Spaces:

jun2024
/

fachuan

Configuration error

fachuan / tests /ci /unit /test_text_parser_coverage.py
jun2024
Upload 2241 files
f8b03c1 verified 7 months ago
24.6 kB
"""Unit tests for apps.client.services.text_parser."""

from __future__ import annotations

import pytest

from apps.client.services.text_parser import (
    _determine_client_type,
    _extract_address,
    _extract_credit_code,
    _extract_id_number,
    _extract_legal_representative,
    _extract_name_smart,
    _extract_parties,
    _extract_phone,
    parse_client_text,
    parse_multiple_clients_text,
)

# ── parse_client_text ─────────────────────────────────────────────────────────

FULL_NATURAL_TEXT = (
    "姓名：张三\n"
    "身份证号码：100000000000000001\n"
    "地址：北京市朝阳区建国路1号\n"
    "联系电话：00000000000"
)

FULL_LEGAL_TEXT = (
    "名称：北京测试科技有限公司\n"
    "统一社会信用代码：91110000MA01ABCD12\n"
    "地址：北京市海淀区中关村大街1号\n"
    "联系电话：010-12345678\n"
    "法定代表人：李四"
)


def test_parse_client_text_natural_person() -> None:
    result = parse_client_text(FULL_NATURAL_TEXT)
    assert result["name"] == "张三"
    assert result["id_number"] == "100000000000000001"
    assert "北京市朝阳区" in result["address"]
    assert result["phone"] == "00000000000"
    assert result["client_type"] == "natural"


def test_parse_client_text_legal_entity() -> None:
    result = parse_client_text(FULL_LEGAL_TEXT)
    assert "测试科技有限公司" in result["name"]
    assert result["id_number"] == "91110000MA01ABCD12"
    assert result["legal_representative"] == "李四"
    assert result["client_type"] == "legal"


def test_parse_client_text_empty_returns_empty_result() -> None:
    result = parse_client_text("")
    assert result["name"] == ""
    assert result["client_type"] == "natural"


def test_parse_client_text_whitespace_only_returns_empty_result() -> None:
    result = parse_client_text("   \n  ")
    assert result["name"] == ""


# ── parse_multiple_clients_text ───────────────────────────────────────────────

MULTI_CLIENT_TEXT = (
    "原告：张三，男，身份证号码：100000000000000001，地址：北京市朝阳区建国路1号\n"
    "被告：北京测试科技有限公司，统一社会信用代码：91110000MA01ABCD12，法定代表人：李四"
)


def test_parse_multiple_clients_text_returns_two_parties() -> None:
    results = parse_multiple_clients_text(MULTI_CLIENT_TEXT)
    assert len(results) >= 2


def test_parse_multiple_clients_text_empty_returns_empty_list() -> None:
    assert parse_multiple_clients_text("") == []


def test_parse_multiple_clients_text_whitespace_returns_empty_list() -> None:
    assert parse_multiple_clients_text("   ") == []


# ── _extract_name_smart ───────────────────────────────────────────────────────

def test_extract_name_smart_from_role_label() -> None:
    text = "原告：张三\n身份证号码：100000000000000001"
    assert _extract_name_smart(text) == "张三"


def test_extract_name_smart_legal_entity() -> None:
    text = "名称：北京测试科技有限公司\n统一社会信用代码：91110000MA01ABCD12"
    result = _extract_name_smart(text)
    assert result is not None
    assert "有限公司" in result


def test_extract_name_smart_natural_person_with_gender() -> None:
    text = "李四，男，身份证号码：100000000000000001"
    result = _extract_name_smart(text)
    assert result == "李四"


def test_extract_name_smart_empty_returns_none() -> None:
    assert _extract_name_smart("") is None


def test_extract_name_smart_only_keywords_returns_none() -> None:
    result = _extract_name_smart("联系电话：00000000000")
    # 不应把"联系电话"当名称
    assert result != "联系电话"


# ── _extract_parties ──────────────────────────────────────────────────────────

def test_extract_parties_single_party() -> None:
    text = "原告：张三\n身份证号码：100000000000000001"
    parties = _extract_parties(text)
    assert len(parties) >= 1
    assert parties[0]["name"] == "张三"


def test_extract_parties_plaintiff_defendant() -> None:
    text = (
        "原告：张三\n身份证号码：100000000000000001\n"
        "被告：李四\n身份证号码：200000000000000002"
    )
    parties = _extract_parties(text)
    names = [p["name"] for p in parties]
    assert "张三" in names
    assert "李四" in names


def test_extract_parties_no_role_label_falls_back() -> None:
    text = "张三\n身份证号码：100000000000000001"
    parties = _extract_parties(text)
    # 无角色标签时兜底解析，可能返回 0 或 1 个
    assert isinstance(parties, list)


# ── _extract_credit_code ──────────────────────────────────────────────────────

def test_extract_credit_code_with_label() -> None:
    text = "统一社会信用代码：91110000MA01ABCD12"
    assert _extract_credit_code(text) == "91110000MA01ABCD12"


def test_extract_credit_code_fallback_with_letters() -> None:
    text = "91110000MA01ABCD12"
    assert _extract_credit_code(text) == "91110000MA01ABCD12"


def test_extract_credit_code_not_confused_with_id_number() -> None:
    text = "身份证号码：100000000000000001"
    # 纯数字18位不应被识别为信用代码
    result = _extract_credit_code(text)
    assert result is None or (result is not None and any(ch.isalpha() for ch in result))


def test_extract_credit_code_none_when_absent() -> None:
    assert _extract_credit_code("张三，男") is None


# ── _extract_id_number ────────────────────────────────────────────────────────

def test_extract_id_number_with_label() -> None:
    text = "身份证号码：100000000000000001"
    assert _extract_id_number(text) == "100000000000000001"


def test_extract_id_number_fallback() -> None:
    text = "100000000000000001"
    assert _extract_id_number(text) == "100000000000000001"


def test_extract_id_number_none_when_absent() -> None:
    assert _extract_id_number("北京市朝阳区") is None


# ── _extract_address ──────────────────────────────────────────────────────────

def test_extract_address_with_label() -> None:
    text = "地址：北京市朝阳区建国路1号\n联系电话：00000000000"
    result = _extract_address(text)
    assert result is not None
    assert "北京市朝阳区" in result


def test_extract_address_fallback_province_line() -> None:
    text = "北京市朝阳区建国路1号院2单元301室"
    result = _extract_address(text)
    assert result is not None


def test_extract_address_none_when_absent() -> None:
    assert _extract_address("张三，男") is None


# ── _extract_phone ────────────────────────────────────────────────────────────

def test_extract_phone_with_label() -> None:
    text = "联系电话：00000000000"
    assert _extract_phone(text) == "00000000000"


def test_extract_phone_fallback_mobile() -> None:
    text = "张三 13900139000"  # nosec
    result = _extract_phone(text)
    assert result == "13900139000"  # nosec


def test_extract_phone_none_when_absent() -> None:
    assert _extract_phone("张三，男") is None


# ── _extract_legal_representative ────────────────────────────────────────────

def test_extract_legal_representative_with_label() -> None:
    text = "法定代表人：王五\n地址：北京市"
    result = _extract_legal_representative(text)
    assert result == "王五"


def test_extract_legal_representative_none_when_absent() -> None:
    assert _extract_legal_representative("张三，男") is None


# ── _determine_client_type ────────────────────────────────────────────────────

def test_determine_client_type_legal_by_credit_code() -> None:
    text = "统一社会信用代码：91110000MA01ABCD12"
    assert _determine_client_type("北京测试有限公司", text) == "legal"


def test_determine_client_type_legal_by_legal_rep() -> None:
    text = "法定代表人：王五"
    assert _determine_client_type("某公司", text) == "legal"


def test_determine_client_type_legal_by_name_keyword() -> None:
    assert _determine_client_type("北京测试有限公司", "") == "legal"


def test_determine_client_type_natural_by_id_number() -> None:
    text = "身份证号码：100000000000000001"
    assert _determine_client_type("张三", text) == "natural"


def test_determine_client_type_natural_default() -> None:
    assert _determine_client_type("张三", "") == "natural"


# ── 边界情况 ──────────────────────────────────────────────────────────────────

def test_parse_client_text_name_only() -> None:
    result = parse_client_text("张三")
    # 只有名称，其他字段为空
    assert result["name"] == "张三"
    assert result["id_number"] == ""
    assert result["phone"] == ""


def test_parse_client_text_invalid_input_no_crash() -> None:
    result = parse_client_text("!!@@##$$")
    assert isinstance(result, dict)
    assert "name" in result


# ── _extract_name_smart 次级兜底分支 ─────────────────────────────────────────

def test_extract_name_smart_jiafang_format() -> None:
    # 甲方：xxx 格式（_SMART_NAME_PATTERN）
    text = "甲方：北京测试科技有限公司\n统一社会信用代码：91110000MA01ABCD12"
    result = _extract_name_smart(text)
    assert result is not None
    assert "有限公司" in result


def test_extract_name_smart_leading_name_before_field() -> None:
    # 名称直接在字段前（_LEADING_NAME_BEFORE_FIELD_PATTERN）
    text = "北京测试科技有限公司统一社会信用代码：91110000MA01ABCD12"
    result = _extract_name_smart(text)
    assert result is not None


def test_extract_name_smart_role_no_colon() -> None:
    # 无冒号角色写法：被告 张三，男（_ROLE_NAME_FALLBACK_PATTERN）
    text = "被告 张三，男，地址：北京市"
    result = _extract_name_smart(text)
    assert result == "张三"


def test_extract_name_smart_leading_person_before_gender() -> None:
    # 自然人名在性别前（_LEADING_PERSON_NAME_PATTERN）
    text = "王小明，男，身份证号码：100000000000000001"
    result = _extract_name_smart(text)
    assert result == "王小明"


def test_extract_name_smart_first_line_fallback() -> None:
    # 兜底：第一行即名称
    text = "某某律师事务所\n联系电话：00000000000"
    result = _extract_name_smart(text)
    assert result is not None
    assert "律师事务所" in result


def test_parse_client_text_falls_back_to_direct_parse() -> None:
    # 无角色标签时走 _parse_fields_directly（覆盖 line 190）
    text = "姓名：张三\n联系电话：00000000000"
    result = parse_client_text(text)
    assert result["name"] == "张三"


# ── _is_valid_name_candidate 边缘分支 ────────────────────────────────────────

def test_is_valid_name_candidate_rejects_pure_digits() -> None:
    from apps.client.services.text_parser import _is_valid_name_candidate
    assert not _is_valid_name_candidate("12345678")


def test_is_valid_name_candidate_rejects_id_number_format() -> None:
    from apps.client.services.text_parser import _is_valid_name_candidate
    assert not _is_valid_name_candidate("100000000000000001")


def test_is_valid_name_candidate_rejects_credit_code_with_letters() -> None:
    from apps.client.services.text_parser import _is_valid_name_candidate
    assert not _is_valid_name_candidate("91110000MA01ABCD12")


def test_is_valid_name_candidate_rejects_role_prefix() -> None:
    from apps.client.services.text_parser import _is_valid_name_candidate

    # "原" 是 "原告" 的前缀
    assert not _is_valid_name_candidate("原告")


def test_is_valid_name_candidate_rejects_non_name_keyword() -> None:
    from apps.client.services.text_parser import _is_valid_name_candidate
    assert not _is_valid_name_candidate("联系电话")


# ── 补充覆盖剩余分支 ──────────────────────────────────────────────────────────

def test_extract_name_smart_smart_name_pattern_valid() -> None:
    # 甲方：xxx 格式，且候选值有效（覆盖 276-278）
    # 需要确保 _extract_name 和 _NAME_FIELD_PATTERN 都不先命中
    text = "甲方：某律师事务所\n地址：北京市"
    result = _extract_name_smart(text)
    assert result is not None
    assert "律师事务所" in result


def test_extract_name_smart_leading_name_valid() -> None:
    # leading name before field 命中且有效（覆盖 289-291）
    text = "某某贸易有限公司地址：北京市朝阳区"
    result = _extract_name_smart(text)
    assert result is not None


def test_extract_name_smart_role_fallback_valid() -> None:
    # 无冒号角色写法命中且有效（覆盖 297）
    text = "\n被告 王小二 地址：北京市"
    result = _extract_name_smart(text)
    assert result is not None


def test_extract_name_smart_leading_person_valid() -> None:
    # leading person name 命中且有效（覆盖 302-304）
    text = "赵小明，男，地址：北京市"
    result = _extract_name_smart(text)
    assert result == "赵小明"


def test_extract_name_smart_legal_entity_only() -> None:
    # 仅公司全称，无任何标签（覆盖 309-311）
    text = "深圳某某科技股份有限公司"
    result = _extract_name_smart(text)
    assert result is not None
    assert "有限公司" in result


def test_extract_name_from_first_meaningful_line_returns_line() -> None:
    # 第一行有效名称（覆盖 334）
    from apps.client.services.text_parser import _extract_name_from_first_meaningful_line
    result = _extract_name_from_first_meaningful_line("某某律师\n联系电话：00000000000")
    assert result == "某某律师"


def test_is_valid_name_candidate_empty_compact() -> None:
    # compact 为空（覆盖 345）
    from apps.client.services.text_parser import _is_valid_name_candidate
    assert not _is_valid_name_candidate("  ")


def test_extract_name_returns_none_when_no_role_label() -> None:
    # _extract_name 无角色标签时返回 None（覆盖 458）
    from apps.client.services.text_parser import _extract_name
    assert _extract_name("北京市朝阳区某路1号") is None


def test_extract_credit_code_fallback_skips_near_id_keyword() -> None:
    # 信用代码前有"身份证"关键词时跳过，但后面有合法信用代码（覆盖 497）
    text = "91110000MA01ABCD12"
    result = _extract_credit_code(text)
    assert result == "91110000MA01ABCD12"


# ── 精确覆盖剩余分支 ──────────────────────────────────────────────────────────

def test_parse_client_text_no_role_label_direct_parse() -> None:
    # 无角色标签，_extract_parties 返回空，走 _parse_fields_directly（覆盖 190, 226）
    # 需要确保文本不含任何角色标签，且能提取到名称
    text = "姓名：王小花\n联系电话：00000000000\n地址：北京市朝阳区"
    result = parse_client_text(text)
    assert result["name"] == "王小花"


def test_extract_name_smart_smart_pattern_no_earlier_match() -> None:
    # 甲方格式，且前面所有模式都不命中（覆盖 276-278）
    # 确保 _extract_name、_NAME_FIELD_PATTERN 都不命中
    text = "甲方：某某律师事务所地址：北京市"
    result = _extract_name_smart(text)
    assert result is not None


def test_extract_name_smart_leading_before_field_no_earlier_match() -> None:
    # leading name before field，且前面模式都不命中（覆盖 289-291）
    text = "某某贸易公司法定代表人：王五"
    result = _extract_name_smart(text)
    assert result is not None


def test_extract_name_smart_role_fallback_no_earlier_match() -> None:
    # 无冒号角色写法，且前面模式都不命中（覆盖 297）
    text = "\n申请人 王小二\n地址：北京市"
    result = _extract_name_smart(text)
    assert result is not None


def test_extract_name_smart_legal_entity_no_earlier_match() -> None:
    # 仅公司全称，无任何标签和字段（覆盖 309-311）
    text = "广州某某贸易有限公司"
    result = _extract_name_smart(text)
    assert result is not None
    assert "有限公司" in result


def test_extract_name_from_first_line_valid() -> None:
    # 第一行是有效名称（覆盖 334）
    from apps.client.services.text_parser import _extract_name_from_first_meaningful_line
    result = _extract_name_from_first_meaningful_line("\n\n某某律师\n联系电话：00000000000")
    assert result == "某某律师"


def test_is_valid_name_candidate_digit_string() -> None:
    # compact.isdigit() 为 True（覆盖 349）
    from apps.client.services.text_parser import _is_valid_name_candidate
    assert not _is_valid_name_candidate("123456")


def test_is_valid_name_candidate_valid_name_returns_true() -> None:
    # 所有检查通过，return True（覆盖 360）
    from apps.client.services.text_parser import _is_valid_name_candidate
    assert _is_valid_name_candidate("某某律师事务所")


def test_extract_name_no_role_label_returns_none() -> None:
    # 无任何角色标签，_extract_name 返回 None（覆盖 458）
    from apps.client.services.text_parser import _extract_name
    result = _extract_name("地址：北京市朝阳区某路1号\n联系电话：00000000000")
    assert result is None


def test_extract_credit_code_fallback_returns_code_with_letters() -> None:
    # 无标签但含字母的18位编码，且前面没有"身份证"关键词（覆盖 497）
    from apps.client.services.text_parser import _extract_credit_code
    result = _extract_credit_code("企业代码：91110000MA01ABCD12")
    assert result == "91110000MA01ABCD12"


# ── 精确覆盖最后剩余分支 ──────────────────────────────────────────────────────

def test_parse_client_text_no_name_falls_back_to_direct_parse() -> None:
    # _extract_parties 返回空（无名称），走 _parse_fields_directly（覆盖 190, 226）
    text = "联系电话：00000000000\n地址：北京市朝阳区"
    result = parse_client_text(text)
    # 无名称，但不崩溃
    assert result["name"] == ""
    assert result["phone"] == "00000000000"


def test_is_valid_name_candidate_short_digit_string() -> None:
    # compact.isdigit() 为 True 且不被身份证 pattern 匹配（覆盖 349）
    from apps.client.services.text_parser import _is_valid_name_candidate
    assert not _is_valid_name_candidate("12345678")


# ── 精确覆盖 annotate 发现的剩余 ! 行 ────────────────────────────────────────

def test_parse_single_party_natural_with_legal_rep_upgrades_to_legal() -> None:
    # natural 类型有法定代表人时升级为 legal（覆盖 _parse_single_party 中 client_type=legal）
    from apps.client.services.text_parser import _parse_single_party
    text = "原告：张三\n身份证号码：100000000000000001\n法定代表人：王五"
    result = _parse_single_party(text, use_smart_name=True)
    assert result["client_type"] == "legal"
    assert result["legal_representative"] == "王五"


def test_is_valid_name_candidate_id_number_fullmatch() -> None:
    # _ID_NUMBER_FALLBACK_PATTERN.fullmatch 命中（覆盖 349 的 return False）
    from apps.client.services.text_parser import _is_valid_name_candidate

    # 18位纯数字，符合身份证格式
    assert not _is_valid_name_candidate("100000000000000001")


def test_extract_credit_code_skips_near_id_keyword() -> None:
    # 信用代码前有"身份证"关键词时 continue（覆盖 497 的 continue 行）
    from apps.client.services.text_parser import _extract_credit_code

    # 身份证关键词在信用代码前20字符内
    text = "身份证91110000MA01ABCD12"
    result = _extract_credit_code(text)
    # 被跳过，返回 None
    assert result is None


def test_extract_name_smart_role_fallback_valid_name() -> None:
    # _ROLE_NAME_FALLBACK_PATTERN 命中且 valid（覆盖 297 body）
    # 需要确保 _extract_name、NAME_FIELD、SMART_NAME、LEADING_NAME 都不命中
    from apps.client.services.text_parser import _extract_name_smart

    # 纯角色+名称，无冒号，无其他字段
    text = "被申请人 某某律师"
    result = _extract_name_smart(text)
    assert result is not None


def test_extract_name_smart_natural_person_pattern() -> None:
    # _NATURAL_PERSON_NAME_PATTERN 命中（"姓名，性别"格式）
    from apps.client.services.text_parser import _extract_name_smart

    # 需要前面所有模式都不命中
    text = "王小明，男"
    result = _extract_name_smart(text)
    assert result == "王小明"


# ── 最后 9 行精确覆盖 ─────────────────────────────────────────────────────────

def test_is_valid_name_too_short() -> None:
    from apps.client.services.text_parser import _is_valid_name_candidate
    assert not _is_valid_name_candidate("张")  # len < 2，覆盖 289 return False


def test_is_valid_name_id_number_fullmatch_rejected() -> None:
    from apps.client.services.text_parser import _is_valid_name_candidate

    # 15位身份证格式，fullmatch 命中（覆盖 349 return False）
    assert not _is_valid_name_candidate("110101900101123")


def test_extract_credit_code_skips_when_id_keyword_before() -> None:
    from apps.client.services.text_parser import _extract_credit_code

    # "身份证" 在信用代码前 20 字符内，触发 continue（覆盖 497 continue）
    # 后面再跟一个合法信用代码让函数能 return
    text = "身份证91110000MA01ABCD12 然后 91220000MB02EFGH34"
    result = _extract_credit_code(text)
    assert result == "91220000MB02EFGH34"


def test_extract_name_smart_leading_name_before_field_valid() -> None:
    from apps.client.services.text_parser import _extract_name_smart

    # 触发 289-291：_extract_name=None, NAME_FIELD=None, SMART_NAME=None
    # LEADING_NAME_BEFORE_FIELD 命中且 valid
    text = "某某贸易公司\n法定代表人：王五"
    result = _extract_name_smart(text)
    assert result is not None


def test_extract_name_smart_legal_entity_pattern_valid() -> None:
    from apps.client.services.text_parser import _extract_name_smart

    # 触发 309-311：前5个模式都不命中，LEGAL_ENTITY 命中
    # 纯公司名，无任何标签、字段、角色
    text = "北京某某科技股份有限公司"
    result = _extract_name_smart(text)
    assert result is not None
    assert "有限公司" in result


def test_extract_name_from_first_meaningful_line_skips_empty() -> None:
    from apps.client.services.text_parser import _extract_name_from_first_meaningful_line

    # 覆盖 334：跳过空行后返回有效名称
    result = _extract_name_from_first_meaningful_line("\n\n\n某某律师事务所")
    assert result == "某某律师事务所"