FunMD_Convert/docling/app/services/pdf_converter.py

"""
PDF Conversion Service
Provides Word-to-PDF and Markdown-to-PDF conversion functionality
"""
from pathlib import Path
from typing import Optional, Dict, Tuple
import io
import tempfile
import logging

from docx import Document as DocxDocument

from app.services.word2markdown import convert_any as convert_word_to_md
from app.services.docling_adapter import md_to_pdf_bytes_with_renderer


logger = logging.getLogger(__name__)


def word_to_pdf_bytes(
    file_path: str | Path,
    toc: bool = False,
    header_text: Optional[str] = None,
    footer_text: Optional[str] = None,
    logo_url: Optional[str] = None,
    copyright_text: Optional[str] = None,
    filename_text: Optional[str] = None,
    cover_src: Optional[str] = None,
    product_name: Optional[str] = None,
    document_name: Optional[str] = None,
    product_version: Optional[str] = None,
    document_version: Optional[str] = None,
) -> bytes:
    """
    Convert Word document (.doc, .docx) to PDF

    Args:
        file_path: Path to Word file
        toc: Enable table of contents
        header_text: Custom header text
        footer_text: Custom footer text
        logo_url: URL to logo image
        copyright_text: Copyright notice
        filename_text: Filename to display
        cover_src: Cover image source
        product_name: Product name for cover
        document_name: Document name for cover
        product_version: Product version
        document_version: Document version

    Returns:
        PDF file as bytes
    """
    logger.info(f"Converting Word to PDF: {file_path}")

    # Convert Word to Markdown first
    path = Path(file_path)
    _, markdown_content = convert_word_to_md(path)

    # Then convert Markdown to PDF
    pdf_bytes = md_to_pdf_bytes_with_renderer(
        md=markdown_content,
        renderer="weasyprint",
        toc=toc,
        header_text=header_text,
        footer_text=footer_text,
        logo_url=logo_url,
        copyright_text=copyright_text,
        filename_text=filename_text or path.stem,
        cover_src=cover_src,
        product_name=product_name,
        document_name=document_name,
        product_version=product_version,
        document_version=document_version,
    )

    logger.info(f"Word to PDF conversion complete: {len(pdf_bytes)} bytes")
    return pdf_bytes


def markdown_to_pdf_bytes(
    markdown_content: str,
    toc: bool = False,
    header_text: Optional[str] = None,
    footer_text: Optional[str] = None,
    logo_url: Optional[str] = None,
    copyright_text: Optional[str] = None,
    filename_text: Optional[str] = None,
    cover_src: Optional[str] = None,
    product_name: Optional[str] = None,
    document_name: Optional[str] = None,
    product_version: Optional[str] = None,
    document_version: Optional[str] = None,
    css_name: Optional[str] = None,
    css_text: Optional[str] = None,
) -> bytes:
    """
    Convert Markdown content to PDF

    Args:
        markdown_content: Markdown text content
        toc: Enable table of contents
        header_text: Custom header text
        footer_text: Custom footer text
        logo_url: URL to logo image
        copyright_text: Copyright notice
        filename_text: Filename to display
        cover_src: Cover image source
        product_name: Product name for cover
        document_name: Document name for cover
        product_version: Product version
        document_version: Document version
        css_name: Name of CSS file in configs/styles
        css_text: Custom CSS as string

    Returns:
        PDF file as bytes
    """
    logger.info("Converting Markdown to PDF")

    pdf_bytes = md_to_pdf_bytes_with_renderer(
        md=markdown_content,
        renderer="weasyprint",
        css_name=css_name,
        css_text=css_text,
        toc=toc,
        header_text=header_text,
        footer_text=footer_text,
        logo_url=logo_url,
        copyright_text=copyright_text,
        filename_text=filename_text,
        cover_src=cover_src,
        product_name=product_name,
        document_name=document_name,
        product_version=product_version,
        document_version=document_version,
    )

    logger.info(f"Markdown to PDF conversion complete: {len(pdf_bytes)} bytes")
    return pdf_bytes


def markdown_file_to_pdf_bytes(
    file_path: str | Path,
    encoding: str = "utf-8",
    **kwargs
) -> bytes:
    """
    Convert Markdown file to PDF

    Args:
        file_path: Path to Markdown file
        encoding: File encoding (default: utf-8)
        **kwargs: Additional arguments passed to markdown_to_pdf_bytes

    Returns:
        PDF file as bytes
    """
    path = Path(file_path)
    markdown_content = path.read_text(encoding=encoding)

    # Set default filename from file path if not provided
    if "filename_text" not in kwargs or not kwargs.get("filename_text"):
        kwargs["filename_text"] = path.stem

    return markdown_to_pdf_bytes(markdown_content, **kwargs)


def read_file_content(file_path: str | Path) -> Tuple[str, bytes]:
    """
    Read file content and detect content type

    Args:
        file_path: Path to file

    Returns:
        Tuple of (detected_type, content_bytes)
    """
    path = Path(file_path)
    if not path.exists():
        raise FileNotFoundError(f"File not found: {file_path}")

    content_bytes = path.read_bytes()

    # Detect by extension
    ext = path.suffix.lower()
    if ext in {".md", ".markdown"}:
        return "markdown", content_bytes
    elif ext in {".doc", ".docx"}:
        return "word", content_bytes
    elif ext in {".txt"}:
        return "text", content_bytes
    else:
        # Try to detect by content
        content_start = content_bytes[:8]
        if content_start.startswith(b"PK\x03\x04"):
            return "word", content_bytes
        return "text", content_bytes