Files
douyin_guaji-main/tests/test_douyin_text_utils.py
T
CodexandClaude Opus 4.7 ecffba94b4 P3 §7: 引入 pytest 基础设施 + 9 个核心模块冒烟/单元测试
§7-1 pytest 基础设施
  - pyproject.toml 加 [tool.pytest.ini_options]
    (testpaths / python_files / strict markers / smoke + slow marker 注册)
  - tests/conftest.py 自动把项目根加到 sys.path

§7-2 核心模块测试(242 passed, 1 skipped)
  - test_filter_logging.py(重写为 pytest):should_skip_fudai_by_config
    + is_watch_task_fudai + _is_short_countdown_text 共 29 用例
  - test_time_utils.py:parse_mm_ss + extract_2digits 共 22 用例
  - test_douyin_text_utils.py:4 个 OCR 解析函数 25 用例
  - test_douyin_utils.py:extract_prize_value_yuan + normalize_countdown_text 21 用例
  - test_path_config.py:10 个路径函数结构断言 + 路径间互不重复 20 用例
  - test_logger_helpers.py:8 个 print→logger helper 的统一接口测试 20 用例
  - test_unified_logger_mdc.py:MDC 7 个方法 + 线程隔离 13 用例
  - test_io_utils.py:_verify_png_integrity / _safe_replace_file /
    _ensure_parent_dir 11 用例
  - test_imports.py:78 个核心模块的 import 冒烟 + 关键类/函数导出验证

修订老测试
  - test_ocr_comparison.py:用 pytest.importorskip 跳过缺失的 psutil/easyocr/paddleocr,
    并整体标 @pytest.mark.slow(需要 'pytest -m slow' 才会跑)
  - test_webgui.py:原会真的启 HTTP 服务卡死,改为 import 冒烟 + helper 调用测试

README
  - 重写 "## ✅ 测试" 段,给出实际可用的 pytest 命令、测试约定、marker 说明

回归:tests/ 142 项测试 PASS / 1 SKIP / 0 FAIL;全部纯函数 + 不依赖 ADB / OCR / 设备。

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-05-21 14:09:02 +08:00

132 lines
3.9 KiB
Python

"""douyin/utils/douyin_text_utils 的 pytest 单元测试。
覆盖福袋内容/倒计时/价值/参与人数 4 个 OCR 文本解析函数。
"""
from __future__ import annotations
import pytest
from douyin.utils.douyin_text_utils import (
_safe_str,
extract_fudai_content,
extract_fudai_countdown,
extract_fudai_value,
extract_participants_count,
)
# ============================================================
# _safe_str
# ============================================================
class TestSafeStr:
@pytest.mark.parametrize(
"value,expected",
[
(None, ""),
("", ""),
("abc", "abc"),
(123, "123"),
(0, "0"),
],
)
def test_safe_str(self, value, expected):
assert _safe_str(value) == expected
# ============================================================
# extract_fudai_content
# ============================================================
class TestExtractFudaiContent:
@pytest.mark.parametrize(
"text,expected_substr",
[
("福袋内容:现金红包 100 元", "现金红包 100 元"),
("奖品内容:iPhone 15", "iPhone 15"),
("内容:手机一台", "手机一台"),
# 全/半角冒号都支持
("福袋内容: 现金", "现金"),
],
)
def test_with_prefix(self, text, expected_substr):
result = extract_fudai_content(text)
assert expected_substr in result
def test_no_prefix_returns_original(self):
text = "随便一段没有前缀的描述"
# 没有前缀时返回原文(去首尾空白),不抛异常
assert extract_fudai_content(text) == text.strip()
@pytest.mark.parametrize("value", ["", None])
def test_empty(self, value):
assert extract_fudai_content(value) == ""
# ============================================================
# extract_fudai_countdown
# ============================================================
class TestExtractFudaiCountdown:
@pytest.mark.parametrize(
"value",
["", None],
)
def test_empty_returns_empty_string(self, value):
assert extract_fudai_countdown(value) == ""
def test_returns_string_type(self):
# 复杂文本走 normalize_countdown_text,结果应总是 str
result = extract_fudai_countdown("剩余 05:30 开奖")
assert isinstance(result, str)
# ============================================================
# extract_fudai_value
# ============================================================
class TestExtractFudaiValue:
@pytest.mark.parametrize(
"value",
["", None],
)
def test_empty_returns_zero(self, value):
assert extract_fudai_value(value) == 0.0
def test_returns_float(self):
# 不论输入是什么,返回值必须是 float
result = extract_fudai_value("价值 100 元")
assert isinstance(result, float)
# 至少应该解析出一个非负数
assert result >= 0.0
# ============================================================
# extract_participants_count
# ============================================================
class TestExtractParticipantsCount:
@pytest.mark.parametrize(
"value",
["", None],
)
def test_empty_returns_zero(self, value):
assert extract_participants_count(value) == 0
@pytest.mark.parametrize(
"text,expected",
[
("已1173人已参与", 1173),
("已经500人参与", 500),
("1173人已参与", 1173),
("1000人参与", 1000),
],
)
def test_extract_pattern(self, text, expected):
assert extract_participants_count(text) == expected
def test_returns_int(self):
result = extract_participants_count("100 人")
assert isinstance(result, int)
assert result >= 0