The unquoted word 'global_' (with trailing underscore) is part of the STAR file syntax (upon which CIF is based), but is explicitly forbidden in the formal CIF syntax:

http://www.iucr.org/resources/cif/spec/version1.1/cifsyntax#global

However, for compatibility with the monomer library which is using the global_ keyword despite it being forbidden in the CIF syntax, we added support for a non-strict mode of the CIF parser in the iotbx which could support global_ data blocks. Of course, if the trailing underscore is lost, then it is a syntax error either way, as the parser is looking for the start of a data block, signified by a data block heading (starting with 'data_').

Cheers,

Richard


On 10 April 2013 16:24, Oleg Dolomanov <oleg@olexsys.org> wrote:
Sorry, yes,'global_' is also a special global data block and should work...

From: nechols@lbl.gov
Date: Wed, 10 Apr 2013 16:17:53 -0700

To: cctbxbb@phenix-online.org
Subject: Re: [cctbxbb] cif parser & ccp4 mon_lib incompatibility

On Wed, Apr 10, 2013 at 4:12 PM, Oleg Dolomanov <oleg@olexsys.org> wrote:
the CIF parser will report this error if any of the CIF items is outside data_ or save_ blocks. So you need to change 'global' to 'data_global' to prevent this error.

This would be true when using the CIF reader on its own, but for macromolecular apps (including the test script that's failing), we disable this error checking specifically for the CCP4 monomer library. �In fact the version we distribute with Phenix ($PHENIX/chem_data/mon_lib/list/mon_lib_list.cif) starts like this:

global_
_lib_name � � � � mon_lib
_lib_version � � �4.11
_lib_update � � � 15/04/05

It is very likely that we may start to distribute the entire chem_data tree from our web site as an optional add-on, which will greatly ease the use of CCTBX in other macromolecular applications (e.g. MolProbity). �Of course the CCP4 version will be more complete and up-to-date, although a) some of those restraints are not especially well-validated, and b) we incorporate additional information in the restraints we distribute.

-Nat

_______________________________________________ cctbxbb mailing list cctbxbb@phenix-online.org http://phenix-online.org/mailman/listinfo/cctbxbb

_______________________________________________
cctbxbb mailing list
cctbxbb@phenix-online.org
http://phenix-online.org/mailman/listinfo/cctbxbb