Lots of added fixtures

Signed-off-by: Christoph Auer <cau@zurich.ibm.com>

Author: cau-git

docling_cvat_tools/cvat_tools/cvat_to_docling.py

@@ -2987,7 +2987,7 @@ def load_document_pages(
         )
         from docling.datamodel.base_models import InputFormat
         from docling.datamodel.document import InputDocument
-        from docling.models.page_preprocessing_model import (
+        from docling.models.stages.page_preprocessing.page_preprocessing_model import (
             PagePreprocessingModel,
             PagePreprocessingOptions,
         )

tests/cvat_to_docling/fixtures/002a_table_order/metadata.json

@@ -2,8 +2,8 @@
   "test_id": "002",
   "name": "002a_table_order",
   "description": "Tables should be included in reading order correctly. Footnotes should appear once.",
-  "observation": "Footnotes are replicated on top and after the table as captions, because the XML annotation uses a to_caption link for the footnote. We should tolerate this error.",
-  "observation_status": "broken",
+  "observation": "Footnotes are correctly placed. They are not visible in the HTML serialization.",
+  "observation_status": "correct",
   "source": {
     "image_identifier": "doc_fade275d3b925d67abc61f56f5bb414c1cab6c3e94c5953cb9cfa8a6f7a9853d_page_000001.png"
   },

tests/cvat_to_docling/fixtures/008_kvp_in_picture_legend/expected.json

@@ -2,8 +2,8 @@
   "test_id": "009",
   "name": "009_rotation_on_table",
   "description": "Rotated table should be correctly rebuilt.",
-  "observation": "The rotated table is garbage.",
-  "observation_status": "broken",
+  "observation": "The rotated table is correctly constructed.",
+  "observation_status": "correct.",
   "source": {
     "image_identifier": "doc_a02617dc490954fe59c8382c7e8dc6203bf230c56b156c7778e09b31f1d43f2b_page_000001.png"
   },

tests/cvat_to_docling/fixtures/009_rotation_on_table/expected.json

@@ -2,8 +2,8 @@
   "test_id": "011",
   "name": "011a_paragraph_continuation_with_picture_between",
   "description": "Paragraph must merge across column wrap, with picture and caption placed afterwards.",
-  "observation": "The paragraph is cut up, and also all paragraphs before the first figure are concatenated to the list that starts at the top.",
-  "observation_status": "broken",
+  "observation": "Correct ordering. Formula transcription is bad from the PDF.",
+  "observation_status": "correct",
   "source": {
     "image_identifier": "doc_f0ac80935cb386d734bee68e317b47050819ab112dadb25fbf78266d3918fbd2_page_000001.png"
   },

tests/cvat_to_docling/fixtures/009_rotation_on_table/metadata.json

@@ -267,8 +267,8 @@
             "coord_origin": "BOTTOMLEFT"
           },
           "charspan": [
-            228,
-            437
+            0,
+            209
           ]
         }
       ],

tests/cvat_to_docling/fixtures/011a_paragraph_continuation_with_picture_between/expected.json

@@ -2,7 +2,7 @@
   "test_id": "013",
   "name": "bad_handling_of_footnote_vs_list_and_merges_inside",
   "description": "The footnote list with sandwiched text must be handled in correct reading order.",
-  "observation": "Observing out-of-order elements in the footnotes.",
+  "observation": "Observing out-of-order elements in the footnotes, some elements lost.",
   "observation_status": "broken",
   "source": {
     "image_identifier": "doc_a4b6343889cd5469d7957e1fcfd97be1299a75c865e12779001180c790620801_page_000001.png"