"""Smoke tests for TextChunker"""

from yuxi.channels.infra.text_chunker import TextChunker, ChunkMode


def test_markdown_code_block_intact_when_chunked():
    tc = TextChunker(max_length=30, mode=ChunkMode.MARKDOWN)
    md_text = (
        "Short intro\n"
        "```python\n"
        'print("hello world this is long")\n'
        'print("another long line here too")\n'
        "```\n"
        "After code block."
    )
    result = tc.chunk(md_text)
    print("MD code block (chunked):")
    for i, c in enumerate(result):
        print(f"  Chunk {i}: {repr(c)}")
    # Code block should not be split mid-block
    for c in result:
        if c.startswith("```"):
            assert "```" in c.rsplit("\n", 1)[-1] or c.count("```") >= 2, (
                f"Code block should not be split mid-block: {repr(c)}"
            )


def test_markdown_table_intact():
    tc = TextChunker(max_length=200, mode=ChunkMode.MARKDOWN)
    table_text = "Header\n| col1 | col2 |\n|------|------|\n| a | b |\n| c | d |\n\nAfter table."
    result = tc.chunk(table_text)
    print("\nMD (table intact):")
    for i, c in enumerate(result):
        print(f"  Chunk {i}: {repr(c)}")
    assert any("|------|------|" in c for c in result), "Table should be in a chunk"


def test_markdown_headings_separate_when_over_limit():
    tc = TextChunker(max_length=30, mode=ChunkMode.MARKDOWN)
    heading_text = "## Hello\nSome content under heading.\n\n## World\nDifferent heading text here."
    result = tc.chunk(heading_text)
    print("\nMD (headings, over limit):")
    for i, c in enumerate(result):
        print(f"  Chunk {i}: {repr(c)}")
    assert len(result) >= 2, "Over-limit headings should separate content"


def test_empty_text():
    tc = TextChunker(max_length=200, mode=ChunkMode.NEWLINE)
    assert tc.chunk("") == []
    assert tc.chunk_as_objects("") == []


def test_single_chunk():
    tc = TextChunker(max_length=200, mode=ChunkMode.LENGTH)
    result = tc.chunk("short text")
    assert result == ["short text"]


def test_chunk_with_index():
    tc = TextChunker(max_length=15, mode=ChunkMode.LENGTH)
    results = list(tc.chunk_with_index("Part one here. Part two also here. Part three goes on."))
    assert len(results) > 1
    assert results[0] == (1, len(results), results[0][2])
    assert results[-1] == (len(results), len(results), results[-1][2])


if __name__ == "__main__":
    test_markdown_code_block_intact_when_chunked()
    test_markdown_table_intact()
    test_markdown_headings_separate_when_over_limit()
    test_empty_text()
    test_single_chunk()
    test_chunk_with_index()
    print("\nALL SMOKE TESTS PASSED")