src/convert_to_pdf.py

"""
Markdown raporlarını PDF formatına çevirir.
Kullanım:
    python convert_to_pdf.py <md_dosyasi>
    python convert_to_pdf.py test_outputs/report.md
    python convert_to_pdf.py  # test_outputs/ klasöründeki tüm .md dosyalarını çevirir
"""

import sys
import os
import io
import re
import base64
from pathlib import Path
import markdown
from xhtml2pdf import pisa

IMAGE_PATTERN = re.compile(
    r'(\*\*Image\*\*:\s*)([^\n]+\.(jpg|jpeg|png|gif|bmp|webp))',
    re.IGNORECASE,
)

IMG_CSS = """
.report-image {
    display: block;
    max-width: 100%;
    max-height: 300px;
    margin: 10px 0 16px 0;
    border: 1px solid #ccc;
    border-radius: 4px;
}
.image-label {
    font-weight: bold;
    color: #003366;
    margin-bottom: 4px;
}
"""


def _image_to_base64(img_path: str) -> str | None:
    path = Path(img_path.strip())
    if not path.exists():
        return None
    mime = {".jpg": "image/jpeg", ".jpeg": "image/jpeg",
            ".png": "image/png", ".gif": "image/gif",
            ".bmp": "image/bmp", ".webp": "image/webp"}.get(path.suffix.lower(), "image/jpeg")
    data = base64.b64encode(path.read_bytes()).decode()
    return f"data:{mime};base64,{data}"


def _embed_images(md_text: str) -> str:
    def replacer(m: re.Match) -> str:
        img_path = m.group(2).strip()
        data_uri = _image_to_base64(img_path)
        if data_uri:
            return (
                f'<p class="image-label">Image</p>'
                f'<img class="report-image" src="{data_uri}" alt="Detection Image">'
            )
        return m.group(0)

    return IMAGE_PATTERN.sub(replacer, md_text)


CSS = """
@page {
    size: A4;
    margin: 2cm 2.5cm 2cm 2.5cm;
}

body {
    font-family: "Helvetica", "Arial", sans-serif;
    font-size: 10pt;
    line-height: 1.6;
    color: #1a1a1a;
}

h1 {
    font-size: 16pt;
    color: #003366;
    border-bottom: 2px solid #003366;
    padding-bottom: 6px;
    margin-bottom: 12px;
}

h2 {
    font-size: 13pt;
    color: #003366;
    margin-top: 18px;
    margin-bottom: 6px;
}

h3 {
    font-size: 11pt;
    color: #1a4d80;
    margin-top: 12px;
    margin-bottom: 4px;
}

p {
    margin: 6px 0;
}

strong {
    color: #111;
}

em {
    color: #333;
}

ul, ol {
    margin: 6px 0 6px 20px;
    padding: 0;
}

li {
    margin-bottom: 3px;
}

hr {
    border: none;
    border-top: 1px solid #cccccc;
    margin: 14px 0;
}

code {
    background: #f4f4f4;
    padding: 1px 4px;
    font-size: 9pt;
    border-radius: 3px;
}

blockquote {
    border-left: 3px solid #003366;
    margin: 8px 0 8px 16px;
    padding-left: 10px;
    color: #555;
}

.header-meta {
    font-size: 9pt;
    color: #666;
    margin-bottom: 16px;
}

table {
    width: 100%;
    border-collapse: collapse;
    margin: 10px 0;
    font-size: 9pt;
}

th {
    background: #003366;
    color: white;
    padding: 5px 8px;
    text-align: left;
}

td {
    border: 1px solid #ccc;
    padding: 4px 8px;
}

tr:nth-child(even) td {
    background: #f5f8ff;
}
"""


def _resize_b64_image(b64: str, max_size: int = 800) -> str:
    from PIL import Image as PILImage
    data = base64.b64decode(b64)
    img = PILImage.open(io.BytesIO(data))
    img.thumbnail((max_size, max_size), PILImage.LANCZOS)
    buf = io.BytesIO()
    img.save(buf, format="PNG")
    return base64.b64encode(buf.getvalue()).decode()


def md_text_to_pdf_bytes(md_text: str, annotated_image_b64: str | None = None) -> bytes:
    """
    Markdown metnini doğrudan PDF baytlarına çevirir (dosya I/O yok).
    annotated_image_b64: base64 kodlanmış PNG — rapor başına eklenir.
    """
    if annotated_image_b64:
        annotated_image_b64 = _resize_b64_image(annotated_image_b64)
        image_block = (
            '<p class="image-label">Detection Image</p>'
            f'<img class="report-image" src="data:image/png;base64,{annotated_image_b64}" alt="Detection Image">'
        )
    else:
        image_block = ""

    html_body = markdown.markdown(
        md_text,
        extensions=["tables", "fenced_code", "nl2br", "sane_lists"],
    )

    full_html = f"""<!DOCTYPE html>
<html lang="tr">
<head>
<meta charset="utf-8">
<style>{CSS}{IMG_CSS}</style>
</head>
<body>
{image_block}
{html_body}
</body>
</html>"""

    buf = io.BytesIO()
    result = pisa.CreatePDF(full_html, dest=buf, encoding="utf-8")
    if result.err:
        raise RuntimeError(f"PDF oluşturulurken hata: {result.err}")
    return buf.getvalue()


def md_to_pdf(md_path: str | Path, pdf_path: str | Path | None = None) -> Path:
    md_path = Path(md_path)
    if not md_path.exists():
        raise FileNotFoundError(f"Dosya bulunamadı: {md_path}")

    if pdf_path is None:
        pdf_path = md_path.with_suffix(".pdf")
    pdf_path = Path(pdf_path)
    pdf_path.parent.mkdir(parents=True, exist_ok=True)

    md_text = md_path.read_text(encoding="utf-8")

    # Görsel yollarını base64 gömülü <img> etiketlerine çevir
    preprocessed = _embed_images(md_text)

    html_body = markdown.markdown(
        preprocessed,
        extensions=["tables", "fenced_code", "nl2br", "sane_lists"],
    )

    full_html = f"""<!DOCTYPE html>
<html lang="tr">
<head>
<meta charset="utf-8">
<style>{CSS}{IMG_CSS}</style>
</head>
<body>
{html_body}
</body>
</html>"""

    with open(pdf_path, "wb") as f:
        result = pisa.CreatePDF(full_html, dest=f, encoding="utf-8")

    if result.err:
        raise RuntimeError(f"PDF oluşturulurken hata: {result.err}")

    print(f"PDF oluşturuldu: {pdf_path}")
    return pdf_path


def convert_all(directory: str | Path = "test_outputs") -> list[Path]:
    directory = Path(directory)
    md_files = list(directory.glob("**/*.md"))

    if not md_files:
        print(f"'{directory}' klasöründe .md dosyası bulunamadı.")
        return []

    results = []
    for md_file in md_files:
        try:
            pdf = md_to_pdf(md_file)
            results.append(pdf)
        except Exception as e:
            print(f"Hata ({md_file.name}): {e}")

    return results


if __name__ == "__main__":
    if len(sys.argv) == 1:
        convert_all("test_outputs")
    elif len(sys.argv) == 2:
        target = sys.argv[1]
        path = Path(target)
        if path.is_dir():
            convert_all(path)
        elif path.suffix == ".md":
            md_to_pdf(path)
        else:
            print(f"Geçersiz dosya: {target}  (.md dosyası veya klasör belirtin)")
            sys.exit(1)
    elif len(sys.argv) == 3:
        md_to_pdf(sys.argv[1], sys.argv[2])
    else:
        print("Kullanım: python convert_to_pdf.py [md_dosyasi] [cikti.pdf]")
        sys.exit(1)