Pytest----caplog的应用场景以及使用方法

【原文链接】Pytest----caplog的应用场景以及使用方法

如何在测试用例中设置日志级别

通过caplog可以对特定的测试函数内设置日志级别，而不影响全局的日志级别，比如如下，首先在pytest.ini中开启实时日志。

[pytest]
log_cli = True

然后测试代码中test_demo和test_demo2中均打印debug、info、warning、error、critical级别的日志，不同的是在test_demo2中通过caplog临时将日志级别设置为debug级别，因为在默认情况下，日志级别为warning，因此根据理论分析，test_demo中的日志打印将采用默认的日志级别，即只会显示warning、error、critical级别的日志，而在test_demo2中由于临时将日志级别设置为debug，因此这里所有的日志都将显示处理。

import logging

def test_demo():
    logging.debug("this is debug log ...")
    logging.info("this is info log ...")
    logging.warning("this is warning log ...")
    logging.error("this is error log ...")
    logging.critical("this is critical log ...")
    assert 1==1

def test_demo2(caplog):
    caplog.set_level(logging.DEBUG)
    logging.debug("this is debug log ...")
    logging.info("this is info log ...")
    logging.warning("this is warning log ...")
    logging.error("this is error log ...")
    logging.critical("this is critical log ...")
    assert 1==1

执行结果如下，可以看出，在test_demo中只显示了warning、error、critical级别的日志，而在test_demo2中则所有的日志都显示出来了，即此时的日志级别已经通过caplog修改为debug级别了。

(demo-HCIhX0Hq) E:\demo>pytest
=================== test session starts ===================
platform win32 -- Python 3.7.9, pytest-7.2.0, pluggy-1.0.0
rootdir: E:\demo, configfile: pytest.ini
plugins: assume-2.4.3, rerunfailures-10.2
collected 2 items

test_demo.py::test_demo
---------------------- live log call ----------------------
WARNING  root:test_demo.py:6 this is warning log ...
ERROR    root:test_demo.py:7 this is error log ...
CRITICAL root:test_demo.py:8 this is critical log ...
PASSED                                               [ 50%]
test_demo.py::test_demo2
---------------------- live log call ----------------------
DEBUG    root:test_demo.py:13 this is debug log ...
INFO     root:test_demo.py:14 this is info log ...
WARNING  root:test_demo.py:15 this is warning log ...
ERROR    root:test_demo.py:16 this is error log ...
CRITICAL root:test_demo.py:17 this is critical log ...
PASSED                                               [100%]

==================== 2 passed in 0.03s ====================

(demo-HCIhX0Hq) E:\demo>

如何对日志级别进行断言

caplog会将日志都记录在records属性中，这样一来就可以在测试脚本末尾，对当前测试脚本中产生的日志级别进行判断，比如脚本中可能存在当某些条件触发时写入error日志，而在脚本末尾则可以对日志级别进行断言，比如要求日志不能有error日志，则可以使用类似如下测试代码。

import logging

def test_demo(caplog):
    logging.warning("this is warning log ...")
    logging.error("this is error log ...")
    logging.critical("this is critical log ...")
    for record in caplog.records:
        assert record.levelname != "ERROR"

执行结果如下，

(demo-HCIhX0Hq) E:\demo>pytest
=================== test session starts ===================
platform win32 -- Python 3.7.9, pytest-7.2.0, pluggy-1.0.0
rootdir: E:\demo, configfile: pytest.ini
plugins: assume-2.4.3, rerunfailures-10.2
collected 1 item

test_demo.py::test_demo
---------------------- live log call ----------------------
WARNING  root:test_demo.py:4 this is warning log ...
ERROR    root:test_demo.py:5 this is error log ...
CRITICAL root:test_demo.py:6 this is critical log ...
FAILED                                               [100%]

======================== FAILURES =========================
________________________ test_demo ________________________

caplog = <_pytest.logging.LogCaptureFixture object at 0x0000029BEF29C608>

    def test_demo(caplog):
        logging.warning("this is warning log ...")
        logging.error("this is error log ...")
        logging.critical("this is critical log ...")
        for record in caplog.records:
>           assert record.levelname != "ERROR"
E           assert 'ERROR' != 'ERROR'
E            +  where 'ERROR' = <LogRecord: root, 40, E:\demo\test_demo.py, 5, "this is error log ...">.levelname

test_demo.py:8: AssertionError
-------------------- Captured log call --------------------
WARNING  root:test_demo.py:4 this is warning log ...
ERROR    root:test_demo.py:5 this is error log ...
CRITICAL root:test_demo.py:6 this is critical log ...
================= short test summary info =================
FAILED test_demo.py::test_demo - assert 'ERROR' != 'ERROR'
==================== 1 failed in 0.08s ====================

(demo-HCIhX0Hq) E:\demo>

如何对日志内容进行断言

caplog同样可以做到对日志的内容进行断言，比如如下测试代码，判断日志中是否有 error log 内容。即通过record的message即可获得日志的内容。

import logging

def test_demo(caplog):
    logging.warning("this is warning log ...")
    logging.error("this is error log ...")
    logging.critical("this is critical log ...")
    for record in caplog.records:
        assert "error log" not in record.meesge

执行结果如下，可以看到断言错误，即日志中含有 error log 内容。

(demo-HCIhX0Hq) E:\demo>pytest -s
=================== test session starts ===================
platform win32 -- Python 3.7.9, pytest-7.2.0, pluggy-1.0.0
rootdir: E:\demo, configfile: pytest.ini
plugins: assume-2.4.3, rerunfailures-10.2
collected 1 item

test_demo.py::test_demo
---------------------- live log call ----------------------
WARNING  root:test_demo.py:4 this is warning log ...
ERROR    root:test_demo.py:5 this is error log ...
CRITICAL root:test_demo.py:6 this is critical log ...
FAILED

======================== FAILURES =========================
________________________ test_demo ________________________

caplog = <_pytest.logging.LogCaptureFixture object at 0x00000260C060D6C8>

    def test_demo(caplog):
        logging.warning("this is warning log ...")
        logging.error("this is error log ...")
        logging.critical("this is critical log ...")
        for record in caplog.records:
>           assert "error log" not in record.meesge
E           AttributeError: 'LogRecord' object has no attribute 'meesge'

test_demo.py:8: AttributeError
-------------------- Captured log call --------------------
WARNING  root:test_demo.py:4 this is warning log ...
ERROR    root:test_demo.py:5 this is error log ...
CRITICAL root:test_demo.py:6 this is critical log ...
================= short test summary info =================
FAILED test_demo.py::test_demo - AttributeError: 'LogRecord' object has no attribute 'me...
==================== 1 failed in 0.08s ====================

(demo-HCIhX0Hq) E:\demo>

如何同时对日志级别和日志内容进行断言

除此以外，caplog还可以对logger，日志级别，日志内容组成的元组进行判决，caplog.record_tuples 保存了所有logger、日志级别、日志内容组成的元组集合。
如下测试代码，这里有一点需要注意的是，当直接使用logging.xxx打印日志的时候，实际上使用的logger是root，这一点需要注意，若对python中logging有更深入的理解，比如定义自己的logger，则在此处就会很容易理解了。

import logging

def test_demo(caplog):
    logging.warning("this is warning log ...")
    logging.error("this is error log ...")
    logging.critical("this is critical log ...")
    assert ("root",logging.ERROR,"this is error log ...") in caplog.record_tuples

执行结果如下，可以看到此时断言成功。

(demo-HCIhX0Hq) E:\demo>pytest
=================== test session starts ===================
platform win32 -- Python 3.7.9, pytest-7.2.0, pluggy-1.0.0
rootdir: E:\demo, configfile: pytest.ini
plugins: assume-2.4.3, rerunfailures-10.2
collected 1 item

test_demo.py::test_demo
---------------------- live log call ----------------------
WARNING  root:test_demo.py:4 this is warning log ...
ERROR    root:test_demo.py:5 this is error log ...
CRITICAL root:test_demo.py:6 this is critical log ...
PASSED                                               [100%]

==================== 1 passed in 0.02s ====================

(demo-HCIhX0Hq) E:\demo>