#!/usr/bin/env python3 """Smart image capture for taste board entries. Given a URL, extracts the best representative image of the creative work featured, rather than screenshotting the blog chrome. Strategy chain: 1. OG image (og:image meta tag) — if large enough and not a logo 2. Largest content image on the page (skip nav, icons, ads) 3. Fallback: screenshot scrolled past the header """ import argparse import os import re import sys import tempfile import urllib.request import urllib.error from pathlib import Path from urllib.parse import urljoin, urlparse from bs4 import BeautifulSoup def _fetch_html(url: str, timeout: int = 10) -> str | None: """Fetch HTML with a browser-like UA.""" req = urllib.request.Request(url, headers={ "User-Agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36", }) try: with urllib.request.urlopen(req, timeout=timeout) as resp: return resp.read().decode("utf-8", errors="replace") except Exception: return None def _download_image(img_url: str, dest: str, timeout: int = 10) -> bool: """Download image to dest. Returns True if file > 50KB.""" req = urllib.request.Request(img_url, headers={ "User-Agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36", }) try: with urllib.request.urlopen(req, timeout=timeout) as resp: data = resp.read() if len(data) < 50_000: return False with open(dest, "wb") as f: f.write(data) return True except Exception: return False def _is_logo_url(url: str) -> bool: """Heuristic: skip URLs that look like site logos/icons.""" low = url.lower() return any(k in low for k in ("logo", "icon", "favicon", "badge", "avatar", "sprite", "banner-ad")) def _try_og_image(url: str, output_path: str) -> bool: """Attempt 1: extract og:image and download it.""" html = _fetch_html(url) if not html: return False soup = BeautifulSoup(html, "html.parser") og = soup.find("meta", property="og:image") if not og: og = soup.find("meta", attrs={"name": "og:image"}) if not og or not og.get("content"): return False img_url = urljoin(url, og["content"].strip()) if _is_logo_url(img_url): return False return _download_image(img_url, output_path) def _try_largest_image(url: str, output_path: str) -> bool: """Attempt 2: use Playwright to find the largest visible content image.""" try: from playwright.sync_api import sync_playwright except ImportError: return False with sync_playwright() as p: browser = p.chromium.launch(headless=True) page = browser.new_page(viewport={"width": 1440, "height": 900}) try: page.goto(url, timeout=15000, wait_until="domcontentloaded") # Dismiss cookie banners / popups _dismiss_popups(page) page.wait_for_timeout(2000) # Find all images, pick the largest by rendered area best = page.evaluate("""() => { const imgs = Array.from(document.querySelectorAll('img')); let best = null, bestArea = 0; for (const img of imgs) { const rect = img.getBoundingClientRect(); const area = rect.width * rect.height; const src = img.currentSrc || img.src || ''; if (area < 10000) continue; // skip tiny if (rect.width < 200 || rect.height < 150) continue; // skip stuff in nav/header/footer const parent = img.closest('nav, header, footer, [role="banner"], [role="navigation"]'); if (parent) continue; const low = src.toLowerCase(); if (/logo|icon|favicon|badge|avatar|sprite|banner-ad|pixel/.test(low)) continue; if (area > bestArea) { bestArea = area; best = src; } } return best; }""") if best: img_url = urljoin(url, best) if _download_image(img_url, output_path): return True except Exception: pass finally: browser.close() return False def _try_scroll_screenshot(url: str, output_path: str) -> bool: """Attempt 3: screenshot the page scrolled past the header.""" try: from playwright.sync_api import sync_playwright except ImportError: return False with sync_playwright() as p: browser = p.chromium.launch(headless=True) page = browser.new_page(viewport={"width": 1440, "height": 900}) try: page.goto(url, timeout=15000, wait_until="domcontentloaded") _dismiss_popups(page) page.wait_for_timeout(2000) # Scroll past header (~400px) page.evaluate("window.scrollTo(0, 400)") page.wait_for_timeout(500) page.screenshot(path=output_path, type="jpeg", quality=90) return os.path.exists(output_path) and os.path.getsize(output_path) > 5000 except Exception: return False finally: browser.close() def _dismiss_popups(page): """Try to dismiss cookie banners and popups.""" selectors = [ 'button:has-text("Accept")', 'button:has-text("accept")', 'button:has-text("OK")', 'button:has-text("Got it")', 'button:has-text("Close")', 'button:has-text("Agree")', 'button:has-text("I agree")', 'button:has-text("Allow")', '[id*="cookie"] button', '[class*="cookie"] button', '[id*="consent"] button', '[class*="consent"] button', '[aria-label="Close"]', '[aria-label="close"]', ] for sel in selectors: try: btn = page.locator(sel).first if btn.is_visible(timeout=500): btn.click(timeout=1000) return except Exception: continue def capture_work_image(url: str, output_dir: str, filename: str) -> str | None: """Capture the best image for a creative work URL. Returns the path to the saved JPG, or None on failure. """ os.makedirs(output_dir, exist_ok=True) if not filename.lower().endswith((".jpg", ".jpeg")): filename += ".jpg" output_path = os.path.join(output_dir, filename) # Strategy 1: OG image if _try_og_image(url, output_path): return output_path # Strategy 2: Largest content image via Playwright if _try_largest_image(url, output_path): return output_path # Strategy 3: Scroll-past-header screenshot if _try_scroll_screenshot(url, output_path): return output_path return None if __name__ == "__main__": parser = argparse.ArgumentParser(description="Smart capture of creative work images from URLs") parser.add_argument("url", help="URL to capture") parser.add_argument("--output", "-o", default=".", help="Output directory") parser.add_argument("--name", "-n", default="capture", help="Output filename (without extension)") args = parser.parse_args() result = capture_work_image(args.url, args.output, args.name) if result: print(f"Saved: {result}") else: print("Failed to capture image", file=sys.stderr) sys.exit(1)