Improve site-check quality: fix broken site configs, add diagnostic utilities, and make self-check report-only by default with opt-in auto-disable. (#2301)

- Fix VK and TradingView checkType; add Reddit and Microsoft Learn API-style probes where appropriate; adjust or disable entries that are unreliable under anti-bot protection.
- Self-check: stop aggressive auto-disable; default to reporting issues only; add --auto-disable and --diagnose for optional fixes and deeper output.
- Tooling: add utils/site_check.py and utils/check_top_n.py (and related helpers) to inspect and rank site behavior against the top-N list
- Scope: aligns with fixing top-traffic / high-impact sites and making diagnostics repeatable without silently flipping disabled flags
This commit is contained in:
Soxoj
2026-03-22 16:48:35 +01:00
committed by GitHub
parent f3b741d283
commit 97cc4b46d9
14 changed files with 1959 additions and 65 deletions
+223
View File
@@ -0,0 +1,223 @@
#!/usr/bin/env python3
"""
Probe likely false-positive sites among the top-N Alexa-ranked entries.
For each of K random *distinct* usernames taken from ``usernameClaimed`` fields in
the Maigret database, runs a clean ``maigret`` scan (``--top-sites N --json simple|ndjson``).
Sites that return CLAIMED in *every* run are reported: unrelated random claimed
handles are unlikely to all exist on the same third-party site, so such sites are
candidates for broken checks.
"""
from __future__ import annotations
import argparse
import json
import random
import shutil
import subprocess
import sys
import tempfile
from pathlib import Path
def repo_root() -> Path:
return Path(__file__).resolve().parent.parent
def load_username_claimed_pool(db_path: Path) -> list[str]:
with db_path.open(encoding="utf-8") as f:
data = json.load(f)
sites = data.get("sites") or {}
seen: set[str] = set()
pool: list[str] = []
for _name, site in sites.items():
u = (site or {}).get("usernameClaimed")
if not u or not isinstance(u, str):
continue
u = u.strip()
if not u or u in seen:
continue
seen.add(u)
pool.append(u)
return pool
def run_maigret(
*,
username: str,
db_path: Path,
out_dir: Path,
top_sites: int,
json_format: str,
quiet: bool,
) -> Path:
"""Run maigret subprocess; return path to the written JSON report."""
safe = username.replace("/", "_")
report_name = f"report_{safe}_{json_format}.json"
report_path = out_dir / report_name
cmd = [
sys.executable,
"-m",
"maigret",
username,
"--db",
str(db_path),
"--top-sites",
str(top_sites),
"--json",
json_format,
"--folderoutput",
str(out_dir),
"--no-progressbar",
"--no-color",
"--no-recursion",
"--no-extracting",
]
sink = subprocess.DEVNULL if quiet else None
proc = subprocess.run(
cmd,
cwd=str(repo_root()),
text=True,
stdout=sink,
stderr=sink,
)
if proc.returncode != 0:
raise RuntimeError(
f"maigret exited with {proc.returncode} for username {username!r}"
)
if not report_path.is_file():
raise FileNotFoundError(f"Expected report missing: {report_path}")
return report_path
def claimed_sites_from_report(path: Path, json_format: str) -> set[str]:
if json_format == "simple":
with path.open(encoding="utf-8") as f:
data = json.load(f)
if not isinstance(data, dict):
return set()
return set(data.keys())
# ndjson: one object per line, each has "sitename"
sites: set[str] = set()
with path.open(encoding="utf-8") as f:
for line in f:
line = line.strip()
if not line:
continue
obj = json.loads(line)
name = obj.get("sitename")
if isinstance(name, str) and name:
sites.add(name)
return sites
def main() -> int:
parser = argparse.ArgumentParser(
description=(
"Pick random distinct usernameClaimed values, run maigret --top-sites N "
"with JSON reports, and list sites that claimed all of them (suspicious FP)."
)
)
parser.add_argument(
"--db",
"-b",
type=Path,
default=repo_root() / "maigret" / "resources" / "data.json",
help="Path to Maigret data.json (a temp copy is used for runs).",
)
parser.add_argument(
"--top-sites",
"-n",
type=int,
default=500,
metavar="N",
help="Value for maigret --top-sites (default: 500).",
)
parser.add_argument(
"--samples",
"-k",
type=int,
default=5,
metavar="K",
help="How many distinct random usernames to draw (default: 5).",
)
parser.add_argument(
"--seed",
type=int,
default=None,
help="RNG seed for reproducible username selection.",
)
parser.add_argument(
"--json",
dest="json_format",
default="simple",
choices=["simple", "ndjson"],
help="JSON report type passed to maigret -J (default: simple).",
)
parser.add_argument(
"--verbose",
"-v",
action="store_true",
default=False,
help="Print maigret stdout/stderr (default: suppress child output).",
)
args = parser.parse_args()
quiet = not args.verbose
db_src = args.db.resolve()
if not db_src.is_file():
print(f"Database not found: {db_src}", file=sys.stderr)
return 2
pool = load_username_claimed_pool(db_src)
if len(pool) < args.samples:
print(
f"Need at least {args.samples} distinct usernameClaimed entries, "
f"found {len(pool)}.",
file=sys.stderr,
)
return 2
rng = random.Random(args.seed)
picked = rng.sample(pool, args.samples)
print(f"Database: {db_src}")
print(f"--top-sites {args.top_sites}, {args.samples} random usernameClaimed:")
for i, u in enumerate(picked, 1):
print(f" {i}. {u}")
site_sets: list[set[str]] = []
with tempfile.TemporaryDirectory(prefix="maigret_fp_probe_") as tmp:
tmp_path = Path(tmp)
db_work = tmp_path / "data.json"
shutil.copyfile(db_src, db_work)
for u in picked:
print(f"\nRunning maigret for {u!r} ...", flush=True)
report = run_maigret(
username=u,
db_path=db_work,
out_dir=tmp_path,
top_sites=args.top_sites,
json_format=args.json_format,
quiet=quiet,
)
sites = claimed_sites_from_report(report, args.json_format)
site_sets.append(sites)
print(f" -> {len(sites)} positive site(s) in JSON", flush=True)
always = set.intersection(*site_sets) if site_sets else set()
print("\n--- Sites with CLAIMED in all runs (candidates for false positives) ---")
if not always:
print("(none)")
else:
for name in sorted(always):
print(name)
return 0
if __name__ == "__main__":
raise SystemExit(main())