Re: [cclib-devel] [cclib-users] g03 oniom

[Noel O'B. <bao...@gm...>]

2009/7/20 Karol M. Langner <kar...@gm...>:
> On Friday 12 June 2009 08:36:43 Manuel Melle Franco wrote:
>> On Fri, Jun 12, 2009 at 12:09 AM, Karol Langner
> <kar...@gm...>wrote:
>> > Hi Manuel,
>> >
>> >  As far as I know, ONIOM is not implemented in an other quantum chem
>> > program
>> > that we support at the moment. That being said, parsing ONIOM-specific
>> > stuff
>> > from only Gaussian would not be very useful in general, as we aim for
>> > extracted the same data from many programs. Correct me if I'm wrong,
>> > Noel.
>> >
>> >  However, if ONIOM output is not parsed by cclib, then that is a bug and
>> > we should be able to fix it.
>>
>> Dear Karol and Noel,
>>
>> As far as I know only gaussian and nwchem have it.
>> I am quite new to it and currently doing the parsing myself (fgrep mostly)
>> but it works only in the cases I use,
>> and it is a bit of a pain in the ass to parse.
>>
>> I agree with Karol, I think it might not be easy to write a general parser
>> for oniom so that it would not be worth the effort to add it to cclib, at
>> list at this point, maybe when more program have it, if that ever happens.
>>
>> best regards
>>
>> Manuel
>
> Manuel,
>
>  I fixed the gaussian parser so that it does not fail now when parsing ONIOM
> calcs. Attached are two Gausian03 test files that I used for testing. Since
> it is not a new feature, I'm not adding them as tests or regressions.

Please do add the files - this is what the regression test suite is for. :-)

>  Please be aware that this does not mean that cclib now parses ONIOM output --
> it does not! It merely does not crash now. Perhaps some of the parsed data
> can be of use, anyway.


>  If you have output that still crashes the parser (using today's revision),
> please send it to us so we can make the parser more failure-proof.
>
> Cheers,
> Karol
>
> --
> written by Karol Langner
> Mon Jul 20 17:23:26 CEST 2009
>