Add pdf meta data to extracted text to add it to full-text index

modules/extract/src/main/scala/docspell/extract/ExtractResult.scala

@@ -28,6 +28,12 @@ object ExtractResult {
 
   case class Success(text: String, pdfMeta: Option[PdfMetaData]) extends ExtractResult {
     val textOption = Some(text)
+    def appendPdfMetaToText: Success =
+      pdfMeta.flatMap(_.asText) match {
+        case Some(m) =>
+          copy(text = text + "\n\n" + m)
+        case None => this
+      }
   }
   def success(text: String, pdfMeta: Option[PdfMetaData]): ExtractResult =
     Success(text, pdfMeta)

modules/extract/src/main/scala/docspell/extract/pdfbox/PdfMetaData.scala

@@ -24,6 +24,15 @@ final case class PdfMetaData(
 
   def keywordList: List[String] =
     keywords.map(kws => kws.split("[,;]\\s*").toList).getOrElse(Nil)
+
+  /** Return all data in lines, except keywords. Keywords are handled separately. */
+  def asText: Option[String] =
+    (title.toList ++ author.toList ++ subject.toList ++ creationDate.toList.map(
+      _.toUtcDate.toString
+    )) match {
+      case Nil  => None
+      case list => Some(list.mkString("\n"))
+    }
 }
 
 object PdfMetaData {

modules/joex/src/main/scala/docspell/joex/process/TextExtraction.scala

@@ -97,7 +97,10 @@ object TextExtraction {
       res  <- extractTextFallback(ctx, cfg, ra, lang)(fids)
       meta = item.changeMeta(
         ra.id,
-        rm => rm.setContentIfEmpty(res.map(_.text.trim).filter(_.nonEmpty))
+        rm =>
+          rm.setContentIfEmpty(
+            res.map(_.appendPdfMetaToText.text.trim).filter(_.nonEmpty)
+          )
       )
       tags = res.flatMap(_.pdfMeta).map(_.keywordList).getOrElse(Nil)
       est <- dst