Caching added

check_clashes.py

@@ -18,7 +18,7 @@ Importable functions:
 from collections import defaultdict
 from concurrent.futures import ThreadPoolExecutor, as_completed
 from pathlib import Path, PurePosixPath
-from typing import Any, Optional, cast
+from typing import Any, cast
 
 from collections.abc import Callable
 from urllib.parse import urlparse, unquote
@@ -131,8 +131,7 @@ def build_url_referers(video_map: dict[str, Any]) -> dict[str, str]:
         parsed = urlparse(page_url)
         referer = f"{parsed.scheme}://{parsed.netloc}/"
         for vid in cast(dict[str, Any], entry).get("videos", []):
-            if isinstance(vid, str):
-                result.setdefault(vid, referer)
+            result.setdefault(vid["url"], referer)
     return result
 
 
@@ -252,12 +251,7 @@ def fetch_sizes(
 
 def main() -> None:
     vm = load_video_map()
-    urls = [
-        u
-        for entry in vm.values()
-        for u in entry.get("videos", [])
-        if u.startswith("http")
-    ]
+    urls = [vid["url"] for entry in vm.values() for vid in entry.get("videos", [])]
 
     clashes = find_clashes(urls)
 

config.py

@@ -1,6 +1,9 @@
 # config.py
 from typing import Final
 
+# How long a cached file size stays valid.  0 = always re-probe; large = effectively forever.
+SIZE_CACHE_TTL: Final[int] = 9_999_999  # seconds (~115 days)
+
 SITES: Final[dict[str, dict[str, str]]] = {
     "jailbirdz": {
         "base_url": "https://www.jailbirdz.com",

download.py

@@ -16,7 +16,7 @@ import re
 import shutil
 from collections import defaultdict
 from concurrent.futures import ThreadPoolExecutor, as_completed
-from typing import Any, Optional
+from typing import Any
 import requests
 
 from check_clashes import (
@@ -264,12 +264,13 @@ def download_one(
 def collect_urls(video_map: dict[str, Any]) -> list[str]:
     urls, seen, skipped = [], set(), 0
     for entry in video_map.values():
-        for video_url in entry.get("videos", []):
-            if video_url in seen:
+        for vid in entry.get("videos", []):
+            u = vid["url"]
+            if u in seen:
                 continue
-            seen.add(video_url)
-            if is_valid_url(video_url):
-                urls.append(video_url)
+            seen.add(u)
+            if is_valid_url(u):
+                urls.append(u)
             else:
                 skipped += 1
     if skipped:
@@ -281,12 +282,22 @@ def build_url_title_map(video_map: dict[str, Any]) -> dict[str, str]:
     url_title = {}
     for entry in video_map.values():
         title = entry.get("title", "")
-        for video_url in entry.get("videos", []):
-            if video_url not in url_title:
-                url_title[video_url] = title
+        for vid in entry.get("videos", []):
+            if vid["url"] not in url_title:
+                url_title[vid["url"]] = title
     return url_title
 
 
+def build_url_to_site() -> dict[str, str]:
+    """Return {cdn_video_url: site_key} by loading each site's map in turn."""
+    result: dict[str, str] = {}
+    for site_key in SITES:
+        for entry in load_video_map(site_key).values():
+            for vid in entry.get("videos", []):
+                result[vid["url"]] = site_key
+    return result
+
+
 # ── Main ─────────────────────────────────────────────────────────────
 
 
@@ -341,11 +352,7 @@ def main() -> None:
     url_referers = build_url_referers(video_map)
     urls = collect_urls(video_map)
 
-    url_to_site: dict[str, str] = {}
-    for site_key in SITES:
-        for entry in load_video_map(site_key).values():
-            for vid_url in entry.get("videos", []):
-                url_to_site[vid_url] = site_key
+    url_to_site = build_url_to_site()
 
     if args.sites:
         selected = set(args.sites)
@@ -409,17 +416,44 @@ def main() -> None:
             print(f"  … and {len(pending) - 20} more")
         return
 
-    print("\n[+] Fetching remote file sizes…")
+    cached_sizes: dict[str, int] = {
+        vid["url"]: vid["size"]
+        for entry in video_map.values()
+        for vid in entry.get("videos", [])
+        if vid.get("size") is not None
+    }
+
+    uncached_pending = [u for u in pending if u not in cached_sizes]
     session = make_session()
-    remote_sizes = fetch_sizes(pending, workers=20, url_referers=url_referers)
+    if uncached_pending:
+        print(
+            f"\n[+] Fetching remote file sizes ({len(uncached_pending)} uncached, {len(pending) - len(uncached_pending)} cached)…"
+        )
+        remote_sizes: dict[str, int | None] = {
+            **cached_sizes,
+            **fetch_sizes(uncached_pending, workers=20, url_referers=url_referers),
+        }
+    else:
+        print(f"\n[+] All {len(pending)} pending sizes cached — skipping probe.")
+        remote_sizes = dict(cached_sizes)
 
     sized = {u: s for u, s in remote_sizes.items() if s is not None}
     total_bytes = sum(sized.values())
     print(f"[+] Download size: {fmt_size(total_bytes)} across {len(pending)} files")
 
     if already:
-        print(f"[+] Verifying {len(already)} existing files…")
-        already_sizes = fetch_sizes(already, workers=20, url_referers=url_referers)
+        uncached_already = [u for u in already if u not in cached_sizes]
+        if uncached_already:
+            print(
+                f"[+] Verifying {len(already)} existing files ({len(uncached_already)} uncached)…"
+            )
+            already_sizes: dict[str, int | None] = {
+                **cached_sizes,
+                **fetch_sizes(uncached_already, workers=20, url_referers=url_referers),
+            }
+        else:
+            print(f"[+] Verifying {len(already)} existing files (all sizes cached)…")
+            already_sizes = dict(cached_sizes)
 
     mismatched = 0
     for url in already:

main.py

@@ -6,7 +6,7 @@ import signal
 import asyncio
 import requests
 from pathlib import PurePosixPath
-from typing import Any, Optional
+from typing import Any
 from urllib.parse import urlparse
 from dotenv import load_dotenv
 from playwright.async_api import async_playwright, BrowserContext
@@ -476,9 +476,13 @@ async def worker(
                 entry = video_map.get(url, {})
                 if title:
                     entry["title"] = title
-                existing_videos = set(entry.get("videos", []))
-                existing_videos.update(all_videos)
-                entry["videos"] = sorted(existing_videos)
+                existing_dict: dict[str, Any] = {
+                    vid["url"]: vid for vid in entry.get("videos", [])
+                }
+                for vid_url in all_videos:
+                    if vid_url not in existing_dict:
+                        existing_dict[vid_url] = {"url": vid_url}
+                entry["videos"] = sorted(existing_dict.values(), key=lambda v: v["url"])
                 mark_done = bool(all_videos) or not expects_video(url)
                 if mark_done:
                     entry["scraped_at"] = int(time.time())
@@ -526,7 +530,9 @@ async def run_for_site(
             site_key, base_url, wp_api, video_map, urls, req_headers
         )
 
-    known = {u for entry in video_map.values() for u in entry.get("videos", [])}
+    known = {
+        vid["url"] for entry in video_map.values() for vid in entry.get("videos", [])
+    }
 
     total = len(urls)
     pending = []

total_size.py

@@ -4,15 +4,19 @@ Importable function:
     summarize_sizes(sizes) - return dict with total, smallest, largest, average, failed
 """
 
-from typing import Optional, TypedDict
+import argparse
+import time
+from typing import Any, TypedDict
 
 from check_clashes import (
     fmt_size,
     fetch_sizes,
     load_video_map,
+    save_video_map,
     build_url_referers,
     VIDEO_MAP_FILE,
 )
+from config import SITES, SIZE_CACHE_TTL
 
 
 class SizeStats(TypedDict):
@@ -25,7 +29,7 @@ class SizeStats(TypedDict):
     failed: list[str]
 
 
-def summarize_sizes(sizes: dict[str, Optional[int]]) -> SizeStats:
+def summarize_sizes(sizes: dict[str, int | None]) -> SizeStats:
     """Given {url: size_or_None}, return a stats dict."""
     known = {u: s for u, s in sizes.items() if s is not None}
     failed = [u for u, s in sizes.items() if s is None]
@@ -51,6 +55,13 @@ def summarize_sizes(sizes: dict[str, Optional[int]]) -> SizeStats:
     }
 
 
+def _is_stale(vid: dict[str, Any], now: int) -> bool:
+    """True if the cached size is absent or older than SIZE_CACHE_TTL seconds."""
+    if vid.get("size") is None:
+        return True
+    return (now - vid.get("size_checked_at", 0)) >= SIZE_CACHE_TTL
+
+
 # --------------- CLI ---------------
 
 
@@ -59,24 +70,7 @@ def _progress(done: int, total: int) -> None:
         print(f"    {done}/{total}")
 
 
-def main() -> None:
-    vm = load_video_map()
-    urls: list[str] = [
-        u
-        for entry in vm.values()
-        for u in entry.get("videos", [])
-        if u.startswith("http")
-    ]
-
-    url_referers = build_url_referers(vm)
-    print(f"[+] {len(urls)} URLs in {VIDEO_MAP_FILE}")
-    print("[+] Fetching file sizes (20 threads)…\n")
-
-    sizes = fetch_sizes(
-        urls, workers=20, on_progress=_progress, url_referers=url_referers
-    )
-    stats = summarize_sizes(sizes)
-
+def _print_stats(stats: SizeStats) -> None:
     print(f"\n{'=' * 45}")
     print(f"  Sized:    {stats['sized']}/{stats['total']} files")
     print(f"  Total:    {fmt_size(stats['total_bytes'])}")
@@ -84,12 +78,108 @@ def main() -> None:
     print(f"  Largest:  {fmt_size(stats['largest'])}")
     print(f"  Average:  {fmt_size(stats['average'])}")
     print(f"{'=' * 45}")
-
     if stats["failed"]:
         print(f"\n[!] {len(stats['failed'])} URL(s) could not be sized:")
         for u in stats["failed"]:
             print(f"    {u}")
 
 
+def _cache_hint(fresh: int, stale: int, missing: int) -> str:
+    parts = [label for count, label in [(fresh, f"{fresh} fresh"), (stale, f"{stale} stale"), (missing, f"{missing} missing")] if count]
+    if stale or missing:
+        suffix = " — run --write to refresh" if stale else " — run --write to probe missing"
+    else:
+        suffix = " — all current"
+    return f"Cache: {', '.join(parts)}{suffix}"
+
+
+def _run_stats() -> None:
+    vm = load_video_map()
+    now = int(time.time())
+    sizes: dict[str, int | None] = {}
+    fresh = stale = missing = 0
+    for entry in vm.values():
+        for vid in entry.get("videos", []):
+            url = vid["url"]
+            if url in sizes:
+                continue
+            sizes[url] = vid.get("size")
+            if vid.get("size") is None:
+                missing += 1
+            elif _is_stale(vid, now):
+                stale += 1
+            else:
+                fresh += 1
+
+    print(f"[+] {len(sizes)} URLs in {VIDEO_MAP_FILE}")
+    print(f"    {_cache_hint(fresh, stale, missing)}")
+    _print_stats(summarize_sizes(sizes))
+
+
+def _apply_fetched(vm: dict[str, Any], fetched: dict[str, int | None], now: int) -> None:
+    for entry in vm.values():
+        for vid in entry.get("videos", []):
+            if vid["url"] in fetched:
+                vid["size"] = fetched[vid["url"]]
+                vid["size_checked_at"] = now
+
+
+def _run_write() -> None:
+    """Probe uncached sizes and write them into video_map.json."""
+    now = int(time.time())
+    all_fetched: dict[str, int | None] = {}
+
+    for site_key in SITES:
+        vm = load_video_map(site_key)
+        if not vm:
+            continue
+
+        url_referers = build_url_referers(vm)
+
+        to_probe: list[str] = [
+            vid["url"]
+            for entry in vm.values()
+            for vid in entry.get("videos", [])
+            if _is_stale(vid, now)
+        ]
+        cached_count = sum(
+            1
+            for entry in vm.values()
+            for vid in entry.get("videos", [])
+            if not _is_stale(vid, now)
+        )
+        print(f"[{site_key}] {cached_count} cached, {len(to_probe)} to probe…")
+
+        fetched: dict[str, int | None] = {}
+        if to_probe:
+            fetched = fetch_sizes(
+                to_probe, workers=20, on_progress=_progress, url_referers=url_referers
+            )
+
+        _apply_fetched(vm, fetched, now)
+        save_video_map(vm, site_key)
+        all_fetched.update(fetched)
+        print(f"[{site_key}] Written.")
+
+    if all_fetched:
+        _print_stats(summarize_sizes(all_fetched))
+
+
+def main() -> None:
+    parser = argparse.ArgumentParser(description="Calculate total video download size")
+    parser.add_argument(
+        "--write",
+        "-w",
+        action="store_true",
+        help="Probe uncached sizes and write them into video_map.json",
+    )
+    args = parser.parse_args()
+
+    if args.write:
+        _run_write()
+    else:
+        _run_stats()
+
+
 if __name__ == "__main__":
     main()

upload.py

@@ -32,11 +32,11 @@ import requests
 from dotenv import load_dotenv
 
 from check_clashes import fmt_size, url_to_filename, VIDEO_EXTS, load_video_map
-from config import SITES
 from download import (
     collect_urls,
     get_paths_for_mode,
     read_mode,
+    build_url_to_site,
     MODE_ORIGINAL,
     DEFAULT_OUTPUT,
 )
@@ -412,11 +412,7 @@ def build_path_to_meta(
     urls = collect_urls(video_map)
     mode = read_mode(input_dir) or MODE_ORIGINAL
 
-    url_to_site: dict[str, str] = {}
-    for site_key in SITES:
-        for entry in load_video_map(site_key).values():
-            for vid_url in entry.get("videos", []):
-                url_to_site[vid_url] = site_key
+    url_to_site = build_url_to_site()
 
     paths = get_paths_for_mode(mode, urls, video_map, input_dir, url_to_site)
 
@@ -429,13 +425,9 @@ def build_path_to_meta(
         title = t if isinstance(t, str) else ""
         desc = d if isinstance(d, str) else ""
 
-        videos_any = entry.get("videos", [])
-        if isinstance(videos_any, list):
-            for video_url_any in videos_any:
-                if not isinstance(video_url_any, str):
-                    continue
-                if video_url_any not in url_meta:
-                    url_meta[video_url_any] = {"title": title, "description": desc}
+        for vid in entry.get("videos", []):
+            if vid["url"] not in url_meta:
+                url_meta[vid["url"]] = {"title": title, "description": desc}
 
     result: dict[Path, dict[str, str]] = {}
     for url, abs_path in paths.items():