Python版spdlog的一些使用例子

    来了CSDN辣么多年，第一次写文章。博主还是个菜鸟，前段时间实习被公司要求研究一下Python版的spdlog怎么用来保存日志，在别的地方搜，但是搜到的资料特别少（大概是因为很少人用），然后自己搞了半天终于搞出了点结果，不过都是些简单的例子。先贡献上代码吧（现在大部分人用的是C++ 的，Python版的估计也没什么人看）

首先，下载安装spdlog。打开cmd，输入以下代码。（在pycharm直接搜索下载也可以）

pip install spdlog

1、在控制台输入日志，不保存文件：

name = 'aha?'  # 名字随便起
logger = ConsoleLogger(name, multithreaded=False, stdout=True, colored=False, async_mode=False)
# or logger = ConsoleLogger(name, False, True, False)  # multithreaded=False, stdout=True, colored=False
logger.trace('Trace what?')
logger.debug('Any thing wrong?')
logger.info('This is a piece of useless information')
logger.warn('Actually,no warn for you~')
logger.error('Always error')
logger.critical('I am so sleepy~~')

运行之后控制台就会输出logger的信息：

2、标准输出、标准错误输出、保存日志文件（普通文件、daily文件、rotate文件）

stdout_logger = spdlog.stdout_sink_st()
# or create a color logger
# stdout_logger = spdlog.stdout_color_sink_mt()
logger_1 = spdlog.SinkLogger("name", [stdout_logger])
logger_1.info("This is a stdout_logger1.")
# info can be replaced by warn, error, critical if it's needed...

stderr_logger = spdlog.stderr_sink_st()
# or create a color logger
# stderr_logger = spdlog.stderr_color_sink_mt()
logger_2 = spdlog.SinkLogger("name", [stderr_logger])
logger_2.info("This is a stderr_logger2.")
# info can be replaced by warn, error, critical if it's needed...

# create a basic logger
spdlog.basic_file_sink_st('filename.txt')
# multithreaded=false, truncate=false
logger_3 = spdlog.FileLogger('name', 'filename.txt', False,
                                     False)
logger_3.info('You have create a basic logger.')
# info can be replaced by warn, error, critical if it's needed...

# hours=8,minutes=30
daily_logger = spdlog.daily_file_sink_st('filename_daily.log', 8, 30)
# update at 8:30am
logger_4 = spdlog.SinkLogger('name', [daily_logger])
logger_4.info('This is a daily logger.')
# info can be replaced by warn, error, critical if it's needed...

spdlog.rotating_file_sink_st('filename_rotate.log', 1024, 5)  
# create a rotate file, size=1024, file number=5

# multithreaded=false, async_mode=False
logger_5 = spdlog.RotatingLogger('name1', 'filename_rotate.log', False, 1024, 5, False)
logger_5.info('This is a rotate logger.')
# info can be replaced by warn, error, critical if it's needed...

spdlog.stdout_sink_st() 和spdlog.stderr_sink_st()仅在控制台输出，其余输出保存在日志文件，位于根目录下面。

 文件内容：

3、设置级别

from spdlog import ConsoleLogger, LogLevel
import unittest


def log_msg(logger, level):
    if level == LogLevel.TRACE:
        logger.trace('I am Trace')
    elif level == LogLevel.DEBUG:
        logger.debug('I am Debug')
    elif level == LogLevel.INFO:
        logger.info('I am Info')
    elif level == LogLevel.WARN:
        logger.warn('I am Warning')
    elif level == LogLevel.ERR:
        logger.error('I am Error')
    else:
        logger.critical('I am Critical')


def set_log_level(logger, level):
    print("Setting Log level to %d" % level)
    logger.set_level(level)


class Test(unittest.TestCase):
    def test_log_level(self):
        # multithreaded=False, stdout=True, colored=False
        logger = ConsoleLogger('Logger', False, True, False)
        for level in (LogLevel.TRACE, LogLevel.DEBUG, LogLevel.INFO, LogLevel.WARN,
                      LogLevel.ERR, LogLevel.CRITICAL):
            set_log_level(logger, level)
            log_msg(logger, level)


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

结果：

在C++中，级别就是trace、debug这些，但是Python级别分别是结果中对应的0-5 

4、通过等级设置日志信息何时出现在文件何时出现在控制台的例子

import spdlog
import unittest


class MyTestCase(unittest.TestCase):
    def test_something(self):
        spdlog.basic_file_sink_st('asynclogger.txt')  # create a basic logger

        logger = spdlog.FileLogger('name', 'asynclogger.txt', False,
                                   False)  # multithreaded=false, truncate=false
        logger.async_mode()  # setting async model
        # queue_size = 2048, thread_count = 4, overflow_policy = 1
        # logger.spdlog.set_async_mode(2048, 4, 1)
        for i in range(100):
            logger.info('You have create a async logger. test_%d' % i)

        # multi sink example
        console_sink = spdlog.stdout_sink_mt()

        # level: trace=0, debug=1, info=2, warn=3, error=4, critical=5, off=6
        console_sink.set_level(3)

        file_sink = spdlog.basic_file_sink_mt('basic.txt')
        file_sink.set_level(0)
        sink = [file_sink, console_sink]
        logger_multi = spdlog.SinkLogger('multi_sink', sink)
        logger_multi.set_level(1)

        # logger.set_pattern('{:%Y-%m-%d --- %H:%M:%S}', spdlog.PatternTimeType(0))
        # while use set_pattern, inside '' must be string

        logger_multi.warn('This should appear in both console and file?')
        logger_multi.info('This message should not appear in the console, only in the file')
        # end of multi sink example

        # binary example
        num = [10, 20, 30, 40]
        binary_logger = spdlog.stdout_sink_st()
        logger_n = spdlog.SinkLogger("name", [binary_logger])
        logger_n.info('{:02X} {:02X} {:02X} {:02X}'.format(*num))
        logger_n.info('{:#X} {:#b} {:#e} {:#o}'.format(*num))
        # end of binary example

        get_logger = spdlog.get('name')
        get_logger.info('test spdlog.get, and this will be showed in name')  # write into 'name' logger

        get_logger.flush()
        get_logger.flush_on(2)  # number is level
        get_logger.set_error_handler(Warning)  # I am not sure

        # null_sink seams to do nothing
        null_logger = spdlog.null_sink_mt()
        logger_1 = spdlog.SinkLogger("null", [null_logger])
        logger_1.info("Notice: this will not be showed.")
        get_logger = spdlog.get('null')
        get_logger.debug('test spdlog.get')


if __name__ == '__main__':
    unittest.main()

控制台结果：

 保存的日志文件：

 basic文件的内容：

以上。禁止转载。

代码已上传GitHub

https://github.com/chole-zhou/spdlog_python