Add Japanese Vertical Support Branch for Tesseract and Ocrmypdf OCR

docker/dockerfiles/joex.dockerfile

@@ -77,7 +77,7 @@ RUN \
   wget https://github.com/tesseract-ocr/tessdata/raw/main/khm.traineddata && \
   mv khm.traineddata /usr/share/tessdata
 
-# Using these data files for japanese, because they work better. See #973
+# Using these data files for japanese, because they work better. Includes vertical data. See #973 and #2445.
 RUN \
   wget https://raw.githubusercontent.com/tesseract-ocr/tessdata_fast/master/jpn_vert.traineddata && \
   wget https://raw.githubusercontent.com/tesseract-ocr/tessdata_fast/master/jpn.traineddata && \

modules/analysis/src/main/scala/docspell/analysis/date/DateFind.scala

@@ -125,6 +125,7 @@ object DateFind {
         case Language.Dutch      => dmy.or(ymd).or(mdy)
         case Language.Latvian    => dmy.or(lavLong).or(ymd)
         case Language.Japanese   => ymd
+        case Language.JpnVert    => ymd
         case Language.Hebrew     => dmy
         case Language.Lithuanian => ymd
         case Language.Polish     => dmy

modules/analysis/src/main/scala/docspell/analysis/date/MonthName.scala

@@ -54,6 +54,8 @@ object MonthName {
         latvian
       case Language.Japanese =>
         japanese
+      case Language.JpnVert =>
+        japanese
       case Language.Hebrew =>
         hebrew
       case Language.Lithuanian =>

modules/analysis/src/test/scala/docspell/analysis/date/DateFindTest.scala

@@ -143,6 +143,7 @@ class DateFindTest extends FunSuite {
     )
   }
 
+  /*The output of a vertical OCR should also be horizontal and be able to be tested in the same way, so an additional test for vertical Japanese is not necessary.*/
   test("find japanese dates") {
     assertEquals(
       DateFind

modules/common/src/main/scala/docspell/common/Language.scala

@@ -123,6 +123,11 @@ object Language {
     val iso3 = "jpn"
   }
 
+  /*It's not an ISO value, but this needs to be unique and tesseract will need jpn_vert for it's scan from the config of /etc/docspell-joex/docspell-joex.conf.*/
+  case object JpnVert extends Language {
+    val iso2 = "ja_vert"
+    val iso3 = "jpn_vert"
+  }
   case object Hebrew extends Language {
     val iso2 = "he"
     val iso3 = "heb"
@@ -172,6 +177,7 @@ object Language {
       Romanian,
       Latvian,
       Japanese,
+      JpnVert,
       Hebrew,
       Lithuanian,
       Polish,

modules/fts-psql/src/main/scala/docspell/ftspsql/FtsRepository.scala

@@ -201,6 +201,7 @@ object FtsRepository extends DoobieMeta {
       case Language.Czech      => "simple"
       case Language.Latvian    => "simple"
       case Language.Japanese   => "simple"
+      case Language.JpnVert    => "simple"
       case Language.Hebrew     => "simple"
       case Language.Lithuanian => "simple"
       case Language.Polish     => "simple"

modules/webapp/src/main/elm/Data/Language.elm

@@ -30,6 +30,7 @@ type Language
     | Dutch
     | Latvian
     | Japanese
+    | JpnVert
     | Hebrew
     | Hungarian
     | Lithuanian
@@ -90,6 +91,9 @@ fromString str =
     else if str == "jpn" || str == "ja" || str == "japanese" then
         Just Japanese
 
+    else if str == "jpn_vert" || str == "ja_vert" || str == "jpnvert" then
+        Just Japanese
+
     else if str == "heb" || str == "he" || str == "hebrew" then
         Just Hebrew
 
@@ -169,6 +173,9 @@ toIso3 lang =
         Japanese ->
             "jpn"
 
+        JpnVert ->
+            "jpn_vert"
+
         Hebrew ->
             "heb"
 
@@ -212,6 +219,7 @@ all =
     , Romanian
     , Latvian
     , Japanese
+    , JpnVert
     , Hebrew
     , Hungarian
     , Lithuanian

modules/webapp/src/main/elm/Messages/Data/Language.elm

@@ -65,6 +65,9 @@ gb lang =
         Japanese ->
             "Japanese"
 
+        JpnVert ->
+            "JpnVert"
+
         Hebrew ->
             "Hebrew"
 
@@ -141,6 +144,9 @@ de lang =
         Japanese ->
             "Japanisch"
 
+        JpnVert ->
+            "JpnSenk"
+
         Hebrew ->
             "Hebräisch"
 
@@ -217,6 +223,9 @@ fr lang =
         Japanese ->
             "Japonnais"
 
+        JpnVert ->
+            "JpnVert"
+
         Hebrew ->
             "Hébreu"