feat(gdocs): extract smart chips (date, person, rich link) when reading documents

gdocs/docs_markdown.py

@@ -122,9 +122,36 @@ def _convert_paragraph_text(
     for elem in para.get("elements", []):
         if "textRun" in elem:
             parts.append(_convert_text_run(elem["textRun"], skip_strikethrough))
+        elif "richLink" in elem:
+            parts.append(_convert_rich_link(elem["richLink"]))
+        elif "person" in elem:
+            parts.append(_convert_person(elem["person"]))
+        elif "dateElement" in elem:
+            props = elem["dateElement"].get("dateElementProperties", {})
+            parts.append(props.get("displayText", ""))
     return "".join(parts).strip()
 
 
+def _convert_rich_link(rich_link: dict[str, Any]) -> str:
+    """Convert a richLink element (URL chip) to markdown."""
+    props = rich_link.get("richLinkProperties", {})
+    title = props.get("title", "")
+    uri = props.get("uri", "")
+    if title and uri:
+        return f"[{title}]({uri})"
+    return title or uri
+
+
+def _convert_person(person: dict[str, Any]) -> str:
+    """Convert a person element (person chip) to markdown."""
+    props = person.get("personProperties", {})
+    name = props.get("name", "")
+    email = props.get("email", "")
+    if name and email:
+        return f"{name} ({email})"
+    return name or email
+
+
 def _convert_text_run(
     text_run: dict[str, Any], skip_strikethrough: bool = False
 ) -> str:

gdocs/docs_structure.py

@@ -92,7 +92,7 @@ def _parse_element(element: dict[str, Any]) -> Optional[dict[str, Any]]:
     if "paragraph" in element:
         paragraph = element["paragraph"]
         element_info["type"] = "paragraph"
-        element_info["text"] = _extract_paragraph_text(paragraph)
+        element_info["text"] = extract_paragraph_text(paragraph)
         element_info["style"] = paragraph.get("paragraphStyle", {})
 
     elif "table" in element:
@@ -162,12 +162,31 @@ def _parse_table_cells(table: dict[str, Any]) -> list[list[dict[str, Any]]]:
     return cells
 
 
-def _extract_paragraph_text(paragraph: dict[str, Any]) -> str:
+def extract_paragraph_text(paragraph: dict[str, Any]) -> str:
     """Extract text from a paragraph element."""
     text_parts = []
     for element in paragraph.get("elements", []):
         if "textRun" in element:
             text_parts.append(element["textRun"].get("content", ""))
+        elif "dateElement" in element:
+            props = element["dateElement"].get("dateElementProperties", {})
+            text_parts.append(props.get("displayText", ""))
+        elif "richLink" in element:
+            props = element["richLink"].get("richLinkProperties", {})
+            title = props.get("title", "")
+            uri = props.get("uri", "")
+            if title and uri:
+                text_parts.append(f"{title} ({uri})")
+            else:
+                text_parts.append(title or uri)
+        elif "person" in element:
+            props = element["person"].get("personProperties", {})
+            name = props.get("name", "")
+            email = props.get("email", "")
+            if name and email:
+                text_parts.append(f"{name} ({email})")
+            else:
+                text_parts.append(name or email)
     return "".join(text_parts)
 
 
@@ -176,7 +195,7 @@ def _extract_cell_text(cell: dict[str, Any]) -> str:
     text_parts = []
     for element in cell.get("content", []):
         if "paragraph" in element:
-            text_parts.append(_extract_paragraph_text(element["paragraph"]))
+            text_parts.append(extract_paragraph_text(element["paragraph"]))
     return "".join(text_parts)
 
 

gdocs/docs_tools.py

@@ -41,6 +41,7 @@
     parse_document_structure,
     find_tables,
     analyze_document_complexity,
+    extract_paragraph_text,
 )
 from gdocs.docs_tables import extract_table_as_data
 from gdocs.docs_markdown import (
@@ -193,13 +194,7 @@ def extract_text_from_elements(elements, tab_name=None, tab_id=None, depth=0):
 
             for element in elements:
                 if "paragraph" in element:
-                    paragraph = element.get("paragraph", {})
-                    para_elements = paragraph.get("elements", [])
-                    current_line_text = ""
-                    for pe in para_elements:
-                        text_run = pe.get("textRun", {})
-                        if text_run and "content" in text_run:
-                            current_line_text += text_run["content"]
+                    current_line_text = extract_paragraph_text(element["paragraph"])
                     if current_line_text.strip():
                         text_lines.append(current_line_text)
                 elif "table" in element: