integrity_scanner_fuer_statische_Webseiten/tests/test_plain.py

"""Tests für die Klartext-Schicht (laienverständliche Ausgabe)."""
import pytest

from scanner.plain import ampel, ampel_zeile, klartext_befunde, was_tun

# Fachbegriffe, die in der Klartext-Ausgabe NICHT vorkommen dürfen
JARGON = ["Score", "Canonical", "JSON-LD", "Meta-Refresh", "Hash", "SHA-256",
          "Whitelist", "Hidden Content", "Exit-Code"]


def _no_jargon(saetze):
    text = " ".join(saetze)
    treffer = [w for w in JARGON if w.lower() in text.lower()]
    assert not treffer, f"Fachbegriff(e) in Klartext gefunden: {treffer}"


class TestAmpel:
    @pytest.mark.parametrize("level", ["green", "yellow", "red"])
    def test_each_level_has_symbol_and_sentence(self, level):
        symbol, satz = ampel(level)
        assert symbol
        assert satz.endswith(".") or satz.endswith(")")

    def test_unknown_level_safe(self):
        symbol, satz = ampel("bogus")
        assert symbol and satz

    def test_ampel_zeile_combines(self):
        zeile = ampel_zeile("green")
        assert zeile.startswith("🟢")


class TestKlartextBefunde:
    def test_empty_diff_says_all_good(self):
        saetze = klartext_befunde({})
        assert any("nichts" in s.lower() for s in saetze)
        _no_jargon(saetze)

    def test_new_external_domain_plain(self):
        saetze = klartext_befunde({"new_external_domains": ["spam.example.com"]})
        assert any("fremde" in s.lower() or "adresse" in s.lower() for s in saetze)
        _no_jargon(saetze)

    def test_hidden_content_plain(self):
        diff = {"page_diffs": [{"url": "x", "new_hidden_content": [{"type": "x"}]}]}
        saetze = klartext_befunde(diff)
        assert any("unsichtbar" in s.lower() for s in saetze)
        _no_jargon(saetze)

    def test_suspicious_filename_is_serious(self):
        saetze = klartext_befunde({"suspicious_filenames": [{"page": "x", "url": "shell.php"}]})
        assert any("warnzeichen" in s.lower() or "verdächtig" in s.lower() for s in saetze)
        _no_jargon(saetze)

    def test_cloaking_extra_link_plain(self):
        cloak = {"findings": [{"url": "x", "extra_links": ["https://spam.com/x"], "extra_text_chars": 0}]}
        saetze = klartext_befunde({}, cloak_diff=cloak)
        assert any("google" in s.lower() for s in saetze)
        _no_jargon(saetze)

    def test_asset_change_plain(self):
        asset = {"changed": [{"url": "x", "type": "script"}]}
        saetze = klartext_befunde({}, asset_diff=asset)
        assert any("datei" in s.lower() for s in saetze)
        _no_jargon(saetze)

    def test_broken_ext_links_plain(self):
        ext = {"broken": 3}
        saetze = klartext_befunde({}, ext_links=ext)
        assert any("funktionieren" in s.lower() or "website" in s.lower() for s in saetze)
        _no_jargon(saetze)

    def test_new_broken_internal_link_plain(self):
        diff = {"new_broken_urls": [{"url": "https://x.de/gone", "status": 404, "pages": []}]}
        saetze = klartext_befunde(diff)
        assert any("leere" in s.lower() or "gefunden" in s.lower() for s in saetze)
        _no_jargon(saetze)

    def test_known_broken_link_still_reported(self):
        """known_broken_urls must appear in Klartext even with score=0."""
        diff = {"known_broken_urls": [{"url": "https://x.de/old", "status": 404, "pages": []}]}
        saetze = klartext_befunde(diff)
        assert any("leere" in s.lower() or "gefunden" in s.lower() for s in saetze), \
            "known_broken_urls müssen auch bei Score=0 im Klartext erscheinen"
        _no_jargon(saetze)


class TestWasTun:
    def test_red_has_urgent_steps(self):
        schritte = was_tun("red")
        assert any("betreut" in s.lower() or "informieren" in s.lower() for s in schritte)

    def test_green_needs_nothing(self):
        schritte = was_tun("green")
        assert any("nichts" in s.lower() for s in schritte)

    def test_no_jargon_anywhere(self):
        for lvl in ("green", "yellow", "red"):
            _no_jargon(was_tun(lvl))