Brand mention monitoring

# brand_mentions.py
# Daily brand-mention sweep with a keyword check and dedupe.
# Keep the known URLs somewhere durable between runs.
# Run with: INSIGHTSOCIAL_API_KEY=isk_live_... python brand_mentions.py
import json
import os
import re
from concurrent.futures import ThreadPoolExecutor
from datetime import date, timedelta
from pathlib import Path

import requests

KEY = os.environ["INSIGHTSOCIAL_API_KEY"]
BASE = "https://api.insightsocial.app/v1"
BRAND = "notion"
# The name is an ordinary word, so a row must also mention one of these.
CONTEXT = ["app", "workspace", "template", "notes", "wiki", "database", "page"]
KNOWN_FILE = Path("known-mentions.json")
YESTERDAY = (date.today() - timedelta(days=1)).isoformat()

SEARCHES = {
    "reddit":    ("reddit/search",          {"max_age_days": 1}),
    "tiktok":    ("tiktok/search",          {"max_age_days": 1}),
    "youtube":   ("youtube/search",         {"max_age_days": 1}),
    "twitter":   ("twitter/search/tweets",  {"max_age_days": 1}),
    "threads":   ("threads/search",         {"start_date": YESTERDAY}),
    "instagram": ("instagram/search/reels", {"max_age_days": 1}),
}

BRAND_RE = re.compile(rf"\b{re.escape(BRAND)}\b", re.IGNORECASE)
CONTEXT_RE = re.compile(r"\b(" + "|".join(map(re.escape, CONTEXT)) + r")\b", re.IGNORECASE)

def on_topic(text):
    # A row with no text (a video with no caption) cannot be checked: leave it out.
    if not text or not BRAND_RE.search(text):
        return False
    return not CONTEXT or bool(CONTEXT_RE.search(text))

def get(path, **params):
    res = requests.get(f"{BASE}/{path}", params=params,
                       headers={"x-api-key": KEY}, timeout=120)
    return res.json()

def search(platform):
    path, extra = SEARCHES[platform]
    return platform, get(path, query=BRAND, **extra)

# Step 1: six searches in parallel.
with ThreadPoolExecutor(max_workers=6) as pool:
    results = dict(pool.map(search, SEARCHES))

# Step 2: keep on-topic rows, dedupe against earlier runs by post URL.
known = set(json.loads(KNOWN_FILE.read_text())) if KNOWN_FILE.exists() else set()
fresh, spent, off_topic = [], 0, 0
for platform, body in results.items():
    spent += body.get("credits_used", 0)
    if not body["success"]:
        print(f"{platform}: {body['error']['type']}")
        continue
    for item in body["data"]["items"]:
        post = item["post"]
        text = post["content"]["text"]
        if not on_topic(text):
            off_topic += 1
            continue
        url = post["url"]
        if url and url not in known:
            known.add(url)
            fresh.append((platform, post))
KNOWN_FILE.write_text(json.dumps(sorted(known)))

# Step 3: alert on what is new.
print(f"{len(fresh)} new mentions of {BRAND!r}, {off_topic} off-topic rows skipped ({spent} credits)")
for platform, post in fresh:
    print(f"[{platform}] {post['content']['text'][:120]}")
    print(f"  {post['url']}")
    # Replace print with a Slack webhook, an email or a database insert.