You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Until now we only clear the fuseki graph for an entire corpus when we update the corpus in total. When updating individual documents of a corpus, ignorer to avoid duplicates, we would have to delete existing triples before posting a new RDF representation to the triple store the following way:
Remove triples of blank node used to hook wikidata-id to the creator of the play:
WITH <https://dracor.org/test> DELETE {
<https://dracor.org/entity/greek000026> ?p ?o}
WHERE {
<https://dracor.org/entity/greek000026> ?p ?o
}
The problem is, that there is currently only a connection from the play to the character; so we have to remove them with a separate query. The information on the author of a play is attached at two points (1. cidoc-like via a creation event, 2. via property of the dracor-ontology dracon:has_author) We maybe could include it only once (1.?), this would save an update operation, but makes it probably more complicated to query the author-info. I should probably think of a way to include the inverse path of schema:character to connect play to character, but I'm not sure if this is possible within schema.org. And, maybe, the queries could be combined into one, but I unfortunately don't know how.
Until now we only clear the fuseki graph for an entire corpus when we update the corpus in total. When updating individual documents of a corpus, ignorer to avoid duplicates, we would have to delete existing triples before posting a new RDF representation to the triple store the following way:
The problem is, that there is currently only a connection from the play to the character; so we have to remove them with a separate query. The information on the author of a play is attached at two points (1. cidoc-like via a creation event, 2. via property of the dracor-ontology
dracon:has_author
) We maybe could include it only once (1.?), this would save an update operation, but makes it probably more complicated to query the author-info. I should probably think of a way to include the inverse path ofschema:character
to connect play to character, but I'm not sure if this is possible within schema.org. And, maybe, the queries could be combined into one, but I unfortunately don't know how.Prefixes I included while testing:
Originally posted by @ingoboerner in #77 (comment)
The text was updated successfully, but these errors were encountered: