master-zapier-plan-draft/research/zapier/scripts/scrape-extras.py
George Lambert b4150c8250 Milestone 0: import zappier billing, Verae middleware, and Zapier research
Compose-ready workspace: packages/zappier (rate card, portal, Stripe),
packages/verae-zapier-middleware (timestamp + NATS), packages/verae-zapier
(CLI app), vendor/zapier-platform, and research/zapier vendor corpus.

Gate 0 structure checks pass. Product code and research are not yet wired.
2026-09-09 02:37:36 -04:00

104 lines
3.3 KiB
Python

#!/usr/bin/env python3
"""Pull featured Zap templates, help article URLs, and overview text from Zapier pages."""
from __future__ import annotations
import json
import re
import ssl
import urllib.request
from concurrent.futures import ThreadPoolExecutor, as_completed
from pathlib import Path
ROOT = Path(__file__).resolve().parents[1]
CATALOG = ROOT / "raw" / "all-apps-full.jsonl"
OUT = ROOT / "raw" / "extras-scrape.jsonl"
def load_done() -> set[str]:
done = set()
if OUT.exists():
for line in OUT.open():
try:
done.add(json.loads(line)["slug"])
except Exception:
continue
return done
def texts(node) -> str:
parts = []
def walk(n):
if isinstance(n, dict):
if n.get("rawText"):
parts.append(n["rawText"])
for c in n.get("childNodes") or []:
walk(c)
walk(node)
return re.sub(r"\s+", " ", " ".join(parts)).strip()
def fetch(slug: str) -> dict:
url = f"https://zapier.com/apps/{slug}/integrations"
req = urllib.request.Request(url, headers={"User-Agent": "Mozilla/5.0 research"})
ctx = ssl.create_default_context()
with urllib.request.urlopen(req, timeout=20, context=ctx) as r:
html = r.read().decode("utf-8", errors="ignore")
m = re.search(r'<script id="__NEXT_DATA__" type="application/json">(.*?)</script>', html)
if not m:
return {"slug": slug, "error": "no_next_data"}
page = json.loads(m.group(1))["props"]["pageProps"]["oneAppPage"]
details = page.get("appDetails") or {}
pt = page.get("paginatedZapTemplates") or {}
featured = []
for t in pt.get("results") or []:
apps = [a.get("slug") for a in (t.get("apps") or []) if a.get("slug")]
featured.append(
{
"id": t.get("id"),
"title": t.get("title"),
"url": "https://zapier.com" + (t.get("canonicalPageRelativePath") or ""),
"description": t.get("metaDescription"),
"apps": apps,
}
)
help_ = []
for h in details.get("helpContent") or []:
help_.append({"title": h.get("title"), "url": h.get("canonicalUrl")})
overview = texts(page.get("integrationOverviewHtmlAst"))[:2000]
return {
"slug": slug,
"template_count_zapier": pt.get("count"),
"featured_templates": featured,
"help_articles": help_,
"overview": overview,
"error": None,
}
def main() -> None:
apps = [json.loads(l) for l in CATALOG.open()]
done = load_done()
todo = [a["slug"] for a in apps if a["slug"] not in done]
print(f"todo={len(todo)} done={len(done)}", flush=True)
n = err = 0
with OUT.open("a") as out, ThreadPoolExecutor(max_workers=28) as ex:
futs = {ex.submit(fetch, s): s for s in todo}
for fut in as_completed(futs):
slug = futs[fut]
try:
rec = fut.result()
except Exception as e:
rec = {"slug": slug, "error": f"{type(e).__name__}: {e}"}
err += 1
out.write(json.dumps(rec, ensure_ascii=False) + "\n")
out.flush()
n += 1
if n % 200 == 0:
print(f"{n}/{len(todo)} err={err} last={slug}", flush=True)
print("finished", n, "err", err)
if __name__ == "__main__":
main()