fix: markdown serialization of hyperlink with code (#434)

* chore: leverage fixtures in tests and add type annotations

Use fixtures for sample DoclingDocument in tests.
Add missing function type annotation when the body contains annotations.

Signed-off-by: Cesar Berrospi Ramis <[email protected]>

* fix(markdown): serialize hyperlink as code always with single backticks

Signed-off-by: Cesar Berrospi Ramis <[email protected]>

---------

Signed-off-by: Cesar Berrospi Ramis <[email protected]>

Author: ceberam

docling_core/transforms/chunker/code_chunking/_language_code_chunkers.py

@@ -38,7 +38,7 @@ def _new_hash(code: str) -> int:
 class _RangeTracker:
     """Handles tracking and management of used byte ranges in code."""
 
-    def __init__(self):
+    def __init__(self) -> None:
         """Initialize the range tracker with an empty list of used ranges."""
         self.used_ranges: List[Tuple[int, int]] = []
 

docling_core/transforms/serializer/markdown.py

@@ -211,7 +211,9 @@ def serialize(
                 text_part = f"{num_hashes * '#'} {text}"
         elif isinstance(item, CodeItem):
             if params.format_code_blocks:
-                text_part = f"`{text}`" if is_inline_scope else f"```\n{text}\n```"
+                # inline items and all hyperlinks: use single backticks
+                bt = is_inline_scope or (params.include_hyperlinks and item.hyperlink)
+                text_part = f"`{text}`" if bt else f"```\n{text}\n```"
             else:
                 text_part = text
             escape_html = False

docling_core/types/doc/document.py

@@ -6138,7 +6138,7 @@ def check_version_is_compatible(cls, v: str) -> str:
         else:
             return CURRENT_VERSION
 
-    @model_validator(mode="after")  # type: ignore
+    @model_validator(mode="after")
     def validate_document(self) -> Self:
         """validate_document."""
         with warnings.catch_warnings():
@@ -6152,7 +6152,7 @@ def validate_document(self) -> Self:
         return self
 
     @model_validator(mode="after")
-    def validate_misplaced_list_items(self):
+    def validate_misplaced_list_items(self) -> Self:
         """validate_misplaced_list_items."""
         # find list items without list parent, putting succesive ones together
         misplaced_list_items: list[list[ListItem]] = []
@@ -6184,7 +6184,7 @@ def validate_misplaced_list_items(self):
             )
 
             # delete list items from document (should not be affected by group addition)
-            self.delete_items(node_items=curr_list_items)
+            self.delete_items(node_items=list(curr_list_items))
 
             # add list items to new group
             for li in curr_list_items: