dnet-hadoop/dhp-workflows/dhp-aggregation/src/main/scala/eu/dnetlib/dhp/collection/mag/SparkCreateMagDenormalizedT...

224 lines
7.7 KiB
Scala

package eu.dnetlib.dhp.collection.mag
import eu.dnetlib.dhp.application.AbstractScalaApplication
import org.apache.spark.sql.functions._
import org.apache.spark.sql.types.{StringType, StructField, StructType}
import org.apache.spark.sql.{Dataset, Row, SparkSession}
import org.slf4j.{Logger, LoggerFactory}
class SparkCreateMagDenormalizedTable(propertyPath: String, args: Array[String], log: Logger)
extends AbstractScalaApplication(propertyPath, args, log: Logger) {
/** Here all the spark applications runs this method
* where the whole logic of the spark node is defined
*/
override def run(): Unit = {
val magBasePath: String = parser.get("magBasePath")
log.info("found parameters magBasePath: {}", magBasePath)
val workingPath: String = parser.get("workingPath")
log.info("found parameters workingPath: {}", workingPath)
generatedDenormalizedMAGTable(spark, magBasePath, workingPath)
}
private def generatedDenormalizedMAGTable(
spark: SparkSession,
magBasePath: String,
workingPath: String
): Unit = {
import spark.implicits._
val schema: StructType = StructType(StructField("DOI", StringType) :: Nil)
//Filter all the MAG Papers that intersect with a Crossref DOI
val magPapers = MagUtility
.loadMagEntity(spark, "Papers", magBasePath)
.withColumn("Doi", lower(col("Doi")))
magPapers.cache()
magPapers.count()
//log.info("Create current abstract")
//Abstract is an inverted list, we define a function that convert in string the abstract and recreate
// a table(PaperId, Abstract)
val paperAbstract = MagUtility
.loadMagEntity(spark, "PaperAbstractsInvertedIndex", magBasePath)
.map(s => (s.getLong(0), MagUtility.convertInvertedIndexString(s.getString(1))))
.withColumnRenamed("_1", "PaperId")
.withColumnRenamed("_2", "Abstract")
//We define Step0 as the result of the Join between PaperIntersection and the PaperAbstract
val step0 = magPapers
.join(paperAbstract, magPapers("PaperId") === paperAbstract("PaperId"), "left")
.select(magPapers("*"), paperAbstract("Abstract"))
.cache()
step0.count()
magPapers.unpersist()
// We have three table Author, Affiliation, and PaperAuthorAffiliation, in the
//next step we create a table containing
val authors = MagUtility.loadMagEntity(spark, "Authors", magBasePath)
val affiliations = MagUtility.loadMagEntity(spark, "Affiliations", magBasePath)
val paperAuthorAffiliations = MagUtility.loadMagEntity(spark, "PaperAuthorAffiliations", magBasePath)
val j1 = paperAuthorAffiliations
.join(authors, paperAuthorAffiliations("AuthorId") === authors("AuthorId"), "inner")
.select(
col("PaperId"),
col("AffiliationId"),
col("AuthorSequenceNumber"),
authors("DisplayName").alias("AuthorName"),
authors("AuthorId")
)
val paperAuthorAffiliationNormalized = j1
.join(affiliations, j1("AffiliationId") === affiliations("AffiliationId"), "left")
.select(j1("*"), affiliations("DisplayName").alias("AffiliationName"), affiliations("GridId"))
.groupBy("PaperId")
.agg(
collect_list(
struct("AffiliationId", "AuthorSequenceNumber", "AffiliationName", "AuthorName", "AuthorId", "GridId")
).alias("authors")
)
val step1 = step0
.join(paperAuthorAffiliationNormalized, step0("PaperId") === paperAuthorAffiliationNormalized("PaperId"), "left")
.select(step0("*"), paperAuthorAffiliationNormalized("authors"))
.cache()
step1.count()
step0.unpersist()
val conference = MagUtility
.loadMagEntity(spark, "ConferenceInstances", magBasePath)
.select(
$"ConferenceInstanceId",
$"DisplayName".as("conferenceName"),
$"Location".as("conferenceLocation"),
$"StartDate".as("conferenceStartDate"),
$"EndDate".as("conferenceEndDate")
)
val step2 = step1
.join(conference, step1("ConferenceInstanceId") === conference("ConferenceInstanceId"), "left")
.select(
step1("*"),
conference("conferenceName"),
conference("conferenceLocation"),
conference("conferenceStartDate"),
conference("conferenceEndDate")
)
.cache()
step2.count()
step1.unpersist()
// val fos = MagUtility
// .loadMagEntity(spark, "FieldsOfStudy", magBasePath)
// .select($"FieldOfStudyId".alias("fos"), $"DisplayName", $"MainType")
//
// val paperFieldsOfStudy = MagUtility
// .loadMagEntity(spark, "PaperFieldsOfStudy", magBasePath)
// .select($"FieldOfStudyId", $"Score", $"PaperId")
//
// val paperFoS = paperFieldsOfStudy
// .join(broadcast(fos), fos("fos") === paperFieldsOfStudy("FieldOfStudyId"))
// .groupBy("PaperId")
// .agg(collect_set(struct("FieldOfStudyId", "DisplayName", "MainType", "Score")).as("FoS"))
//
// val step3 = step2
// .join(paperFoS, step2("PaperId") === paperFoS("PaperId"), "left")
// .select(step2("*"), paperFoS("FoS"))
// .cache()
// step3.count()
//
// step2.unpersist()
val journals = MagUtility
.loadMagEntity(spark, "Journals", magBasePath)
.select(
$"JournalId",
$"DisplayName".as("journalName"),
$"Issn".as("journalIssn"),
$"Publisher".as("journalPublisher")
)
val step3 = step2
.join(journals, step2("JournalId") === journals("JournalId"), "left")
.select(
step2("*"),
journals("journalName"),
journals("journalIssn"),
journals("journalPublisher")
)
.cache
step3.count()
val paper_urls = MagUtility
.loadMagEntity(spark, "PaperUrls", magBasePath)
.groupBy("PaperId")
.agg(slice(collect_set("SourceUrl"), 1, 6).alias("urls"))
.cache
paper_urls.count
step3
.join(paper_urls, step3("PaperId") === paper_urls("PaperId"))
.select(step3("*"), paper_urls("urls"))
.select(
$"PaperId".as("paperId"),
$"Doi".as("doi"),
$"DocType".as("docType"),
$"PaperTitle".as("paperTitle"),
$"OriginalTitle".as("originalTitle"),
$"BookTitle".as("bookTitle"),
$"Year".as("year"),
$"Date".as("date"),
$"OnlineDate".as("onlineDate"),
$"Publisher".as("publisher"),
$"JournalId".as("journalId"),
$"ConferenceSeriesId".as("conferenceSeriesId"),
$"ConferenceInstanceId".as("conferenceInstanceId"),
$"Volume".as("volume"),
$"Issue".as("issue"),
$"FirstPage".as("firstPage"),
$"LastPage".as("lastPage"),
$"ReferenceCount".as("referenceCount"),
$"CitationCount".as("citationCount"),
$"EstimatedCitation".as("estimatedCitation"),
$"OriginalVenue".as("originalVenue"),
$"FamilyId".as("familyId"),
$"FamilyRank".as("familyRank"),
$"DocSubTypes".as("docSubTypes"),
$"CreatedDate".as("createdDate"),
$"Abstract".as("abstractText"),
$"authors".as("authors"),
$"conferenceName".as("conferenceName"),
$"conferenceLocation".as("conferenceLocation"),
$"conferenceStartDate".as("conferenceStartDate"),
$"conferenceEndDate".as("conferenceEndDate"),
$"journalName".as("journalName"),
$"journalIssn".as("journalIssn"),
$"journalPublisher".as("journalPublisher"),
$"urls"
)
.write
.mode("OverWrite")
.save(s"$workingPath/mag")
step3.unpersist()
}
}
object SparkCreateMagDenormalizedTable {
val log: Logger = LoggerFactory.getLogger(SparkCreateMagDenormalizedTable.getClass)
def main(args: Array[String]): Unit = {
new SparkCreateMagDenormalizedTable(
"/eu/dnetlib/dhp/collection/mag/create_MAG_denormalized_table_properties.json",
args,
log
).initialize().run()
}
}