feat: pdf reader

backend/apps/books/services/process_ebook.py

@@ -89,7 +89,69 @@ def _process_epub(file_path: str) -> dict[str, Any]:
 
 
 def _process_pdf(file_path: str) -> dict[str, Any]:
-    return {"format": "pdf", "page_count": 0, "metadata": {}, "toc": []}
+    from pypdf import PdfReader
+
+    reader = PdfReader(file_path)
+    page_count = len(reader.pages)
+    metadata = _extract_pdf_metadata(reader)
+    toc = _extract_pdf_outline(reader)
+    if not toc and page_count > 0:
+        toc = [
+            {"title": f"Page {i}", "href": f"pdf:page:{i}", "children": []}
+            for i in range(1, page_count + 1)
+        ]
+    return {
+        "format": "pdf",
+        "page_count": page_count,
+        "metadata": metadata,
+        "toc": toc,
+    }
+
+
+def _extract_pdf_metadata(reader) -> dict[str, Any]:
+    metadata: dict[str, Any] = {}
+    info = reader.metadata
+    if not info:
+        return metadata
+    title = getattr(info, "title", None)
+    author = getattr(info, "author", None)
+    if title:
+        metadata["title"] = str(title)
+    if author:
+        metadata["author"] = str(author)
+    return metadata
+
+
+def _extract_pdf_outline(reader) -> list[dict[str, Any]]:
+    outline = getattr(reader, "outline", None)
+    if not outline:
+        return []
+    return _walk_pdf_outline(reader, outline)
+
+
+def _walk_pdf_outline(reader, outline: list[Any]) -> list[dict[str, Any]]:
+    entries: list[dict[str, Any]] = []
+    i = 0
+    while i < len(outline):
+        item = outline[i]
+        if isinstance(item, list):
+            if entries:
+                entries[-1]["children"] = _walk_pdf_outline(reader, item)
+            i += 1
+            continue
+        title = getattr(item, "title", None) or "Section"
+        page_num = 1
+        try:
+            page_num = reader.get_destination_page_number(item) + 1
+        except Exception:
+            logger.debug("Could not resolve PDF outline destination", exc_info=True)
+        entries.append({
+            "title": str(title),
+            "href": f"pdf:page:{page_num}",
+            "children": [],
+        })
+        i += 1
+    return entries
 
 
 def _extract_epub_metadata(book) -> dict[str, Any]: