feat: PDF 파서 추가

src/ureca_document_parser/pdf/__init__.py

@@ -0,0 +1,5 @@
+"""PDF parser using pymupdf."""
+
+from .parser import PdfParser
+
+__all__ = ["PdfParser"]

src/ureca_document_parser/pdf/parser.py

@@ -0,0 +1,90 @@
+"""PDF parser implementation using pymupdf (fitz)."""
+
+from __future__ import annotations
+
+from pathlib import Path
+from typing import TYPE_CHECKING
+
+from ..models import Document, Metadata, Paragraph, ParseError
+
+if TYPE_CHECKING:
+    from os import PathLike
+
+
+class PdfParser:
+    """Parser for PDF files using pymupdf.
+
+    Extracts text content from PDF files and converts to Document model.
+    """
+
+    @staticmethod
+    def can_parse(file_path: str | PathLike[str]) -> bool:
+        """Check if file can be parsed as PDF."""
+        path = Path(file_path)
+        return path.suffix.lower() == ".pdf"
+
+    @staticmethod
+    def parse(file_path: str | PathLike[str]) -> Document:
+        """Parse PDF file and return Document.
+
+        Args:
+            file_path: Path to PDF file
+
+        Returns:
+            Document with extracted content
+
+        Raises:
+            ParseError: If file cannot be parsed
+        """
+        try:
+            import fitz  # pymupdf
+        except ImportError:
+            raise ParseError(
+                "PDF support requires pymupdf. "
+                "Install with: pip install ureca_document_parser[pdf]"
+            ) from None
+
+        path = Path(file_path)
+
+        try:
+            # Open PDF
+            doc = fitz.open(path)
+
+            # Extract metadata
+            metadata = Metadata(
+                title=doc.metadata.get("title") or path.stem,
+                author=doc.metadata.get("author") or "",
+                source_format="pdf",
+                extra={
+                    "pages": doc.page_count,
+                    "producer": doc.metadata.get("producer", ""),
+                    "creator": doc.metadata.get("creator", ""),
+                },
+            )
+
+            # Extract text from each page
+            elements = []
+            for page in doc:
+                text = page.get_text()
+
+                # Split by paragraphs (double newline)
+                paragraphs = text.split("\n\n")
+
+                for para_text in paragraphs:
+                    # Clean up whitespace
+                    cleaned = " ".join(para_text.split())
+
+                    if cleaned:  # Skip empty paragraphs
+                        elements.append(
+                            Paragraph(
+                                text=cleaned,
+                                heading_level=0,
+                            )
+                        )
+
+            doc.close()
+
+            return Document(elements=elements, metadata=metadata)
+
+        except Exception as e:
+            raise ParseError(f"Failed to parse PDF: {e}") from e