recipes/bin/task3-frontmatter.py

#!/tmp/recipe-venv/bin/python3
"""Add frontmatter and images to 9 recipes without frontmatter."""

import re, sys, time
from pathlib import Path

import requests
from bs4 import BeautifulSoup
from ddgs import DDGS

WORKDIR = Path("/home/jbrechtel/recipes/.worktrees/recipe-images")
HEADERS = {"User-Agent": "Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36"}

# (filename, slug, title, source_url_or_empty)
recipes = [
    ("Black Pepper Tofu (Ottolenghi).cook", "black-pepper-tofu-ottolenghi", "Black Pepper Tofu (Ottolenghi)", ""),
    ("Black Pepper Tofu with Bok Choy.cook", "black-pepper-tofu-with-bok-choy", "Black Pepper Tofu with Bok Choy", ""),
    ("butternut-turkey-chili.cook", "butternut-turkey-chili", "Butternut & Turkey Chili", ""),
    ("Caramel Cake.cook", "caramel-cake", "Caramel Cake", ""),
    ("Caramel Icing.cook", "caramel-icing", "Caramel Icing", ""),
    ("example-fried-rice.cook", "example-fried-rice", "Fried Rice", ""),
    ("indian-spiced-chicken.cook", "indian-spiced-chicken", "Indian Spiced Chicken", ""),
    ("Italian Sausage and Delicata Rigatoni.cook", "italian-sausage-and-delicata-rigatoni", "Italian Sausage & Delicata Rigatoni", ""),
    ("shrimp-feta.cook", "shrimp-feta", "Orzo with shrimp, tomato and marinated feta", "https://thehappyfoodie.co.uk/recipes/ottolenghis-orzo-with-prawns-tomato-and-marinated-feta/"),
]


def fetch_og_image(url: str) -> str | None:
    try:
        resp = requests.get(url, headers=HEADERS, timeout=15)
        resp.raise_for_status()
    except:
        return None
    soup = BeautifulSoup(resp.text, "html.parser")
    for meta in soup.find_all("meta"):
        c = meta.get("content", "")
        p = (meta.get("property") or "").lower()
        n = (meta.get("name") or "").lower()
        if ("og:image" in p or "og:image" in n or "twitter:image" in n) and c:
            return c
    # JSON-LD
    for script in soup.find_all("script", type="application/ld+json"):
        try:
            import json
            data = json.loads(script.string)
            items = data if isinstance(data, list) else [data]
            for item in items:
                if isinstance(item, dict) and item.get("@type") == "Recipe":
                    img = item.get("image")
                    if isinstance(img, str): return img
                    if isinstance(img, list):
                        for i in img:
                            if isinstance(i, str): return i
                            if isinstance(i, dict):
                                u = i.get("contentUrl") or i.get("url", "")
                                if u: return u
                    if isinstance(img, dict):
                        u = img.get("contentUrl") or img.get("url", "")
                        if u: return u
        except:
            pass
    return None


def search_and_scrape(query: str) -> str | None:
    try:
        results = list(DDGS().text(query, max_results=5))
    except:
        return None
    for r in results:
        url = r.get("href", "")
        if not url or "pinterest" in url or "facebook" in url:
            continue
        img = fetch_og_image(url)
        if img:
            return img
        time.sleep(0.3)
    return None


def download_image(url: str, dest: Path) -> bool:
    try:
        resp = requests.get(url, headers=HEADERS, timeout=30)
        resp.raise_for_status()
        ct = resp.headers.get("content-type", "")
        if not ct.startswith("image/"):
            return False
        content = resp.content
        if len(content) < 5000:
            return False
        dest.write_bytes(content)
        return True
    except:
        return False


def add_frontmatter(filepath: Path, slug: str, title: str):
    url = f"https://roux.roo.lol/recipe-images/{slug}.webp"
    content = filepath.read_text(encoding="utf-8")

    frontmatter = f"""---
title: {title}
description: 
servings: 1
category: uncategorized
source: Unknown
image: {url}
---

"""
    filepath.write_text(frontmatter + content, encoding="utf-8")
    print(f"  Added frontmatter to {filepath.name}", file=sys.stderr)


def main():
    for filename, slug, title, src_url in recipes:
        fp = WORKDIR / filename
        dest = WORKDIR / "recipe-images" / f"{slug}.webp"

        print(f"[{slug}] Processing {filename}", file=sys.stderr)

        # Find and download image
        img_url = None
        if src_url:
            print(f"  Trying source URL: {src_url}", file=sys.stderr)
            img_url = fetch_og_image(src_url)
            time.sleep(0.3)

        if not img_url:
            print(f"  Searching: {title} recipe", file=sys.stderr)
            img_url = search_and_scrape(f"{title} recipe")
            time.sleep(0.3)

        if not img_url:
            print(f"  Searching (alt): {title}", file=sys.stderr)
            img_url = search_and_scrape(title)
            time.sleep(0.3)

        if img_url:
            print(f"  Found: {img_url[:80]}", file=sys.stderr)
            if download_image(img_url, dest):
                print(f"  Downloaded ({dest.stat().st_size} bytes)", file=sys.stderr)
            else:
                print(f"  Download failed, will still add frontmatter", file=sys.stderr)
        else:
            print(f"  No image found, will still add frontmatter", file=sys.stderr)

        # Add frontmatter (with or without image)
        add_frontmatter(fp, slug, title)
        print(f"  Done", file=sys.stderr)

    print("\n=== Task 3 complete ===", file=sys.stderr)


if __name__ == "__main__":
    main()