"""测试所有 PPTX Readers 的一致性。"""

import pytest
from scripts.readers.pptx import (
    docling,
    markitdown,
    native_xml,
    python_pptx,
    unstructured,
)


class TestPptxReadersConsistency:
    """验证所有 PPTX Readers 解析同一文件时核心文字内容一致。"""

    def test_all_readers_parse_same_content(self, temp_pptx):
        """测试所有 Readers 解析同一文件时核心内容一致。"""
        file_path = temp_pptx(slides=[
            ("测试标题", "这是测试幻灯片内容。"),
            ("第二页", "第二页的内容。")
        ])

        parsers = [
            ("docling", docling.parse),
            ("markitdown", markitdown.parse),
            ("native_xml", native_xml.parse),
            ("python_pptx", python_pptx.parse),
            ("unstructured", unstructured.parse),
        ]

        successful_results = []
        for name, parser in parsers:
            content, error = parser(file_path)
            if content is not None and content.strip():
                successful_results.append((name, content))

        assert len(successful_results) > 0, "没有任何 reader 成功解析文件"

        core_texts = ["测试标题", "幻灯片", "内容", "第二页"]
        for name, content in successful_results:
            assert any(text in content for text in core_texts), \
                f"{name} 解析结果不包含核心内容"