reformatted and fixed bugs

Signed-off-by: Peter Staar <[email protected]>

docling_core/types/doc/base.py

@@ -10,6 +10,7 @@
 from pydantic import BaseModel, Field, PositiveInt, StrictStr
 
 from docling_core.search.mapping import es_field
+from docling_core.types.doc.tokens import DocumentToken
 from docling_core.utils.alias import AliasModel
 
 CellData = tuple[float, float, float, float, str, str]
@@ -249,31 +250,29 @@ def export_to_html(self) -> str:
 
         return body
 
-    def export_to_document_tokens(self, new_line:str="\n", loc_str:str=""):
-
+    def export_to_document_tokens(self, new_line: str = "\n", loc_str: str = ""):
+        """Export table to document tokens format."""
         body = ""
-        
+
         body += f"{DocumentToken.BEG_TABLE.value}{loc_str}"
 
         if self.text is not None and len(self.text) > 0:
             body += f"{DocumentToken.BEG_CAPTION.value}"
-            body += (
-                f"{self.text}{DocumentToken.END_CAPTION.value}{new_line}"
-            )
+            body += f"{self.text}{DocumentToken.END_CAPTION.value}{new_line}"
 
         if self.data is not None and len(self.data) > 0:
             for i, row in enumerate(self.data):
                 body += f"<row_{i}>"
                 for j, col in enumerate(row):
                     text = col.text
                     body += f"<col_{j}>{text}</col_{j}>"
-                    
+
                 body += f"</row_{i}>{new_line}"
 
         body += f"{DocumentToken.BEG_TABLE.value}{new_line}"
 
         return body
-        
+
 
 # FIXME: let's add some figure specific data-types later
 class Figure(BaseCell):

docling_core/types/doc/document.py

@@ -6,11 +6,8 @@
 """Models for the Docling Document data type."""
 
 from datetime import datetime
-from enum import Enum
 from typing import Generic, Optional, Tuple, Union
 
-
-
 from pydantic import (
     AnyHttpUrl,
     BaseModel,
@@ -34,9 +31,6 @@
     LanguageT,
     Log,
 )
-
-from docling_core.types.doc.tokens import DocumentToken
-
 from docling_core.types.doc.base import (
     BaseCell,
     BaseText,
@@ -48,6 +42,7 @@
     S3Data,
     Table,
 )
+from docling_core.types.doc.tokens import DocumentToken
 from docling_core.utils.alias import AliasModel
 
 

docling_core/types/doc/tokens.py

@@ -6,6 +6,8 @@
 """Tokens used in the docling document model."""
 
 from enum import Enum
+from typing import Tuple
+
 
 class DocumentToken(Enum):
     """Class to represent an LLM friendly representation of a Document."""
@@ -104,7 +106,7 @@ def get_col_token(col: int, beg=bool) -> str:
             return f"<col_{col}>"
         else:
             return f"</col_{col}>"
-        
+
     @staticmethod
     def get_page_token(page: int):
         """Function to get page tokens."""
@@ -122,5 +124,3 @@ def get_location_token(val: float, rnorm: int = 100):
             return f"<loc_{rnorm}>"
 
         return f"<loc_{val_}>"
-
-

docling_core/types/rec/statement.py

@@ -21,6 +21,7 @@
 from docling_core.types.rec.attribute import Attribute
 from docling_core.types.rec.subject import Subject
 
+
 class StatementToken(Enum):
     """Class to represent an LLM friendly representation of statements."""