"""Read three small-web Bluesky topic feeds (public API, no login) and save their posts.
Vonk, project 90. Read-only, 0.7 s between calls, stops at MAX_DAYS old or MAX_PAGES."""
import json, time, urllib.request
from datetime import datetime, timezone, timedelta

FEEDS = {
    "neocities": "at://did:plc:gy4mgzgeqsg5h47z5wx4gy62/app.bsky.feed.generator/aaangtgczuo7a",
    "indieweb": "at://did:plc:kn4dz4f67maytvsbcnbry36k/app.bsky.feed.generator/aaact2ln2gffk",
    "www": "at://did:plc:bsxmru6yec5sy42azm7cnrqq/app.bsky.feed.generator/aaai2qt6lgrng",
}
ME = "did:plc:e46yjn4y2jktr6riufqzx5fc"
MAX_DAYS, MAX_PAGES = 14, 12
UA = {"User-Agent": "Vonk (AI character) reading public feeds, project 90"}

def get(url):
    return json.load(urllib.request.urlopen(urllib.request.Request(url, headers=UA), timeout=25))

cut = datetime.now(timezone.utc) - timedelta(days=MAX_DAYS)
posts, seen = [], set()
for name, feed in FEEDS.items():
    cursor, pages, n = None, 0, 0
    while pages < MAX_PAGES:
        url = f"https://api.bsky.app/xrpc/app.bsky.feed.getFeed?feed={feed}&limit=100" + (f"&cursor={cursor}" if cursor else "")
        r = get(url); pages += 1
        old = False
        for it in r.get("feed", []):
            p = it["post"]; rec = p.get("record", {})
            t = rec.get("createdAt", p.get("indexedAt", ""))
            try:
                dt = datetime.fromisoformat(t.replace("Z", "+00:00"))
            except Exception:
                dt = None
            if dt and dt < cut:
                old = True; continue
            if p["author"]["did"] == ME or p["uri"] in seen:
                continue
            seen.add(p["uri"]); n += 1
            rkey = p["uri"].split("/")[-1]
            posts.append({"feed": name, "uri": p["uri"], "url": f"https://bsky.app/profile/{p['author']['handle']}/post/{rkey}",
                          "handle": p["author"]["handle"], "time": t, "text": rec.get("text", ""),
                          "reply": "reply" in rec, "likes": p.get("likeCount", 0), "replies": p.get("replyCount", 0),
                          "labels": [l.get("val") for l in p["author"].get("labels", [])]})
        cursor = r.get("cursor")
        if old or not cursor or not r.get("feed"):
            break
        time.sleep(0.7)
    print(name, "posts", n, "pages", pages)
json.dump(posts, open("posts.json", "w", encoding="utf-8"), ensure_ascii=False, indent=1)
print("total", len(posts))