# Trouve-KA — tests du scoring Québec
# Author: Simon-Pierre Boucher
# Contact: contact@spboucher.ai

from trouveka.classifier import score_page
from trouveka.types import ParsedPage


def _page(**kwargs) -> ParsedPage:
    defaults = {"url": "https://example.com/page", "title": "", "body": ""}
    return ParsedPage(**(defaults | kwargs))


def test_quebec_business_scores_high():
    page = _page(
        url="https://plomberiexyz.qc.ca/services",
        title="Plomberie XYZ — Plombier à Gatineau",
        description="Service de plomberie résidentielle en Outaouais",
        body="Plomberie XYZ dessert Gatineau et l'Outaouais. 123 rue Principale, Gatineau (Québec) J8X 3X3. "
             "Appelez-nous au 819-555-1234. Urgences 24h partout au Québec.",
        headings=["Plombier Gatineau", "Nos services"],
        language="fr",
    )
    signals = score_page(page, "plomberiexyz.qc.ca")
    assert signals.score > 0.7
    assert "gatineau" in signals.locations
    assert "tld_quebec" in signals.reasons
    assert "code_postal_qc" in signals.reasons


def test_ontario_business_scores_low():
    page = _page(
        url="https://ottawaplumbing.ca/",
        title="Ottawa Plumbing Services",
        description="Plumbing services in Ottawa, Ontario",
        body="We serve Ottawa, Kanata and Nepean. 456 Bank Street, Ottawa, ON K1S 3T4. Call 613-555-9999.",
        headings=["Plumbing Ottawa"],
        language="en",
    )
    signals = score_page(page, "ottawaplumbing.ca")
    assert signals.score < 0.2


def test_nyt_article_about_montreal_gets_page_signal():
    # Un article international SUR Montréal : le domaine n'est pas québécois,
    # mais la page a un signal réel (page_quebec_score > 0, sans être maximal).
    page = _page(
        url="https://www.nytimes.com/2026/01/01/travel/montreal.html",
        title="36 Hours in Montreal",
        description="What to do in Montreal, Quebec's largest city",
        body="Montreal is the largest city in Quebec. From the Plateau-Mont-Royal to Old Montreal, "
             "the city offers poutine, festivals and more. Nearby Laval and Longueuil...",
        headings=["36 Hours in Montreal"],
        language="en",
    )
    signals = score_page(page, "nytimes.com")
    assert 0.2 < signals.score < 0.8
    assert "montreal" in signals.locations


def test_generic_english_page_scores_near_zero():
    page = _page(
        url="https://techblog.com/post",
        title="Understanding Rust lifetimes",
        body="Rust lifetimes are a way to express the scope of references in your program. " * 20,
        language="en",
    )
    signals = score_page(page, "techblog.com")
    assert signals.score < 0.1


def test_french_alone_is_not_proof():
    # Une page française de France ne doit pas passer le seuil sur la langue seule
    page = _page(
        url="https://lemonde.fr/article",
        title="Actualités françaises",
        body="La France annonce de nouvelles mesures. Paris, Lyon et Marseille concernées. " * 10,
        language="fr",
    )
    signals = score_page(page, "lemonde.fr")
    assert signals.score < 0.3


def test_score_saturates_below_one():
    body = "Montréal Québec Gatineau Sherbrooke Laval " * 100 + " H2X 1Y6 G1R 4S9 514-555-0000 Hydro-Québec"
    page = _page(url="https://x.qc.ca/", title="Montréal Québec", body=body, language="fr")
    signals = score_page(page, "x.qc.ca")
    assert 0.8 < signals.score <= 1.0