from bindery.diffing import diff_html, html_to_blocks, text_hash


def test_random_class_names_do_not_count_as_changes():
    a = '<div><p class="cnMwZG">Hello <em>there</em>.</p><p class="x1">Second</p></div>'
    b = '<div><p class="zzQ99">Hello <em>there</em>.</p><p class="y2">Second</p></div>'
    assert text_hash(a) == text_hash(b)
    assert html_to_blocks(a) == ["Hello there.", "Second"]


def test_text_changes_change_the_hash():
    assert text_hash("<p>one</p>") != text_hash("<p>two</p>")


def test_diff_counts_and_word_level_changes():
    old = "<p>Same.</p><p>The cat sat.</p><p>Gone.</p>"
    new = "<p>Same.</p><p>The dog sat.</p><p>Gone.</p><p>Added.</p>"
    result = diff_html(old, new)
    assert (result["changed"], result["added"], result["removed"]) == (1, 1, 0)
    assert "<del>cat</del><ins>dog</ins>" in result["html"]
    assert not result["identical"]


def test_diff_output_is_escaped():
    result = diff_html("<p>a</p>", "<p>&lt;script&gt;alert(1)&lt;/script&gt;</p>")
    assert "<script>" not in result["html"]
    assert "&lt;script&gt;" in result["html"]


def test_source_diff_sees_markup_changes():
    old, new = "<p>Hi</p>", '<p class="x">Hi</p>'
    assert diff_html(old, new)["identical"]
    assert not diff_html(old, new, source=True)["identical"]