# Trouve-KA — tests de détection de pièges
# Author: Simon-Pierre Boucher
# Contact: contact@spboucher.ai

from trouveka.crawler.traps import looks_like_trap


def test_session_ids():
    assert looks_like_trap("https://x.ca/p?PHPSESSID=abc123")
    assert looks_like_trap("https://x.ca/p?sid=9f8e7d")


def test_too_many_params():
    url = "https://x.ca/p?" + "&".join(f"f{i}={i}" for i in range(12))
    assert looks_like_trap(url)


def test_repeated_facet_param():
    assert looks_like_trap("https://x.ca/s?f=a&f=b&f=c&f=d&f=e")


def test_deep_paths_and_loops():
    assert looks_like_trap("https://x.ca/" + "/".join(["seg"] * 15))
    assert looks_like_trap("https://x.ca/a/b/a/b/a/b/a/b")


def test_far_future_calendar():
    assert looks_like_trap("https://x.ca/events/2085/05/17")
    assert not looks_like_trap("https://x.ca/nouvelles/2024/06/12")


def test_excessive_pagination():
    assert looks_like_trap("https://x.ca/liste?page=9999")
    assert not looks_like_trap("https://x.ca/liste?page=3")


def test_normal_urls_pass():
    assert not looks_like_trap("https://www.quebec.ca/services/permis?type=conduire")
    assert not looks_like_trap("https://ici.radio-canada.ca/nouvelle/2024/ceci-est-un-titre")