omnisvg-lite/tests/test_validator.py

"""Юнит-тесты для validator.py.

Покрывают контрактные правила из design.md §5 и явные требования задачи:
- валидный SVG → ok=True
- отсутствует viewBox → ok=False (missing_viewbox)
- есть <script> → ok=False (disallowed_tag)
- есть <foreignObject> → ok=False (disallowed_tag)
- http:// ссылка в href → ok=False (external_ref)
- on*= атрибут → ok=False (event_handler)

Плюс несколько дополнительных кейсов, чтобы покрыть size-limit, парсинг
из обёртки с пояснением и edge cases.

Запуск:
    python -m pytest tests/ -v
"""

from __future__ import annotations

import sys
from pathlib import Path

# Делаем корень проекта доступным как `import validator`.
ROOT = Path(__file__).resolve().parent.parent
if str(ROOT) not in sys.path:
    sys.path.insert(0, str(ROOT))

import pytest  # noqa: E402

from validator import (  # noqa: E402
    MAX_BYTES,
    validate,
    validate_svg,
    extract_svg,
    ValidatorError,
)


# ---------------------------------------------------------------------------
# Положительные кейсы
# ---------------------------------------------------------------------------


VALID_ICON = """<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 64 64" width="64" height="64">
  <circle cx="26" cy="26" r="14" fill="#3B82F6"/>
  <rect x="36" y="34" width="6" height="20" rx="3" fill="#3B82F6" transform="rotate(45 39 44)"/>
</svg>"""


def test_valid_svg_returns_ok_and_cleaned():
    ok, reason, cleaned = validate_svg(VALID_ICON, mode="icon")
    assert ok is True
    assert reason == ""
    assert "<svg" in cleaned
    assert cleaned.endswith("</svg>")
    # viewBox должен сохраниться после сериализации.
    assert 'viewBox="0 0 64 64"' in cleaned


def test_valid_svg_wrapped_in_prose_is_extracted():
    """Модель иногда отвечает `Here is your icon: <svg>...</svg>`. Должны вытащить."""
    wrapped = (
        "Sure! Here is your SVG icon:\n"
        f"{VALID_ICON}\n"
        "Hope that helps."
    )
    ok, reason, cleaned = validate_svg(wrapped, mode="icon")
    assert ok is True, f"expected ok, got reason={reason!r}"
    assert reason == ""
    assert "<svg" in cleaned


def test_valid_illustration_with_filter():
    """В illustration разрешены <filter>, <feGaussianBlur> и т.п."""
    ill = """<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 512 512" width="512" height="512">
  <defs>
    <filter id="b" x="-10%" y="-10%" width="120%" height="120%">
      <feGaussianBlur stdDeviation="2"/>
    </filter>
  </defs>
  <rect x="100" y="100" width="200" height="200" fill="#ff0000" filter="url(#b)"/>
</svg>"""
    ok, reason, _ = validate_svg(ill, mode="illustration")
    assert ok is True, f"reason={reason!r}"


# ---------------------------------------------------------------------------
# Негативные кейсы
# ---------------------------------------------------------------------------


def test_missing_viewbox_rejected():
    bad = """<svg xmlns="http://www.w3.org/2000/svg">
  <circle cx="10" cy="10" r="5"/>
</svg>"""
    ok, reason, cleaned = validate_svg(bad, mode="icon")
    assert ok is False
    assert reason == "missing_viewbox"
    assert cleaned == ""


def test_bad_viewbox_rejected():
    bad = """<svg xmlns="http://www.w3.org/2000/svg" viewBox="oops">
  <circle cx="10" cy="10" r="5"/>
</svg>"""
    ok, reason, _ = validate_svg(bad, mode="icon")
    assert ok is False
    assert reason == "bad_viewbox"


def test_script_tag_rejected():
    bad = """<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 64 64">
  <script>alert(1)</script>
</svg>"""
    ok, reason, _ = validate_svg(bad, mode="icon")
    assert ok is False
    assert reason == "disallowed_tag"


def test_foreign_object_rejected():
    bad = """<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 64 64">
  <foreignObject width="100" height="100">
    <div xmlns="http://www.w3.org/1999/xhtml">Hi</div>
  </foreignObject>
</svg>"""
    ok, reason, _ = validate_svg(bad, mode="icon")
    assert ok is False
    assert reason == "disallowed_tag"


def test_http_ref_rejected():
    bad = """<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 64 64">
  <use href="http://example.com/sprite.svg#icon"/>
</svg>"""
    ok, reason, _ = validate_svg(bad, mode="icon")
    assert ok is False
    assert reason == "external_ref"


def test_onclick_attribute_rejected():
    bad = """<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 64 64">
  <rect x="0" y="0" width="10" height="10" onclick="alert(1)"/>
</svg>"""
    ok, reason, _ = validate_svg(bad, mode="icon")
    assert ok is False
    assert reason == "event_handler"


def test_url_with_https_rejected():
    """https:// в url(...) тоже блокируется."""
    bad = """<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 64 64">
  <defs>
    <pattern id="p" width="4" height="4" patternUnits="userSpaceOnUse">
      <rect width="4" height="4" fill="url('https://example.com/tex.png')"/>
    </pattern>
  </defs>
</svg>"""
    ok, reason, _ = validate_svg(bad, mode="icon")
    assert ok is False
    # Сейчас external_ref ловится на `url(` независимо от http(s).
    assert reason in ("external_ref", "disallowed_tag", "unknown_tag")


def test_too_large_rejected():
    """SVG > MAX_BYTES[mode] → too_large."""
    # Соберём раздутый SVG с большим комментарием.
    pad = "<!-- " + ("x" * (MAX_BYTES["icon"] + 100)) + " -->"
    bad = (
        '<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 64 64">'
        + pad
        + "</svg>"
    )
    ok, reason, _ = validate_svg(bad, mode="icon")
    assert ok is False
    assert reason == "too_large"


def test_empty_input_rejected():
    ok, reason, _ = validate_svg("", mode="icon")
    assert ok is False
    assert reason == "empty_input"


def test_no_svg_block_rejected():
    ok, reason, _ = validate_svg("Sorry, I cannot help with that.", mode="icon")
    assert ok is False
    assert reason == "not_svg"


# ---------------------------------------------------------------------------
# Прямые проверки API (validate() бросает, extract_svg() утилитарный)
# ---------------------------------------------------------------------------


def test_validate_strict_raises():
    with pytest.raises(ValidatorError) as excinfo:
        validate("plain text", mode="icon")
    assert excinfo.value.code == "not_svg"


def test_extract_svg_finds_block():
    text = "noise <svg viewBox='0 0 1 1'></svg> noise"
    found = extract_svg(text)
    assert found.startswith("<svg")
    assert found.endswith("</svg>")


def test_validate_svg_returns_tuple_for_unexpected_exception():
    """Даже если что-то странное, обёртка возвращает кортеж (ok=False, ...)."""
    ok, reason, cleaned = validate_svg(None, mode="icon")  # type: ignore[arg-type]
    assert ok is False
    assert reason in ("empty_input", "internal:TypeError")
    assert cleaned == ""