Use keywords in pdfs to search for existing tags

During processing, keywords stored in PDF metadata are used to look
them up in the tag database and associate any existing tags to the
item.

See #175

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

@@ -53,6 +53,9 @@ case class Column(name: String, ns: String = "", alias: String = "") {
   def isIn[A: Put](values: NonEmptyList[A]): Fragment =
     isIn(values.map(a => sql"$a").toList)
 
+  def isLowerIn[A: Put](values: NonEmptyList[A]): Fragment =
+    fr"lower(" ++ f ++ fr") IN (" ++ commas(values.map(a => sql"$a").toList) ++ fr")"
+
   def isIn(frag: Fragment): Fragment =
     f ++ fr"IN (" ++ frag ++ fr")"
 

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

@@ -314,6 +314,9 @@ object RItem {
   def findByIdAndCollective(itemId: Ident, coll: Ident): ConnectionIO[Option[RItem]] =
     selectSimple(all, table, and(id.is(itemId), cid.is(coll))).query[RItem].option
 
+  def checkByIdAndCollective(itemId: Ident, coll: Ident): ConnectionIO[Option[Ident]] =
+    selectSimple(Seq(id), table, and(id.is(itemId), cid.is(coll))).query[Ident].option
+
   def removeFolder(folderId: Ident): ConnectionIO[Int] = {
     val empty: Option[Ident] = None
     updateRow(table, folder.is(folderId), folder.setTo(empty)).update.run

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

@@ -1,5 +1,8 @@
 package docspell.store.records
 
+import cats.data.NonEmptyList
+import cats.implicits._
+
 import docspell.common._
 import docspell.store.impl.Implicits._
 import docspell.store.impl._
@@ -101,6 +104,21 @@ object RTag {
     ) ++ orderBy(name.prefix("t").asc)).query[RTag].to[Vector]
   }
 
+  def findAllByNameOrId(
+      nameOrIds: List[String],
+      coll: Ident
+  ): ConnectionIO[Vector[RTag]] = {
+    val idList =
+      NonEmptyList.fromList(nameOrIds.flatMap(s => Ident.fromString(s).toOption)).toSeq
+    val nameList = NonEmptyList.fromList(nameOrIds.map(_.toLowerCase)).toSeq
+
+    val cond = idList.flatMap(ids => Seq(tid.isIn(ids))) ++
+      nameList.flatMap(ns => Seq(name.isLowerIn(ns)))
+
+    if (cond.isEmpty) Vector.empty.pure[ConnectionIO]
+    else selectSimple(all, table, and(cid.is(coll), or(cond))).query[RTag].to[Vector]
+  }
+
   def delete(tagId: Ident, coll: Ident): ConnectionIO[Int] =
     deleteFrom(table, and(tid.is(tagId), cid.is(coll))).update.run
 }

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

@@ -1,5 +1,6 @@
 package docspell.store.records
 
+import cats.data.NonEmptyList
 import cats.implicits._
 
 import docspell.common._
@@ -43,4 +44,28 @@ object RTagItem {
 
   def findByItem(item: Ident): ConnectionIO[Vector[RTagItem]] =
     selectSimple(all, table, itemId.is(item)).query[RTagItem].to[Vector]
+
+  def findAllIn(item: Ident, tags: Seq[Ident]): ConnectionIO[Vector[RTagItem]] =
+    NonEmptyList.fromList(tags.toList) match {
+      case Some(nel) =>
+        selectSimple(all, table, and(itemId.is(item), tagId.isIn(nel)))
+          .query[RTagItem]
+          .to[Vector]
+      case None =>
+        Vector.empty.pure[ConnectionIO]
+    }
+
+  def setAllTags(item: Ident, tags: Seq[Ident]): ConnectionIO[Int] =
+    if (tags.isEmpty) 0.pure[ConnectionIO]
+    else
+      for {
+        entities <- tags.toList.traverse(tagId =>
+          Ident.randomId[ConnectionIO].map(id => RTagItem(id, item, tagId))
+        )
+        n <- insertRows(
+          table,
+          all,
+          entities.map(v => fr"${v.tagItemId},${v.itemId},${v.tagId}")
+        ).update.run
+      } yield n
 }