Re: [ddlm-group] Dictionary conformance (was Re: Second proposal toallow looping of 'Set' categories)
- To: Group finalising DDLm and associated dictionaries <ddlm-group@iucr.org>
- Subject: Re: [ddlm-group] Dictionary conformance (was Re: Second proposal toallow looping of 'Set' categories)
- From: "Bollinger, John C" <John.Bollinger@STJUDE.ORG>
- Date: Thu, 16 Jun 2016 13:51:21 +0000
- Accept-Language: en-US
- authentication-results: spf=none (sender IP is )smtp.mailfrom=John.Bollinger@STJUDE.ORG;
- DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=SJCRH.onmicrosoft.com; s=selector1-stjude-org;h=From:Date:Subject:Message-ID:Content-Type:MIME-Version;bh=xcMcbxzDBUqJXseYH+9T+w9in2vT7MskxCKn8mjDP5s=;b=IbqRICcL4i7jjKCPpOLcTahHCF54ojFtrKzeAA6H01rUPatUGvQ2JQNJQIpk1WBZtjYJCJYkK78n8/57fT34RQJESLeDb8v9SBo7PNcrmym9bQebXvlQgYCanFFYlM7Z7Oq8llJ+3cCQUhSSxcEFAMUtTIWP8n98N2M2itnLx34=
- In-Reply-To: <CAM+dB2fAd2WmpVtqksZ7UA3tJkG-LiLtUfr0Ytm2U4O1G7jQmQ@mail.gmail.com>
- References: <CAM+dB2e-p0gORJGKUXMf8m+h4jhqEaGKbVgQuTPKG1KQobysew@mail.gmail.com><1699752094.6098741.1465993722575.JavaMail.yahoo@mail.yahoo.com><BY2PR0401MB09361117BEC19164C92892B2E0550@BY2PR0401MB0936.namprd04.prod.outlook.com><1573396202.6231045.1465999233495.JavaMail.yahoo@mail.yahoo.com><BY2PR0401MB093601E709C28EFB7BDB620DE0550@BY2PR0401MB0936.namprd04.prod.outlook.com><CAM+dB2fAd2WmpVtqksZ7UA3tJkG-LiLtUfr0Ytm2U4O1G7jQmQ@mail.gmail.com>
- spamdiagnosticmetadata: NSPM
- spamdiagnosticoutput: 1:99
Dear James and Colleagues, I agree that we should not require data files to replicate in audit_conform the imports performed by the dictionaries with which they specify direct conformance.
Thus, I agree that in the example presented, it is sufficient to specify conformance with just magcif and pdcif, omitting core_cif, etc.. I see no reason to forbid data files from specifying additional dictionary conformance, however, such as if the example
also declared direct conformance with the core. I don’t think such a prohibition is intended, but my agreement is predicated on that understanding. Complications may arise if audit_conform rows designate conformance with dictionary versions (via _audit_conform.dict_version) that rely on different versions
of the same dictionary, or if one directly designates conformance with a different version of a dictionary (maybe cif_core) than other dictionaries it claims conformance with rely upon. I am prepared to ignore that issue, however, on the basis that a CIF
presenting such conformance assertions is inconsistent. In such a case there is anyway a reasonable chance that the import semantics described would yield a combined dictionary suitable for validating the data file. I also observe that when a data file documents conformance with a specific dictionary version, that seems to require the corresponding notional _import.get element
to identify the target dictionary by a URI referencing the specific version designated. That information could be drawn from _audit_conform.dict_location, if that is provided, but my understanding is that there is (or was or should have been) an online registry
maintained by IUCr for identifying dictionary locations from name and version (see http://www.iucr.org/__data/iucr/lists/cif-developers/msg00044.html, section 2). I do not recommend that reliance on such a registry be explicitly written into the DDLm definition
of audit_conform; rather, I suggest that the means by which the needed dictionary is identified be left unspecified, as indeed it is in the DDL1 core. Regards, John From: ddlm-group [mailto:ddlm-group-bounces@iucr.org]
On Behalf Of James Hester I agree that audit_conform must be looped. In order to write the definition we must answer some questions about the semantic interpretation of multiple dictionaries. To make this concrete, let's suppose that
we have a powder diffraction result that reports a magnetic structure. This datablock will have items from pd_cif and mag_cif. Both pd_cif and mag_cif internally import core_cif, as well as templ_enum and templ_attr. Which of these dictionaries appear in the audit_conform loop? I think it should be pd_cif and mag_cif, and that the semantic interpretation is as if there was a virtual dictionary with an import.get in 'Full'
mode of the listed dictionaries into its HEAD category. For our example, it would be as if the following lines were present in the virtual dictionary HEAD, where the "file" entries point to the precise URL given in _audit_conform: _import.get [{"file":magcif.dic "save":MAGCIF "mode":Full "dupl":Ignore "miss":Exit} Note that we must have "dupl":Ignore (or Replace) in order to account for the fact that core_cif definitions will be notionally present in both dictionaries. Does this sound reasonable? James. On 16 June 2016 at 01:03, Bollinger, John C <John.Bollinger@stjude.org> wrote:
T +61 (02) 9717 9907 |
_______________________________________________ ddlm-group mailing list ddlm-group@iucr.org http://mailman.iucr.org/cgi-bin/mailman/listinfo/ddlm-group
Reply to: [list | sender only]
- Follow-Ups:
- References:
- [ddlm-group] Dictionary conformance (was Re: Second proposal toallow looping of 'Set' categories) (James Hester)
- Re: [ddlm-group] Dictionary conformance (was Re: Second proposal toallow looping of 'Set' categories) (SIMON WESTRIP)
- Re: [ddlm-group] Dictionary conformance (was Re: Second proposal toallow looping of 'Set' categories) (Bollinger, John C)
- Re: [ddlm-group] Dictionary conformance (was Re: Second proposal toallow looping of 'Set' categories) (SIMON WESTRIP)
- Re: [ddlm-group] Dictionary conformance (was Re: Second proposal toallow looping of 'Set' categories) (Bollinger, John C)
- Re: [ddlm-group] Dictionary conformance (was Re: Second proposal toallow looping of 'Set' categories) (James Hester)
- Prev by Date: Re: [ddlm-group] Dictionary conformance (was Re: Second proposal toallow looping of 'Set' categories)
- Next by Date: Re: [ddlm-group] Dictionary conformance (was Re: Second proposal toallow looping of 'Set' categories)
- Prev by thread: Re: [ddlm-group] Dictionary conformance (was Re: Second proposal toallow looping of 'Set' categories)
- Next by thread: Re: [ddlm-group] Dictionary conformance (was Re: Second proposal toallow looping of 'Set' categories)
- Index(es):