mews.page/tests/test_lint_rules.py

"""The checks a DTD cannot express: prologue, head, sizes, images, transport."""

import pytest

from mews.fetch import Fetched
from mews.lint import SIZE_MUST, validate_bytes

CASES = {
    "no-marker": ("3.2", "must", "marker"),
    "two-markers": ("3.2", "must", "marker"),
    "marker-wrong-version": ("3.2", "must", "marker-version"),
    "meta-bad-name": ("3.3", "must", "meta-name"),
    "two-stylesheets": ("3.3", "must", "link-stylesheet-count"),
    "bad-feed-type": ("3.3", "must", "link-alternate-type"),
    "no-xml-decl": ("3.1", "must", "xml-declaration"),
    "bom": ("3.1", "must", "xml-declaration"),
    "wrong-doctype": ("3.1", "must", "doctype"),
    "internal-subset": ("3.1", "must", "internal-subset"),
    "billion-laughs": ("3.1", "must", "internal-subset"),
    "not-wellformed": ("3", "must", "well-formed"),
    "wrong-namespace": ("3.1", "must", "root-element"),
    "data-uri-img": ("4.2", "must", "image-data-uri"),
    "no-lang": ("3.1", "should", "lang"),
    "no-viewport": ("3.3", "should", "viewport"),
}


@pytest.mark.parametrize(("name", "expected"), CASES.items())
def test_rule_failure_maps_to_its_section(fail_page, name, expected):
    report = validate_bytes(fail_page(name))
    section, level, code = expected
    match = [f for f in report.findings if f.code == code]
    assert match, f"{name}: no {code} in {[f.code for f in report.findings]}"
    assert (match[0].section, match[0].level) == (section, level)


def test_entity_declarations_are_refused_before_parsing(fail_page):
    """A page declaring its own entities is rejected without expanding them."""
    report = validate_bytes(fail_page("billion-laughs"))
    assert [f.code for f in report.findings] == ["internal-subset"]


def test_should_level_findings_still_conform(fail_page):
    assert validate_bytes(fail_page("no-viewport")).conforms
    assert validate_bytes(fail_page("no-lang")).conforms


def test_offsite_image_needs_the_page_address(fail_page):
    """Without a URL there is no site to compare against, so nothing is claimed."""
    page = fail_page("offsite-img")
    assert not [f for f in validate_bytes(page).findings if f.code == "image-offsite"]
    report = validate_bytes(page, url="https://example.com/x.html")
    assert [f.code for f in report.findings if f.code == "image-offsite"]


def test_subdomains_of_one_site_are_the_same_site(conforming):
    page = conforming(body='<p><img src="https://img.example.com/a.png" alt="a" /></p>')
    report = validate_bytes(page, url="https://example.com/x.html")
    assert not [f for f in report.findings if f.code == "image-offsite"]


def test_oversized_page_is_reported_not_truncated(conforming):
    filler = "<p>" + ("x" * 200) + "</p>\n    "
    page = conforming(body=filler * (SIZE_MUST // 200 + 20))
    report = validate_bytes(page)
    assert "size-must" in [f.code for f in report.findings]
    assert not report.conforms


def test_page_between_the_two_size_limits_only_warns(conforming):
    filler = "<p>" + ("x" * 200) + "</p>\n    "
    page = conforming(body=filler * 400)
    report = validate_bytes(page)
    assert "size-should" in [f.code for f in report.findings]
    assert report.conforms


def _response(url="https://example.com/", headers=None, status=200):
    import httpx

    return Fetched(url, status, httpx.Headers(headers or {}), b"", False)


def test_a_cookie_on_the_page_is_a_failure(conforming):
    report = validate_bytes(
        conforming(),
        url="https://example.com/",
        response=_response(
            headers={"set-cookie": "a=1", "etag": '"x"', "content-type": "text/html"}
        ),
    )
    cookie = [f for f in report.findings if f.code == "cookie"]
    assert cookie and cookie[0].section == "7.4"
    assert not report.conforms


def test_missing_cache_validators_warn(conforming):
    report = validate_bytes(
        conforming(),
        url="https://example.com/",
        response=_response(headers={"content-type": "text/html"}),
    )
    assert "validators" in [f.code for f in report.findings]
    assert report.conforms


def test_plain_http_warns(conforming):
    report = validate_bytes(
        conforming(),
        url="http://example.com/",
        response=_response(
            url="http://example.com/",
            headers={"content-type": "text/html", "etag": '"x"'},
        ),
    )
    assert "https" in [f.code for f in report.findings]


def test_wrong_content_type_warns(conforming):
    report = validate_bytes(
        conforming(),
        url="https://example.com/",
        response=_response(headers={"content-type": "text/plain", "etag": '"x"'}),
    )
    assert "content-type" in [f.code for f in report.findings]


def test_title_and_description_are_picked_up(conforming):
    report = validate_bytes(conforming())
    assert report.title == "Test site"
    assert report.description == "A test page."
    assert report.language == "en"


def test_control_characters_are_stripped_from_the_title():
    page = """<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE html PUBLIC "-//WAPFORUM//DTD XHTML Mobile 1.2//EN"
  "http://www.openmobilealliance.org/tech/DTD/xhtml-mobile12.dtd">
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
  <head>
    <title>Good‮evil   site</title>
    <meta name="mews-profile" content="0.1" />
  </head>
  <body><p>x</p></body>
</html>
""".encode()
    report = validate_bytes(page)
    assert report.title == "Goodevil site"