tests: use workspace manifesto→aufklaerung (1→2 pages), binarize ad h…

…oc where needed
OCR-D · Aug 30, 2024 · c73b3ef · c73b3ef
1 parent 32b2e9c
commit c73b3ef
Show file tree

Hide file tree

Showing 2 changed files with 31 additions and 13 deletions.
diff --git a/tests/test_recognize.py b/tests/test_recognize.py
@@ -2,14 +2,23 @@
 
 from ocrd import run_processor
 from ocrd_kraken.recognize import KrakenRecognize
+from ocrd_kraken.binarize import KrakenBinarize
 
 
-def test_recognize(workspace_manifesto):
+def test_recognize(workspace_aufklaerung):
+    # some models (like default en) require binarized images
+    run_processor(KrakenBinarize,
+                  input_file_grp="OCR-D-GT-PAGE",
+                  output_file_grp="OCR-D-GT-PAGE-BIN",
+                  **workspace_aufklaerung,
+    )
     run_processor(KrakenRecognize,
-                  input_file_grp="OCR-D-SEG-KRAKEN",
+                  # re-use layout, overwrite text:
+                  input_file_grp="OCR-D-GT-PAGE-BIN",
                   output_file_grp="OCR-D-OCR-KRAKEN",
-                  **workspace_manifesto,
+                  parameter={'overwrite_text': True},
+                  **workspace_aufklaerung,
     )
-    ws = workspace_manifesto['workspace']
+    ws = workspace_aufklaerung['workspace']
     ws.save_mets()
     # FIXME: add result assertions (find_files, parsing PAGE etc)
diff --git a/tests/test_segment.py b/tests/test_segment.py
@@ -2,23 +2,25 @@
 
 from ocrd import run_processor
 from ocrd_kraken.segment import KrakenSegment
+from ocrd_kraken.binarize import KrakenBinarize
 
 
-def test_run_blla(workspace_manifesto):
+def test_run_blla(workspace_aufklaerung):
     run_processor(KrakenSegment,
-                  input_file_grp="OCR-D-IMG-BIN",
+                  input_file_grp="OCR-D-IMG",
                   output_file_grp="OCR-D-SEG-LINE-KRAKEN",
                   parameter={'maxcolseps': 0, 'use_legacy': False},
-                  **workspace_manifesto,
+                  **workspace_aufklaerung,
     )
-    ws = workspace_manifesto['workspace']
+    ws = workspace_aufklaerung['workspace']
     ws.save_mets()
     # FIXME: add result assertions (find_files, parsing PAGE etc)
 
 def test_run_blla_regionlevel(workspace_aufklaerung_region):
     run_processor(KrakenSegment,
                   input_file_grp="OCR-D-GT-SEG-REGION",
                   output_file_grp="OCR-D-SEG-LINE-KRAKEN",
+                  # only 1 page (takes 3min per page without GPU)
                   page_id="phys_0005",
                   parameter={'maxcolseps': 0, 'use_legacy': False},
                   **workspace_aufklaerung_region,
@@ -27,13 +29,20 @@ def test_run_blla_regionlevel(workspace_aufklaerung_region):
     ws.save_mets()
     # FIXME: add result assertions (find_files, parsing PAGE etc)
 
-def test_run_legacy(workspace_manifesto):
+def test_run_legacy(workspace_aufklaerung):
+    # legacy segmentation requires binarized images
+    run_processor(KrakenBinarize,
+                  input_file_grp="OCR-D-GT-PAGE",
+                  output_file_grp="OCR-D-GT-PAGE-BIN",
+                  **workspace_aufklaerung,
+    )
     run_processor(KrakenSegment,
-                  input_file_grp="OCR-D-IMG-BIN",
+                  # overwrite layout:
+                  input_file_grp="OCR-D-GT-PAGE-BIN",
                   output_file_grp="OCR-D-SEG-LINE-KRAKEN",
-                  parameter={'maxcolseps': 0, 'use_legacy': True},
-                  **workspace_manifesto,
+                  parameter={'maxcolseps': 0, 'use_legacy': True, 'overwrite_segments': True},
+                  **workspace_aufklaerung,
     )
-    ws = workspace_manifesto['workspace']
+    ws = workspace_aufklaerung['workspace']
     ws.save_mets()
     # FIXME: add result assertions (find_files, parsing PAGE etc)