Finally got around to looking in to this. The patch would overwrite the cif
reader with the mmcif reader, like so:
bool CIFFormat::ReadMolecule(OBBase* pOb, OBConversion* pConv)
{
OBFormat *obformat = OBFormat::FindType("mmcif");
return obformat->ReadMolecule(pOb, pConv);
}
Here’s the history of cifformat edits
<https://github.com/openbabel/openbabel/commits/master/src/formats/cifformat.cpp>,
and two pull requests (88 <https://github.com/openbabel/openbabel/pull/88>
and 89 <https://github.com/openbabel/openbabel/pull/89>) incorporating this
into mmcif. I read through the cifformat history and I think I got
everything, but I’m only testing against a directory of random cifs, ie.
I’m sure this patch would break someone’s workflow.
Anecdotally, I’ve been using the above patch for ~2 months without issue.
It performs better in my use case, which is cifs from the CCDC
<http://www.ccdc.cam.ac.uk/pages/Home.aspx>. I think it's worthwhile, but
want to see if there are any strong opinions against workflow-breaking
edits before making a pull request.
Pat
On Thu, Aug 21, 2014 at 10:37 AM, Patrick Fuller <patrickful...@gmail.com>
wrote:
> > Since you and others work wit CIFs more than myself, perhaps you’d be
> willing to take up some “maintenance” of MMCIF? I can also ask Chris Wilmer
> now that he’s here at Pitt and may have some ideas.
>
> I'm willing to help here. A better cif parser means less dropping down to
> mercury / materials studio / etc., which is better for my group's workflow.
>
> > I’m well aware that the CIF parser has received many more patches and
> modifications than MMCIF (e.g., for minerals, odd space groups, missing
> tags) so I’d suggest being careful and asking folks like Jean Brefort
> and/or Jure Varlec, who have contributed more.
>
> I'll read through the commit history and the `ReadMolecule`s of the two
> files and see how far apart they are. And I definitely welcome comments
> from the more experienced developer community before I go and make some
> potentially breaking changes.
>
> Pat
>
>
> On Thu, Aug 21, 2014 at 9:58 AM, Geoffrey Hutchison <
> geoff.hutchi...@gmail.com> wrote:
>
>> > Outside of two small hiccups (one pull request here, another to come),
>> it works amazingly well. Reading through the source explains this to an
>> extent, as it looks like the mmcif ReadMolecule is more built up than the
>> cif analog.
>> > A folder of random cifs is not a thorough test, so I was wondering if
>> anyone has a workflow where the cif parser works and the mmcif does not. If
>> not, would it be a good idea to merge the cif and mmcif parsers?
>>
>> I know very little about MMCIF, only that it’s designed for
>> macromolecules and has different dictionaries. I think in principal, the
>> formats could be merged, since MMCIF is simply an extension of CIF.
>>
>> I’m quite surprised that people have been doing this (and not asking
>> “hey, why does the MMCIF seem to work better?”).
>>
>> Since you and others work wit CIFs more than myself, perhaps you’d be
>> willing to take up some “maintenance” of MMCIF? I can also ask Chris Wilmer
>> now that he’s here at Pitt and may have some ideas.
>>
>> I’m well aware that the CIF parser has received many more patches and
>> modifications than MMCIF (e.g., for minerals, odd space groups, missing
>> tags) so I’d suggest being careful and asking folks like Jean Brefort
>> and/or Jure Varlec, who have contributed more.
>>
>> -Geoff
>
>
>
------------------------------------------------------------------------------
Meet PCI DSS 3.0 Compliance Requirements with EventLog Analyzer
Achieve PCI DSS 3.0 Compliant Status with Out-of-the-box PCI DSS Reports
Are you Audit-Ready for PCI DSS 3.0 Compliance? Download White paper
Comply to PCI DSS 3.0 Requirement 10 and 11.5 with EventLog Analyzer
http://p.sf.net/sfu/Zoho
_______________________________________________
OpenBabel-Devel mailing list
OpenBabel-Devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/openbabel-devel
