tests: add actual assertions

OCR-D · Aug 30, 2024 · fd15e2a · fd15e2a
1 parent ae6445b
commit fd15e2a
Show file tree

Hide file tree

Showing 3 changed files with 53 additions and 6 deletions.
diff --git a/tests/test_binarize.py b/tests/test_binarize.py
@@ -1,15 +1,31 @@
 # pylint: disable=import-error
 
 import json
+import os
 
 from ocrd import run_processor
+from ocrd_utils import MIMETYPE_PAGE
+from ocrd_models.constants import NAMESPACES
+from ocrd_modelfactory import page_from_file
+
 from ocrd_kraken.binarize import KrakenBinarize
 
 from .assets import assets
 
 
 PARAM_JSON = assets.url_of('param-binarize.json')
 
+def analyse_result(ws, level):
+    assert os.path.isdir(os.path.join(ws.directory, 'OCR-D-BIN-KRAKEN'))
+    out_files = list(ws.find_files(fileGrp="OCR-D-BIN-KRAKEN", mimetype=MIMETYPE_PAGE))
+    assert len(out_files), "found no output PAGE file"
+    out_images = list(ws.find_files(fileGrp="OCR-D-BIN-KRAKEN", mimetype="//^image/.*"))
+    assert len(out_images), "found no output image file"
+    out_pcgts = page_from_file(out_files[0])
+    assert out_pcgts is not None
+    out_images = out_pcgts.etree.xpath('//page:%s/page:AlternativeImage[contains(@comments,"binarized")]' % level, namespaces=NAMESPACES)
+    assert len(out_images) > 0, "found no binarized AlternativeImages in output PAGE file"
+
 def test_param_json(workspace_sbb):
     run_processor(KrakenBinarize,
                   input_file_grp="OCR-D-IMG",
@@ -19,6 +35,7 @@ def test_param_json(workspace_sbb):
     )
     ws = workspace_sbb['workspace']
     ws.save_mets()
+    analyse_result(ws, 'Page')
 
 def test_binarize_regions(workspace_aufklaerung):
     run_processor(KrakenBinarize,
@@ -29,7 +46,7 @@ def test_binarize_regions(workspace_aufklaerung):
     )
     ws = workspace_aufklaerung['workspace']
     ws.save_mets()
-    # FIXME: add result assertions (find_files, parsing PAGE etc)
+    analyse_result(ws, 'TextRegion')
 
 def test_binarize_lines(workspace_aufklaerung):
     run_processor(KrakenBinarize,
@@ -40,4 +57,5 @@ def test_binarize_lines(workspace_aufklaerung):
     )
     ws = workspace_aufklaerung['workspace']
     ws.save_mets()
-    # FIXME: add result assertions (find_files, parsing PAGE etc)
+    analyse_result(ws, 'TextLine')
+
diff --git a/tests/test_recognize.py b/tests/test_recognize.py
@@ -1,6 +1,12 @@
 # pylint: disable=import-error
 
+import os
+
 from ocrd import run_processor
+from ocrd_utils import MIMETYPE_PAGE
+from ocrd_models.constants import NAMESPACES
+from ocrd_modelfactory import page_from_file
+
 from ocrd_kraken.recognize import KrakenRecognize
 from ocrd_kraken.binarize import KrakenBinarize
 
@@ -21,4 +27,10 @@ def test_recognize(workspace_aufklaerung):
     )
     ws = workspace_aufklaerung['workspace']
     ws.save_mets()
-    # FIXME: add result assertions (find_files, parsing PAGE etc)
+    assert os.path.isdir(os.path.join(ws.directory, 'OCR-D-OCR-KRAKEN'))
+    results = ws.find_files(file_grp='OCR-D-OCR-KRAKEN', mimetype=MIMETYPE_PAGE)
+    result0 = next(results, False)
+    assert result0, "found no output PAGE file"
+    result0 = page_from_file(result0)
+    text0 = result0.etree.xpath('//page:Glyph/page:TextEquiv/page:Unicode', namespaces=NAMESPACES)
+    assert len(text0) > 0, "found no glyph text in output PAGE file"
diff --git a/tests/test_segment.py b/tests/test_segment.py
@@ -1,10 +1,27 @@
 # pylint: disable=import-error
 
+import os
+
 from ocrd import run_processor
+from ocrd_utils import MIMETYPE_PAGE
+from ocrd_models.constants import NAMESPACES
+from ocrd_modelfactory import page_from_file
+
 from ocrd_kraken.segment import KrakenSegment
 from ocrd_kraken.binarize import KrakenBinarize
 
 
+def analyse_result(ws):
+    assert os.path.isdir(os.path.join(ws.directory, 'OCR-D-SEG-LINE-KRAKEN'))
+    out_files = list(ws.find_files(fileGrp="OCR-D-SEG-LINE-KRAKEN", mimetype=MIMETYPE_PAGE))
+    assert len(out_files), "found no output PAGE file"
+    out_pcgts = page_from_file(out_files[0])
+    assert out_pcgts is not None
+    out_regions = out_pcgts.etree.xpath('//page:TextRegion/page:Coords', namespaces=NAMESPACES)
+    assert len(out_regions) > 0, "found no text regions in output PAGE file"
+    out_lines = out_pcgts.get_Page().get_AllTextLines()
+    assert len(out_lines), "found no text lines in output PAGE file"
+
 def test_run_blla(workspace_aufklaerung):
     run_processor(KrakenSegment,
                   input_file_grp="OCR-D-IMG",
@@ -14,7 +31,7 @@ def test_run_blla(workspace_aufklaerung):
     )
     ws = workspace_aufklaerung['workspace']
     ws.save_mets()
-    # FIXME: add result assertions (find_files, parsing PAGE etc)
+    analyse_result(ws)
 
 def test_run_blla_regionlevel(workspace_aufklaerung_region):
     run_processor(KrakenSegment,
@@ -27,7 +44,7 @@ def test_run_blla_regionlevel(workspace_aufklaerung_region):
     )
     ws = workspace_aufklaerung_region['workspace']
     ws.save_mets()
-    # FIXME: add result assertions (find_files, parsing PAGE etc)
+    analyse_result(ws)
 
 def test_run_legacy(workspace_aufklaerung):
     # legacy segmentation requires binarized images
@@ -45,4 +62,4 @@ def test_run_legacy(workspace_aufklaerung):
     )
     ws = workspace_aufklaerung['workspace']
     ws.save_mets()
-    # FIXME: add result assertions (find_files, parsing PAGE etc)
+    analyse_result(ws)