EnterpriseArchitect/knowledge/运维/shared-scripts/test_pre_trim_multica_comment_v1.0.1.py

"""
test_pre_trim_multica_comment.py — pre_trim_multica_comment 单元测试

覆盖 4 类样本（COO 要求）：
  1. 长 markdown
  2. 嵌套表
  3. 多代码块
  4. 中英混排

版本：v1.0
作者：严维序（opengineer）
"""

import os
import sys
import unittest

_SCRIPT_DIR = os.path.dirname(os.path.abspath(__file__))
sys.path.insert(0, _SCRIPT_DIR)

from pre_trim_multica_comment import (
    pre_trim_for_multica_comment,
    DEFAULT_MAX_BYTES,
    __version__,
)


class TestPreTrimLongMarkdown(unittest.TestCase):
    """长 markdown 测试"""

    def test_long_paragraphs_get_truncated(self):
        long_text = "这是中文段落 mixed with English text. " * 100
        result = pre_trim_for_multica_comment(long_text)
        self.assertLessEqual(result["trimmed_bytes"], DEFAULT_MAX_BYTES)
        self.assertTrue(result["truncated"])

    def test_short_markdown_not_truncated(self):
        short = "# 标题\n\n这是一段简短说明。\n\n## 子标题\n\n- 列表项 1\n- 列表项 2\n"
        result = pre_trim_for_multica_comment(short)
        self.assertFalse(result["truncated"])
        self.assertEqual(result["trimmed_bytes"], result["original_bytes"])


class TestPreTrimNestedTable(unittest.TestCase):
    """嵌套表测试"""

    def test_nested_table_cells_collapsed(self):
        nested_table = (
            "| 字段 | 值 |\n|------|----|\n"
            + "| 配置 | `a=1|b=2|c=3|d=4|e=5|f=6|g=7|h=8|i=9|j=10|k=11` |\n"
            + "| 备注 | normal cell |\n"
        )
        result = pre_trim_for_multica_comment(nested_table)
        # 嵌套单元格被折叠为 `...`
        self.assertIn("`...`", result["trimmed"])
        self.assertIn("normal cell", result["trimmed"])

    def test_simple_table_preserved(self):
        simple_table = "| A | B |\n|---|---|\n| 1 | 2 |\n"
        result = pre_trim_for_multica_comment(simple_table)
        self.assertIn("| A | B |", result["trimmed"])
        self.assertIn("| 1 | 2 |", result["trimmed"])


class TestPreTrimMultipleCodeBlocks(unittest.TestCase):
    """多代码块测试"""

    def test_multiple_code_blocks_merged(self):
        text = (
            "段落1\n\n"
            "```python\nprint(1)\n```\n\n"
            "段落2\n\n"
            "```bash\necho 1\n```\n\n"
            "段落3\n\n"
            "```js\nconsole.log(1)\n```\n\n"
            "段落4\n\n"
            "```yaml\nfoo: bar\n```\n\n"
            "结尾"
        )
        result = pre_trim_for_multica_comment(text)
        # 应保留首尾两个代码块，中间合并
        self.assertIn("print(1)", result["trimmed"])
        self.assertIn("foo: bar", result["trimmed"])
        # 中间代码块应被省略（echo 1 和 console.log 不应出现）
        # 注意：合并后保留 first + last，所以 echo/console 可能保留
        self.assertIn("已省略", result["trimmed"])

    def test_two_code_blocks_preserved(self):
        text = "段落\n```python\nprint(1)\n```\n段落\n```bash\necho 1\n```\n"
        result = pre_trim_for_multica_comment(text)
        self.assertIn("print(1)", result["trimmed"])
        self.assertIn("echo 1", result["trimmed"])


class TestPreTrimMixedLanguage(unittest.TestCase):
    """中英混排测试"""

    def test_chinese_english_mixed_counted_by_bytes(self):
        text = "中文 Hello 混合 123 测试。\n" * 20
        result = pre_trim_for_multica_comment(text)
        # 中文 3 字节/字符，UTF-8 字节数正确
        self.assertGreater(result["original_bytes"], len(text))
        self.assertLessEqual(result["trimmed_bytes"], DEFAULT_MAX_BYTES)

    def test_chinese_only_no_truncation_under_4kb(self):
        # 约 1000 个中文字符 ≈ 3000 字节，未超 4KB
        text = "运维工程师严维序负责系统稳定性保障。" * 30
        result = pre_trim_for_multica_comment(text)
        self.assertFalse(result["truncated"])


class TestPreTrimMentionLinks(unittest.TestCase):
    """mention 链接测试"""

    def test_agent_mention_stripped(self):
        text = "[@徐聪](mention://agent/46bdd4a6-5c64-475a-92ef-36a763602fa1) 已就绪"
        result = pre_trim_for_multica_comment(text)
        self.assertIn("@徐聪", result["trimmed"])
        self.assertNotIn("mention://", result["trimmed"])

    def test_issue_mention_preserved(self):
        text = "[MUL-104](mention://issue/abc-123) 已派发"
        result = pre_trim_for_multica_comment(text)
        self.assertIn("MUL-104", result["trimmed"])


class TestPreTrimLogPersist(unittest.TestCase):
    """.log 落盘测试"""

    def test_log_persisted_when_path_provided(self, log_path="/tmp/test_pre_trim.log"):
        if os.path.exists(log_path):
            os.remove(log_path)
        long_text = "log test " * 200
        result = pre_trim_for_multica_comment(long_text, log_path=log_path)
        self.assertIsNotNone(result["log_path"])
        self.assertTrue(os.path.exists(log_path))
        with open(log_path, "r", encoding="utf-8") as f:
            content = f.read()
        self.assertIn("log test", content)
        # 清理
        os.remove(log_path)


class TestPreTrimVersion(unittest.TestCase):
    """版本戳测试（BIZ-38 合规）"""

    def test_version_stamp_present(self):
        self.assertIsNotNone(__version__)
        self.assertRegex(__version__, r"^\d+\.\d+\.\d+$")


class TestPreTrimCJKSafeBoundary(unittest.TestCase):
    """CJK-safe 截断边界测试（BIZ-104 Phase ④）"""

    def test_no_split_mid_rune(self):
        """确保截断点不在 CJK 多字节序列中间"""
        # 强制截断以验证边界点
        # 1 个中文字 = 3 字节
        # 200 个中文字 = 600 字节
        text = "中" * 200  # 600 字节
        result = pre_trim_for_multica_comment(text, max_bytes=100)
        # 验证输出是合法 UTF-8
        out_bytes = result["trimmed"].encode("utf-8")
        # 验证不能被错误丢弃的字节产生非法序列
        try:
            decoded = out_bytes.decode("utf-8")
            self.assertIsInstance(decoded, str)
        except UnicodeDecodeError:
            self.fail("Output contains invalid UTF-8")

    def test_cjk_density_warning_added(self):
        """高 CJK 密度时动作链含 cjk_density_warn"""
        # 200 个中文字 + 少量 ASCII，CJK 占比 > 95%
        text = "运维工程师严维序负责系统稳定性保障。" * 50
        result = pre_trim_for_multica_comment(text)
        cjk_warnings = [a for a in result["actions"] if "cjk_density" in a]
        self.assertGreater(len(cjk_warnings), 0)


if __name__ == "__main__":
    unittest.main(verbosity=2)