semantic-kraus / dw-data

converts XML/TEI data into CIDOCish RDF
https://semantic-kraus.github.io/dw-data/
MIT License
0 stars 1 forks source link

Deleting some wholeperiodicals #26

Closed BOberreither closed 1 year ago

BOberreither commented 1 year ago

Right now, when there is a citedRange[@wholePeriodical="yes"], an F24 is built from title j key; that's correct.

But also, an F24 is built from bibl/@xml:id and from citedRange/@xml:id, including respective F30, E52, F22, E42 statements. IF there is a citedRange[@wholePeriodical="yes"] in th bibl element, all these statements should not be built.

BergListe commented 1 year ago

Could you share an example, please?

BOberreither commented 1 year ago

@BergListe Sure:

From title j key (correct):

    <https://sk.acdh.oeaw.ac.at/DWbibl04277/published-expression> a frbroo:F24_Publication_Expression ;
        rdfs:label "Periodical: Ostara"@en .

From bibl xml:id (wrong):

<https://sk.acdh.oeaw.ac.at/DWbibl03236/publication> a frbroo:F30_Publication_Event ;
        rdfs:label "Publication of: Published Expression: "@en ;
        cidoc:P4_has_time-span <https://sk.acdh.oeaw.ac.at/DWbibl03236/publication/time-span> ;
        cidoc:P7_took_place_at sk:DWplace00085 ;
        cidoc:R24_created <https://sk.acdh.oeaw.ac.at/DWbibl03236/published-expression> .

    <https://sk.acdh.oeaw.ac.at/DWbibl03236/publication/time-span> a cidoc:E52_Time-Span ;
        rdfs:label "1905 - 1917"@en ;
        cidoc:P4i_is_time-span_of <https://sk.acdh.oeaw.ac.at/DWbibl03236/publication> ;
        cidoc:P82a_begin_of_the_begin "1905"^^xsd:gYear ;
        cidoc:P82b_end_of_the_end "1917"^^xsd:gYear .

    <https://sk.acdh.oeaw.ac.at/DWbibl03236/published-expression> a frbroo:F24_Publication_Expression ;
        rdfs:label "Published Expression: "@en ;
        cidoc:P165_incorporates sk:DWbibl03237 ;
        cidoc:P1_is_identified_by <https://sk.acdh.oeaw.ac.at/DWbibl03237/appellation/0> ;
        cidoc:R24i_was_created_through <https://sk.acdh.oeaw.ac.at/DWbibl03237/publication> .

Having F30 and E52 with the right URIs would be nice, but is not at the top of my priority list right now ...

F22 from citedRange wholePeriodical (also wrong):

    sk:DWbibl03237 a frbroo:F22_Self-Contained_Expression ;
        rdfs:label "Expression: [Expression]"@en .
BergListe commented 1 year ago

... including respective F30, E52, F22, E42 statements.

I think, I fixed this issue but I didn't find a case of an oddly generated E42. If you stumble upon one please share.

BOberreither commented 1 year ago

@BergListe So, i found a bug - in my comment above: The last wrongly built statement including attached E42 and F28 is built not from wholeText citedRange, but from wholePeriodical citedRange. If there is one of these, don't create F22, E42 and F28 from it. The F22 is built from the title j key, and thats enough from a bibl containing a wholePeriodical.

BergListe commented 1 year ago

Now only items with citedRange[@wholePeriodical="yes"] are excluded from the creation of the above mentioned statements.