Skip pdf conversion if a converted file exists

For images the conversion also returns the extracted text. If this
would have failed to be saved, it is extracted in the following
text-extraction step.

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

@@ -38,10 +38,20 @@ object ConvertPdf {
       item: ItemData
   ): Task[F, ProcessItemArgs, ItemData] =
     Task { ctx =>
-      def convert(ra: RAttachment) =
-        findMime(ctx)(ra).flatMap(m =>
-          convertSafe(cfg, JsoupSanitizer.clean, ctx, item)(ra, m)
-        )
+      def convert(ra: RAttachment): F[(RAttachment, Option[RAttachmentMeta])] =
+        isConverted(ctx)(ra).flatMap {
+          case true =>
+            ctx.logger.info(
+              s"Conversion to pdf already done for attachment ${ra.name}."
+            ) *>
+              ctx.store
+                .transact(RAttachmentMeta.findById(ra.id))
+                .map(rmOpt => (ra, rmOpt))
+          case false =>
+            findMime(ctx)(ra).flatMap(m =>
+              convertSafe(cfg, JsoupSanitizer.clean, ctx, item)(ra, m)
+            )
+        }
 
       for {
         ras <- item.attachments.traverse(convert)
@@ -51,6 +61,11 @@ object ConvertPdf {
 
     }
 
+  def isConverted[F[_]: Sync](ctx: Context[F, ProcessItemArgs])(
+      ra: RAttachment
+  ): F[Boolean] =
+    ctx.store.transact(RAttachmentSource.isConverted(ra.id))
+
   def findMime[F[_]: Functor](ctx: Context[F, _])(ra: RAttachment): F[Mimetype] =
     OptionT(ctx.store.transact(RFileMeta.findById(ra.fileId)))
       .map(_.mimetype)

modules/store/src/main/scala/docspell/store/impl/Column.scala

@@ -44,6 +44,9 @@ case class Column(name: String, ns: String = "", alias: String = "") {
   def isNot[A: Put](value: A): Fragment =
     f ++ fr"<> $value"
 
+  def isNot(c: Column): Fragment =
+    f ++ fr"<>" ++ c.f
+
   def isNull: Fragment =
     f ++ fr"is null"
 

modules/store/src/main/scala/docspell/store/records/RAttachmentMeta.scala

@@ -46,6 +46,9 @@ object RAttachmentMeta {
   def exists(attachId: Ident): ConnectionIO[Boolean] =
     selectCount(id, table, id.is(attachId)).query[Int].unique.map(_ > 0)
 
+  def findById(attachId: Ident): ConnectionIO[Option[RAttachmentMeta]] =
+    selectSimple(all, table, id.is(attachId)).query[RAttachmentMeta].option
+
   def upsert(v: RAttachmentMeta): ConnectionIO[Int] =
     for {
       n0 <- update(v)

modules/store/src/main/scala/docspell/store/records/RAttachmentSource.scala

@@ -48,6 +48,21 @@ object RAttachmentSource {
       .unique
       .map(_ > 0)
 
+  def isConverted(attachId: Ident): ConnectionIO[Boolean] = {
+    val sId   = Columns.id.prefix("s")
+    val sFile = Columns.fileId.prefix("s")
+    val aId   = RAttachment.Columns.id.prefix("a")
+    val aFile = RAttachment.Columns.fileId.prefix("a")
+
+    val from = table ++ fr"s INNER JOIN" ++
+      RAttachment.table ++ fr"a ON" ++ aId.is(sId)
+
+    selectCount(aId, from, and(aId.is(attachId), aFile.isNot(sFile)))
+      .query[Int]
+      .unique
+      .map(_ > 0)
+  }
+
   def delete(attachId: Ident): ConnectionIO[Int] =
     deleteFrom(table, id.is(attachId)).update.run