如何关闭日志语句而不从代码中删除它们

Question

If there are logging statements spread throughout a codebase, how do I set up the logger so I don't have to comment out each call to the logger when deploying the code into production?

Here's my current code:

import logging


logging.basicConfig(filename='./example.log', level=logging.DEBUG, 
                    format='%(asctime)s %(name)-12s %(levelname)-8s %(message)s',
                    datefmt='%m-%d %H:%M')

logging.debug('debug failed')
logging.info('info failed')
logging.warning('A warning')

Answer 1

Instead of using the basicConfig , you can set up the logger more explicitly with the handlers you want, based on any criteria.

import logging

log = logging.getLogger("FOO")
log.setLevel(logging.DEBUG)

# needs a handler
log.info('info')
#No handlers could be found for logger "FOO"

ch = logging.StreamHandler()
log.addHandler(ch)
log.info('info')
# info

log.removeHandler(ch)

noop = logging.NullHandler()
log.addHandler(noop)
# nothing happens here
log.info('info')

You can have a conditional statement that either adds the handler you want if you are running in debug mode, or you can add a NullHandler that just absorbs the log messages. You can also configure the levels individually of each handler, so that you would always see warnings and above. Each handler can have its own level, in addition to the main logger.

You can refer to the tutorials on how to get more specific with levels, handlers, and formatting.

Answer 2

There are a couple of really simple answers here. The first is to simply comment out the basicConfig(...) statement. That will have the effect not setting up any of the loggers, handlers or formatters which means your debug(), info(), etc calls will be effectively no-ops.

Another simple answer is to set the logging level in the basicConfig() call to something higher than DEBUG . CRITICAL + 1 would make sure that you never see a log message.

However you mentioned something about moving this code into production, so what you probably want to do is provide -q and -v command line options (assuming this is a CLI tool). My usual approach is to start at WARNING level, and for each -q move towards quieter logging by increasing the filter level. Conversely, for each -v , move towards more verbose logging. Here is a snippet of code that does exactly that.

from argparse import ArgumentParser
from logging import basicConfig, CRITICAL, ERROR, WARNING, INFO, DEBUG

parser = ArgumentParser()
parser.add_argument("-v", "--verbose", action="count")
parser.add_argument("-q", "--quiet", action="count")

arguments = parser.parse_args()

raw_log_level = 2 + (arguments.verbose or 0) - (arguments.quiet or 0)
if raw_log_level <= 0: 
    log_level = CRITICAL
elif raw_log_level == 1:
    log_level = ERROR
elif raw_log_level == 2:     # default
    log_level = WARNING
elif raw_log_level == 3: 
    log_level = INFO
else:         
    log_level = DEBUG

basicConfig(level=log_level)

Answer 3

There are levels of logging. Based on the severity of the logging level, it will print it.

Level   Numeric value
CRITICAL    50
ERROR   40
WARNING     30
INFO    20
DEBUG   10
NOTSET  0

Based on the logging level it will print the statements.

the level you have specified here is level=logging.DEBUG. So all but notset logging levels should print out. If you would want to print out only critical levels, please change the level=logging.CRITICAL

http://docs.python.org/release/2.5/lib/module-logging.html has more information