antonis.lempesis
/
dnet-hadoop
forked from D-Net/dnet-hadoop
1
1
Fork 1
Code Issues Pull Requests Projects Releases Wiki Activity

removed alternate identifier in resolutionMap

This commit is contained in:
Sandro La Bruzzo 2021-11-12 09:56:45 +01:00
parent 2ca0a436ad
commit a7763d2492
1 changed files with 16 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

24
dhp-workflows/dhp-graph-mapper/src/main/java/eu/dnetlib/dhp/oa/graph/resolution/SparkResolveRelation.scala
View File

@ -96,6 +96,21 @@ object SparkResolveRelation {
.text(s"$graphBasePath/relation")
}
def extractInstanceCF(input: String): List[(String, String)] = {
implicit lazy val formats: DefaultFormats.type = org.json4s.DefaultFormats
lazy val json: json4s.JValue = parse(input)
val result: List[(String, String)] = for {
JObject(iObj) <- json \ "instance"
JField("collectedfrom", JObject(cf)) <- iObj
JField("instancetype", JObject(instancetype)) <- iObj
JField("value", JString(collectedFrom)) <- cf
JField("classname", JString(classname)) <- instancetype
} yield (classname, collectedFrom)
result
}
def extractPidsFromRecord(input: String): (String, List[(String, String)]) = {
implicit lazy val formats: DefaultFormats.type = org.json4s.DefaultFormats
@ -108,14 +123,7 @@ object SparkResolveRelation {
JField("classid", JString(pidType)) <- qualifier
} yield (pidValue, pidType)
val alternateIds: List[(String, String)] = for {
JObject(pids) <- json \\ "alternateIdentifier"
JField("value", JString(pidValue)) <- pids
JField("qualifier", JObject(qualifier)) <- pids
JField("classid", JString(pidType)) <- qualifier
} yield (pidValue, pidType)
(id, result ::: alternateIds)
(id, result)
}