozzi, Dept of Physics, University of Udine
> via delle Scienze 208, 33100 Udine, Italy
> Phone +39-0432-558216, fax +39-0432-558222
> 
> 
>