thermograph/frontend/tests/unit/test_rendering.py

"""Unit tier: the SSR rendering layer (content.py + format.py) fed committed backend
fixtures via the hermetic `client` fixture (tests/conftest.py). Asserts the rendering
*contract* — pages render and carry the expected structure — not exact climate numbers
(those are the backend's responsibility). Ported from the old backend-coupled
test_content.py; refresh the fixtures with scripts/capture-fixtures.sh.
"""
import re

import pytest

B = "/thermograph"                 # matches THERMOGRAPH_BASE in tests/conftest.py
SLUG = "london-england-gb"
MONTH = "july"


def test_robots_txt(client):
    r = client.get(f"{B}/robots.txt")
    assert r.status_code == 200
    assert "Sitemap:" in r.text and "/sitemap.xml" in r.text
    assert f"Disallow: {B}/api/" in r.text


def test_sitemap_lists_city_urls(client):
    r = client.get(f"{B}/sitemap.xml")
    assert r.status_code == 200
    assert "<urlset" in r.text
    assert f"/climate/{SLUG}</loc>" in r.text
    assert f"/climate/{SLUG}/{MONTH}</loc>" in r.text
    assert f"/climate/{SLUG}/records</loc>" in r.text


# The public site is HTTPS-only behind Caddy, which terminates TLS and proxies
# plain HTTP -- so x-forwarded-proto / request.url.scheme read "http". A request
# arriving on the configured public host (THERMOGRAPH_BASE_URL defaults to
# https://thermograph.org in the test env) must still emit https:// in every
# frontend-built absolute URL, or canonicals self-conflict and the sitemap lists
# redirecting URLs.
_PUBLIC = {"host": "thermograph.org", "x-forwarded-proto": "http"}


def test_public_host_canonical_and_og_are_https(client):
    b = client.get(f"{B}/climate/{SLUG}", headers=_PUBLIC).text
    assert f'<link rel="canonical" href="https://thermograph.org{B}/climate/{SLUG}"' in b
    assert f'<meta property="og:url" content="https://thermograph.org{B}/climate/{SLUG}"' in b
    assert 'property="og:image" content="https://thermograph.org' in b


def test_public_host_sitemap_locs_are_https(client):
    body = client.get(f"{B}/sitemap.xml", headers=_PUBLIC).text
    assert f"<loc>https://thermograph.org{B}/climate/{SLUG}</loc>" in body
    assert "<loc>http://thermograph.org" not in body


def test_public_host_robots_sitemap_is_https(client):
    body = client.get(f"{B}/robots.txt", headers=_PUBLIC).text
    assert f"Sitemap: https://thermograph.org{B}/sitemap.xml" in body


def test_non_public_host_keeps_forwarded_scheme(client):
    # A LAN/dev host never matches the configured public host, so the observed
    # (plain-http) scheme is preserved -- http development is unaffected.
    b = client.get(f"{B}/climate/{SLUG}",
                   headers={"host": "192.168.1.10:8000", "x-forwarded-proto": "http"}).text
    assert f'<link rel="canonical" href="http://192.168.1.10:8000{B}/climate/{SLUG}"' in b


def test_proxied_forwarded_host_still_https(client):
    # Reached via backend's proxy fallback: Host is the internal hop, the real
    # browser host arrives in X-Forwarded-Host -- which still resolves to the
    # public host and must go https.
    b = client.get(f"{B}/climate/{SLUG}",
                   headers={"host": "frontend:8080", "x-forwarded-host": "thermograph.org",
                            "x-forwarded-proto": "http"}).text
    assert f'<link rel="canonical" href="https://thermograph.org{B}/climate/{SLUG}"' in b


def test_city_page_renders_structure(client):
    r = client.get(f"{B}/climate/{SLUG}")
    assert r.status_code == 200
    b = r.text
    assert "London" in b and "climate" in b.lower()
    assert 'rel="canonical"' in b and f"/climate/{SLUG}" in b
    assert '"@type":"Dataset"' in b                              # JSON-LD dataset block
    assert "average temperatures by month" in b.lower()
    assert re.search(r"-?\d+°C</span>", b)                       # GB city -> Celsius


@pytest.mark.parametrize("path", ["", f"/{MONTH}", "/records"])
def test_city_subpages_render(client, path):
    r = client.get(f"{B}/climate/{SLUG}{path}")
    assert r.status_code == 200
    assert "text/html" in r.headers["content-type"]
    assert "__ORIGIN__" not in r.text                            # origin placeholder filled in


def test_homepage_renders(client):
    r = client.get(f"{B}/")
    assert r.status_code == 200
    assert "text/html" in r.headers["content-type"]


def test_hub_renders(client):
    r = client.get(f"{B}/climate")
    assert r.status_code == 200
    assert "text/html" in r.headers["content-type"]


def test_unknown_city_is_404(client):
    assert client.get(f"{B}/climate/nope-not-a-city").status_code == 404


def test_unknown_month_is_404(client):
    assert client.get(f"{B}/climate/{SLUG}/nonemonth").status_code == 404


# --- backend-down resilience (migrated from main's hardening of the old
# tests/test_content.py, which the unit-tier restructure deleted) ------------
def _unreachable(*_a, **_k):
    import httpx
    req = httpx.Request("GET", "http://backend.invalid/x")
    raise httpx.ConnectError("connection refused", request=req)


@pytest.mark.parametrize("path, attr", [
    (f"/climate/{SLUG}", "city"),
    (f"/climate/{SLUG}/july", "city_month"),
    (f"/climate/{SLUG}/records", "city_records"),
    ("/climate", "hub"),
    ("/", "home"),
    ("/sitemap.xml", "sitemap"),
])
def test_backend_unreachable_maps_to_503(client, monkeypatch, path, attr):
    """A backend that's down/timing out (httpx.ConnectError, TimeoutException,
    ...) never raises httpx.HTTPStatusError -- it doesn't even get a response
    -- so it must be caught separately from the 404/503 status-code passthrough
    or it escapes as a raw framework 500 instead of a clean 503."""
    import api_client
    monkeypatch.setattr(api_client, attr, _unreachable)
    assert client.get(f"{B}{path}").status_code == 503