Initial backup: folder structure manifest, tree summary, scraper scripts, and text metadata

pmvhaven/downloader.py

@@ -0,0 +1,36 @@
+import sys
+from pathlib import Path
+from urllib.parse import urlparse
+
+# Add parent workspace directory to path to import generic_downloader
+sys.path.append(str(Path(__file__).resolve().parents[1]))
+import generic_downloader
+
+def is_video_link(url):
+    # Match watch/video URLs like: /video/slug_id, /videos/slug_id, or /watch/slug_id
+    return "/video" in url or "/watch" in url
+
+if __name__ == "__main__":
+    uploader_eval = """() => {
+        const el = Array.from(document.querySelectorAll('a[href*="/user/"], a[href*="/users/"], a[href*="/profile/"], .uploader-name'))
+                        .find(a => a.innerText.trim() !== '');
+        if (el) {
+            const h3 = el.querySelector('h3');
+            if (h3) {
+                const clone = h3.cloneNode(true);
+                const spans = clone.querySelectorAll('span');
+                spans.forEach(s => s.remove());
+                return clone.textContent.trim();
+            }
+            return el.innerText.replace(/uploader/gi, '').trim();
+        }
+        return 'unknown';
+    }"""
+    
+    generic_downloader.run(
+        site_name="PMVHaven",
+        is_video_link_fn=is_video_link,
+        next_page_selector="a:has-text('Next'), .next, a.pagination-next",
+        video_selector="video",
+        uploader_eval_js=uploader_eval
+    )