Closed acka47 closed 7 years ago
I just took a look at a subset to find out what values can be found in 531.z. Taking the following steps (see also https://wiki1.hbz-nrw.de/pages/viewpage.action?pageId=9867212):
/files/open_data/closed/hbzvk/snapshot$ for i in $(find 01852 -name "HT01852*.bz2"); do printf $i ; bzcat $i | grep --color -A 4 "<datafield tag=\"531\" ind1=\"z\" ind2=\".\"><subfield code=\"p\">" ; done
<datafield tag=\"531\" ind1=\"z\" ind2=\".\"><subfield code=\"p\">[^<]*
sort 531values.txt | uniq -c
Here is the result:
1 4. Aufl. u.d.T.
1 Als Vorg. gilt --->
5 Darin aufgeg. --->
7 Druckausg. u. Vorg. --->
2 Früher u.d.T.
4 Früher u.d. T.
2 Hervorgeg. aus --->
1 Vorg. als Druckausg. --->
16 Vorg. --->
103 Vorg.: --->
Another sample (or i in $(find 00449 -name "HT00449*.bz2")) gives the following results:
1 2004 - 2011 Beil., dann darin aufgeg. --->
1 2009 darin aufgeg. --->
1 44.1995 - 45.1996 Beil., dann darin aufgeg. --->
1 6. Aufl. u.d.T.
1 Ab 1984 darin aufgeg. --->
3 Bd. 1 u.d.T.
1 Beil. u. darin aufgeg. --->
1 Beil zu, dann darin aufgeg. --->
3 Daraus hervorgeg. u. darin aufgeg. --->
6 Darin aufgeg. --->
1 Darin, später darin aufgeg. --->
1 Praxisreihe u. darin aufgeg. --->
82 Vorg. --->
34 Vorg.: --->
1 Vorg u. 23/24.1988 --->
Obviously, this field generally contains information about predecessors, so we should actually use http://rdaregistry.info/Elements/u/P60683 to model this in JSON-LD:
"@context": {
"hasPredecessor": {
"@type": "@id",
"@id": "http://rdaregistry.info/Elements/u/P60683"
}
},
"id" : "http://lobid.org/resources/HT018885805#!",
"hasPredecessor": {
"id": "http://lobid.org/resources/HT018885808",
"label": "Elternarbeit in Rheinland-Pfalz"
}
}
Would the approach outlined in the previous comment work for you, @jschnasse ?
We might also add the info from subfield p
as rdfs:comment
or similar in the case edoweb stakeholders want to show this information.
{
"@context": {
"hasPredecessor": {
"@type": "@id",
"@id": "http://rdaregistry.info/Elements/u/P60683"
},
"comment": "http://www.w3.org/2000/01/rdf-schema#comment"
},
"id" : "http://lobid.org/resources/HT018885805#!",
"hasPredecessor": {
"id": "http://lobid.org/resources/HT018885808",
"label": "Elternarbeit in Rheinland-Pfalz",
"comment": "Fortsetzung von"
}
}
Predecessors and successors are already added. The contents of 531z1.p
or 533z1.p
aren't standardized and mostly don't contain additional information. Therefore I'm closing this issue.
Request by edoweb stakeholders (edited for better readability):
From the example:
Generally, field 531 is about "related entries" and predecessor/successor relations are only one kind of relationships, I assume. Thus, it is hard to capture this in RDF without loosing the special character of the relationshhip named in subfield
p
(here: "Fortsetzung von"). We could do something like this (example for API 2.0):@jschnasse says he would also need this in API 1.0.