22

✏️ Editing: logging-cookbook.html

<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
 "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">

<html xmlns="http://www.w3.org/1999/xhtml">
 <head>
 <meta http-equiv="X-UA-Compatible" content="IE=Edge" />
 <meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
 <title>Logging Cookbook &#8212; Python 3.6.7 documentation</title>
 <link rel="stylesheet" href="../_static/pydoctheme.css" type="text/css" />
 <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
 <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
 <script type="text/javascript" src="../_static/jquery.js"></script>
 <script type="text/javascript" src="../_static/underscore.js"></script>
 <script type="text/javascript" src="../_static/doctools.js"></script>
 <script type="text/javascript" src="../_static/sidebar.js"></script>
 <link rel="search" type="application/opensearchdescription+xml"
 title="Search within Python 3.6.7 documentation"
 href="../_static/opensearch.xml"/>
 <link rel="author" title="About these documents" href="../about.html" />
 <link rel="index" title="Index" href="../genindex.html" />
 <link rel="search" title="Search" href="../search.html" />
 <link rel="copyright" title="Copyright" href="../copyright.html" />
 <link rel="next" title="Regular Expression HOWTO" href="regex.html" />
 <link rel="prev" title="Logging HOWTO" href="logging.html" />
 <link rel="shortcut icon" type="image/png" href="../_static/py.png" />
 <link rel="canonical" href="https://docs.python.org/3/howto/logging-cookbook.html" />
 
 <script type="text/javascript" src="../_static/copybutton.js"></script>

</head><body> 
 <div class="related" role="navigation" aria-label="related navigation">
 <h3>Navigation</h3>
 <ul>
 <li class="right" style="margin-right: 10px">
 <a href="../genindex.html" title="General Index"
 accesskey="I">index</a></li>
 <li class="right" >
 <a href="../py-modindex.html" title="Python Module Index"
 >modules</a> |</li>
 <li class="right" >
 <a href="regex.html" title="Regular Expression HOWTO"
 accesskey="N">next</a> |</li>
 <li class="right" >
 <a href="logging.html" title="Logging HOWTO"
 accesskey="P">previous</a> |</li>
 <li><img src="../_static/py.png" alt=""
 style="vertical-align: middle; margin-top: -1px"/></li>
 <li><a href="https://www.python.org/">Python</a> &#187;</li>
 <li>
 <a href="../index.html">3.6.7 Documentation</a> &#187;
 </li>

<li class="nav-item nav-item-1"><a href="index.html" accesskey="U">Python HOWTOs</a> &#187;</li>
 <li class="right">

<div class="inline-search" style="display: none" role="search">
 <form class="inline-search" action="../search.html" method="get">
 <input placeholder="Quick search" type="text" name="q" />
 <input type="submit" value="Go" />
 <input type="hidden" name="check_keywords" value="yes" />
 <input type="hidden" name="area" value="default" />
 </form>
 </div>
 <script type="text/javascript">$('.inline-search').show(0);</script>
 |
 </li>

</ul>
 </div>

<div class="document">
 <div class="documentwrapper">
 <div class="bodywrapper">
 <div class="body" role="main">
 
 <div class="section" id="logging-cookbook">
<h1>Logging Cookbook<a class="headerlink" href="#logging-cookbook" title="Permalink to this headline">¶</a></h1>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Author:</th><td class="field-body">Vinay Sajip &lt;vinay_sajip at red-dove dot com&gt;</td>
</tr>
</tbody>
</table>
This page contains a number of recipes related to logging, which have been found
useful in the past.
<div class="section" id="using-logging-in-multiple-modules">
<h2>Using logging in multiple modules<a class="headerlink" href="#using-logging-in-multiple-modules" title="Permalink to this headline">¶</a></h2>
Multiple calls to <code class="docutils literal notranslate">logging.getLogger('someLogger')</code> return a reference to the
same logger object. This is true not only within the same module, but also
across modules as long as it is in the same Python interpreter process. It is
true for references to the same object; additionally, application code can
define and configure a parent logger in one module and create (but not
configure) a child logger in a separate module, and all logger calls to the
child will pass up to the parent. Here is a main module:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>import logging
import auxiliary_module

# create logger with &#39;spam_application&#39;
logger = logging.getLogger(&#39;spam_application&#39;)
logger.setLevel(logging.DEBUG)
# create file handler which logs even debug messages
fh = logging.FileHandler(&#39;spam.log&#39;)
fh.setLevel(logging.DEBUG)
# create console handler with a higher log level
ch = logging.StreamHandler()
ch.setLevel(logging.ERROR)
# create formatter and add it to the handlers
formatter = logging.Formatter(&#39;%(asctime)s - %(name)s - %(levelname)s - %(message)s&#39;)
fh.setFormatter(formatter)
ch.setFormatter(formatter)
# add the handlers to the logger
logger.addHandler(fh)
logger.addHandler(ch)

logger.info(&#39;creating an instance of auxiliary_module.Auxiliary&#39;)
a = auxiliary_module.Auxiliary()
logger.info(&#39;created an instance of auxiliary_module.Auxiliary&#39;)
logger.info(&#39;calling auxiliary_module.Auxiliary.do_something&#39;)
a.do_something()
logger.info(&#39;finished auxiliary_module.Auxiliary.do_something&#39;)
logger.info(&#39;calling auxiliary_module.some_function()&#39;)
auxiliary_module.some_function()
logger.info(&#39;done with auxiliary_module.some_function()&#39;)
</pre></div>
</div>
Here is the auxiliary module:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>import logging

# create logger
module_logger = logging.getLogger(&#39;spam_application.auxiliary&#39;)

class Auxiliary:
 def __init__(self):
 self.logger = logging.getLogger(&#39;spam_application.auxiliary.Auxiliary&#39;)
 self.logger.info(&#39;creating an instance of Auxiliary&#39;)

def do_something(self):
 self.logger.info(&#39;doing something&#39;)
 a = 1 + 1
 self.logger.info(&#39;done doing something&#39;)

def some_function():
 module_logger.info(&#39;received a call to &quot;some_function&quot;&#39;)
</pre></div>
</div>
The output looks like this:
<div class="highlight-none notranslate"><div class="highlight"><pre>2005-03-23 23:47:11,663 - spam_application - INFO -
 creating an instance of auxiliary_module.Auxiliary
2005-03-23 23:47:11,665 - spam_application.auxiliary.Auxiliary - INFO -
 creating an instance of Auxiliary
2005-03-23 23:47:11,665 - spam_application - INFO -
 created an instance of auxiliary_module.Auxiliary
2005-03-23 23:47:11,668 - spam_application - INFO -
 calling auxiliary_module.Auxiliary.do_something
2005-03-23 23:47:11,668 - spam_application.auxiliary.Auxiliary - INFO -
 doing something
2005-03-23 23:47:11,669 - spam_application.auxiliary.Auxiliary - INFO -
 done doing something
2005-03-23 23:47:11,670 - spam_application - INFO -
 finished auxiliary_module.Auxiliary.do_something
2005-03-23 23:47:11,671 - spam_application - INFO -
 calling auxiliary_module.some_function()
2005-03-23 23:47:11,672 - spam_application.auxiliary - INFO -
 received a call to &#39;some_function&#39;
2005-03-23 23:47:11,673 - spam_application - INFO -
 done with auxiliary_module.some_function()
</pre></div>
</div>
</div>
<div class="section" id="logging-from-multiple-threads">
<h2>Logging from multiple threads<a class="headerlink" href="#logging-from-multiple-threads" title="Permalink to this headline">¶</a></h2>
Logging from multiple threads requires no special effort. The following example
shows logging from the main (initial) thread and another thread:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>import logging
import threading
import time

def worker(arg):
 while not arg[&#39;stop&#39;]:
 logging.debug(&#39;Hi from myfunc&#39;)
 time.sleep(0.5)

def main():
 logging.basicConfig(level=logging.DEBUG, format=&#39;%(relativeCreated)6d %(threadName)s %(message)s&#39;)
 info = {&#39;stop&#39;: False}
 thread = threading.Thread(target=worker, args=(info,))
 thread.start()
 while True:
 try:
 logging.debug(&#39;Hello from main&#39;)
 time.sleep(0.75)
 except KeyboardInterrupt:
 info[&#39;stop&#39;] = True
 break
 thread.join()

logger = logging.getLogger(&#39;simple_example&#39;)
logger.setLevel(logging.DEBUG)
# create file handler which logs even debug messages
fh = logging.FileHandler(&#39;spam.log&#39;)
fh.setLevel(logging.DEBUG)
# create console handler with a higher log level
ch = logging.StreamHandler()
ch.setLevel(logging.ERROR)
# create formatter and add it to the handlers
formatter = logging.Formatter(&#39;%(asctime)s - %(name)s - %(levelname)s - %(message)s&#39;)
ch.setFormatter(formatter)
fh.setFormatter(formatter)
# add the handlers to logger
logger.addHandler(ch)
logger.addHandler(fh)

# &#39;application&#39; code
logger.debug(&#39;debug message&#39;)
logger.info(&#39;info message&#39;)
logger.warn(&#39;warn message&#39;)
logger.error(&#39;error message&#39;)
logger.critical(&#39;critical message&#39;)
</pre></div>
</div>
Notice that the ‘application’ code does not care about multiple handlers. All
that changed was the addition and configuration of a new handler named fh.
The ability to create new handlers with higher- or lower-severity filters can be
very helpful when writing and testing an application. Instead of using many
<code class="docutils literal notranslate">print</code> statements for debugging, use <code class="docutils literal notranslate">logger.debug</code>: Unlike the print
statements, which you will have to delete or comment out later, the logger.debug
statements can remain intact in the source code and remain dormant until you
need them again. At that time, the only change that needs to happen is to
modify the severity level of the logger and/or handler to debug.
</div>
<div class="section" id="logging-to-multiple-destinations">
<h2>Logging to multiple destinations<a class="headerlink" href="#logging-to-multiple-destinations" title="Permalink to this headline">¶</a></h2>
Let’s say you want to log to console and file with different message formats and
in differing circumstances. Say you want to log messages with levels of DEBUG
and higher to file, and those messages at level INFO and higher to the console.
Let’s also assume that the file should contain timestamps, but the console
messages should not. Here’s how you can achieve this:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>import logging

# set up logging to file - see previous section for more details
logging.basicConfig(level=logging.DEBUG,
 format=&#39;%(asctime)s %(name)-12s %(levelname)-8s %(message)s&#39;,
 datefmt=&#39;%m-%d %H:%M&#39;,
 filename=&#39;/temp/myapp.log&#39;,
 filemode=&#39;w&#39;)
# define a Handler which writes INFO messages or higher to the sys.stderr
console = logging.StreamHandler()
console.setLevel(logging.INFO)
# set a format which is simpler for console use
formatter = logging.Formatter(&#39;%(name)-12s: %(levelname)-8s %(message)s&#39;)
# tell the handler to use this format
console.setFormatter(formatter)
# add the handler to the root logger
logging.getLogger(&#39;&#39;).addHandler(console)

# Now, we can log to the root logger, or any other logger. First the root...
logging.info(&#39;Jackdaws love my big sphinx of quartz.&#39;)

# Now, define a couple of other loggers which might represent areas in your
# application:

logger1 = logging.getLogger(&#39;myapp.area1&#39;)
logger2 = logging.getLogger(&#39;myapp.area2&#39;)

logger1.debug(&#39;Quick zephyrs blow, vexing daft Jim.&#39;)
logger1.info(&#39;How quickly daft jumping zebras vex.&#39;)
logger2.warning(&#39;Jail zesty vixen who grabbed pay from quack.&#39;)
logger2.error(&#39;The five boxing wizards jump quickly.&#39;)
</pre></div>
</div>
When you run this, on the console you will see
<div class="highlight-none notranslate"><div class="highlight"><pre>root : INFO Jackdaws love my big sphinx of quartz.
myapp.area1 : INFO How quickly daft jumping zebras vex.
myapp.area2 : WARNING Jail zesty vixen who grabbed pay from quack.
myapp.area2 : ERROR The five boxing wizards jump quickly.
</pre></div>
</div>
and in the file you will see something like
<div class="highlight-none notranslate"><div class="highlight"><pre>10-22 22:19 root INFO Jackdaws love my big sphinx of quartz.
10-22 22:19 myapp.area1 DEBUG Quick zephyrs blow, vexing daft Jim.
10-22 22:19 myapp.area1 INFO How quickly daft jumping zebras vex.
10-22 22:19 myapp.area2 WARNING Jail zesty vixen who grabbed pay from quack.
10-22 22:19 myapp.area2 ERROR The five boxing wizards jump quickly.
</pre></div>
</div>
As you can see, the DEBUG message only shows up in the file. The other messages
are sent to both destinations.
This example uses console and file handlers, but you can use any number and
combination of handlers you choose.
</div>
<div class="section" id="configuration-server-example">
<h2>Configuration server example<a class="headerlink" href="#configuration-server-example" title="Permalink to this headline">¶</a></h2>
Here is an example of a module using the logging configuration server:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>import logging
import logging.config
import time
import os

# read initial config file
logging.config.fileConfig(&#39;logging.conf&#39;)

# create and start listener on port 9999
t = logging.config.listen(9999)
t.start()

try:
 # loop through logging calls to see the difference
 # new configurations make, until Ctrl+C is pressed
 while True:
 logger.debug(&#39;debug message&#39;)
 logger.info(&#39;info message&#39;)
 logger.warn(&#39;warn message&#39;)
 logger.error(&#39;error message&#39;)
 logger.critical(&#39;critical message&#39;)
 time.sleep(5)
except KeyboardInterrupt:
 # cleanup
 logging.config.stopListening()
 t.join()
</pre></div>
</div>
And here is a script that takes a filename and sends that file to the server,
properly preceded with the binary-encoded length, as the new logging
configuration:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>#!/usr/bin/env python
import socket, sys, struct

with open(sys.argv[1], &#39;rb&#39;) as f:
 data_to_send = f.read()

HOST = &#39;localhost&#39;
PORT = 9999
s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
print(&#39;connecting...&#39;)
s.connect((HOST, PORT))
print(&#39;sending config...&#39;)
s.send(struct.pack(&#39;&gt;L&#39;, len(data_to_send)))
s.send(data_to_send)
s.close()
print(&#39;complete&#39;)
</pre></div>
</div>
</div>
<div class="section" id="dealing-with-handlers-that-block">
<h2>Dealing with handlers that block<a class="headerlink" href="#dealing-with-handlers-that-block" title="Permalink to this headline">¶</a></h2>
Sometimes you have to get your logging handlers to do their work without
blocking the thread you’re logging from. This is common in Web applications,
though of course it also occurs in other scenarios.
A common culprit which demonstrates sluggish behaviour is the
<a class="reference internal" href="../library/logging.handlers.html#logging.handlers.SMTPHandler" title="logging.handlers.SMTPHandler"><code class="xref py py-class docutils literal notranslate">SMTPHandler</code></a>: sending emails can take a long time, for a
number of reasons outside the developer’s control (for example, a poorly
performing mail or network infrastructure). But almost any network-based
handler can block: Even a <a class="reference internal" href="../library/logging.handlers.html#logging.handlers.SocketHandler" title="logging.handlers.SocketHandler"><code class="xref py py-class docutils literal notranslate">SocketHandler</code></a> operation may do a
DNS query under the hood which is too slow (and this query can be deep in the
socket library code, below the Python layer, and outside your control).
One solution is to use a two-part approach. For the first part, attach only a
<a class="reference internal" href="../library/logging.handlers.html#logging.handlers.QueueHandler" title="logging.handlers.QueueHandler"><code class="xref py py-class docutils literal notranslate">QueueHandler</code></a> to those loggers which are accessed from
performance-critical threads. They simply write to their queue, which can be
sized to a large enough capacity or initialized with no upper bound to their
size. The write to the queue will typically be accepted quickly, though you
will probably need to catch the <a class="reference internal" href="../library/queue.html#queue.Full" title="queue.Full"><code class="xref py py-exc docutils literal notranslate">queue.Full</code></a> exception as a precaution
in your code. If you are a library developer who has performance-critical
threads in their code, be sure to document this (together with a suggestion to
attach only <code class="docutils literal notranslate">QueueHandlers</code> to your loggers) for the benefit of other
developers who will use your code.
The second part of the solution is <a class="reference internal" href="../library/logging.handlers.html#logging.handlers.QueueListener" title="logging.handlers.QueueListener"><code class="xref py py-class docutils literal notranslate">QueueListener</code></a>, which has been
designed as the counterpart to <a class="reference internal" href="../library/logging.handlers.html#logging.handlers.QueueHandler" title="logging.handlers.QueueHandler"><code class="xref py py-class docutils literal notranslate">QueueHandler</code></a>. A
<a class="reference internal" href="../library/logging.handlers.html#logging.handlers.QueueListener" title="logging.handlers.QueueListener"><code class="xref py py-class docutils literal notranslate">QueueListener</code></a> is very simple: it’s passed a queue and some handlers,
and it fires up an internal thread which listens to its queue for LogRecords
sent from <code class="docutils literal notranslate">QueueHandlers</code> (or any other source of <code class="docutils literal notranslate">LogRecords</code>, for that
matter). The <code class="docutils literal notranslate">LogRecords</code> are removed from the queue and passed to the
handlers for processing.
The advantage of having a separate <a class="reference internal" href="../library/logging.handlers.html#logging.handlers.QueueListener" title="logging.handlers.QueueListener"><code class="xref py py-class docutils literal notranslate">QueueListener</code></a> class is that you
can use the same instance to service multiple <code class="docutils literal notranslate">QueueHandlers</code>. This is more
resource-friendly than, say, having threaded versions of the existing handler
classes, which would eat up one thread per handler for no particular benefit.
An example of using these two classes follows (imports omitted):
<div class="highlight-python3 notranslate"><div class="highlight"><pre>que = queue.Queue(-1) # no limit on size
queue_handler = QueueHandler(que)
handler = logging.StreamHandler()
listener = QueueListener(que, handler)
root = logging.getLogger()
root.addHandler(queue_handler)
formatter = logging.Formatter(&#39;%(threadName)s: %(message)s&#39;)
handler.setFormatter(formatter)
listener.start()
# The log output will display the thread which generated
# the event (the main thread) rather than the internal
# thread which monitors the internal queue. This is what
# you want to happen.
root.warning(&#39;Look out!&#39;)
listener.stop()
</pre></div>
</div>
which, when run, will produce:
<div class="highlight-none notranslate"><div class="highlight"><pre>MainThread: Look out!
</pre></div>
</div>
<div class="versionchanged">
Changed in version 3.5: Prior to Python 3.5, the <a class="reference internal" href="../library/logging.handlers.html#logging.handlers.QueueListener" title="logging.handlers.QueueListener"><code class="xref py py-class docutils literal notranslate">QueueListener</code></a> always passed every message
received from the queue to every handler it was initialized with. (This was
because it was assumed that level filtering was all done on the other side,
where the queue is filled.) From 3.5 onwards, this behaviour can be changed
by passing a keyword argument <code class="docutils literal notranslate">respect_handler_level=True</code> to the
listener’s constructor. When this is done, the listener compares the level
of each message with the handler’s level, and only passes a message to a
handler if it’s appropriate to do so.
</div>
</div>
<div class="section" id="sending-and-receiving-logging-events-across-a-network">
<h2>Sending and receiving logging events across a network<a class="headerlink" href="#sending-and-receiving-logging-events-across-a-network" title="Permalink to this headline">¶</a></h2>
Let’s say you want to send logging events across a network, and handle them at
the receiving end. A simple way of doing this is attaching a
<a class="reference internal" href="../library/logging.handlers.html#logging.handlers.SocketHandler" title="logging.handlers.SocketHandler"><code class="xref py py-class docutils literal notranslate">SocketHandler</code></a> instance to the root logger at the sending end:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>import logging, logging.handlers

rootLogger = logging.getLogger(&#39;&#39;)
rootLogger.setLevel(logging.DEBUG)
socketHandler = logging.handlers.SocketHandler(&#39;localhost&#39;,
 logging.handlers.DEFAULT_TCP_LOGGING_PORT)
# don&#39;t bother with a formatter, since a socket handler sends the event as
# an unformatted pickle
rootLogger.addHandler(socketHandler)

# Now, define a couple of other loggers which might represent areas in your
# application:

class LogRecordStreamHandler(socketserver.StreamRequestHandler):
 &quot;&quot;&quot;Handler for a streaming logging request.

This basically logs the record using whatever logging policy is
 configured locally.
 &quot;&quot;&quot;

def handle(self):
 &quot;&quot;&quot;
 Handle multiple requests - each expected to be a 4-byte length,
 followed by the LogRecord in pickle format. Logs the record
 according to whatever policy is configured locally.
 &quot;&quot;&quot;
 while True:
 chunk = self.connection.recv(4)
 if len(chunk) &lt; 4:
 break
 slen = struct.unpack(&#39;&gt;L&#39;, chunk)[0]
 chunk = self.connection.recv(slen)
 while len(chunk) &lt; slen:
 chunk = chunk + self.connection.recv(slen - len(chunk))
 obj = self.unPickle(chunk)
 record = logging.makeLogRecord(obj)
 self.handleLogRecord(record)

def unPickle(self, data):
 return pickle.loads(data)

def handleLogRecord(self, record):
 # if a name is specified, we use the named logger rather than the one
 # implied by the record.
 if self.server.logname is not None:
 name = self.server.logname
 else:
 name = record.name
 logger = logging.getLogger(name)
 # N.B. EVERY record gets logged. This is because Logger.handle
 # is normally called AFTER logger-level filtering. If you want
 # to do filtering, do it at the client end to save wasting
 # cycles and network bandwidth!
 logger.handle(record)

class LogRecordSocketReceiver(socketserver.ThreadingTCPServer):
 &quot;&quot;&quot;
 Simple TCP socket-based logging receiver suitable for testing.
 &quot;&quot;&quot;

allow_reuse_address = True

def __init__(self, host=&#39;localhost&#39;,
 port=logging.handlers.DEFAULT_TCP_LOGGING_PORT,
 handler=LogRecordStreamHandler):
 socketserver.ThreadingTCPServer.__init__(self, (host, port), handler)
 self.abort = 0
 self.timeout = 1
 self.logname = None

def serve_until_stopped(self):
 import select
 abort = 0
 while not abort:
 rd, wr, ex = select.select([self.socket.fileno()],
 [], [],
 self.timeout)
 if rd:
 self.handle_request()
 abort = self.abort

if __name__ == &#39;__main__&#39;:
 main()
</pre></div>
</div>
First run the server, and then the client. On the client side, nothing is
printed on the console; on the server side, you should see something like:
<div class="highlight-none notranslate"><div class="highlight"><pre>About to start TCP server...
 59 root INFO Jackdaws love my big sphinx of quartz.
 59 myapp.area1 DEBUG Quick zephyrs blow, vexing daft Jim.
 69 myapp.area1 INFO How quickly daft jumping zebras vex.
 69 myapp.area2 WARNING Jail zesty vixen who grabbed pay from quack.
 69 myapp.area2 ERROR The five boxing wizards jump quickly.
</pre></div>
</div>
Note that there are some security issues with pickle in some scenarios. If
these affect you, you can use an alternative serialization scheme by overriding
the <code class="xref py py-meth docutils literal notranslate">makePickle()</code> method and implementing your
alternative there, as well as adapting the above script to use your alternative
serialization.
</div>
<div class="section" id="adding-contextual-information-to-your-logging-output">
<h2>Adding contextual information to your logging output<a class="headerlink" href="#adding-contextual-information-to-your-logging-output" title="Permalink to this headline">¶</a></h2>
Sometimes you want logging output to contain contextual information in
addition to the parameters passed to the logging call. For example, in a
networked application, it may be desirable to log client-specific information
in the log (e.g. remote client’s username, or IP address). Although you could
use the extra parameter to achieve this, it’s not always convenient to pass
the information in this way. While it might be tempting to create
<code class="xref py py-class docutils literal notranslate">Logger</code> instances on a per-connection basis, this is not a good idea
because these instances are not garbage collected. While this is not a problem
in practice, when the number of <code class="xref py py-class docutils literal notranslate">Logger</code> instances is dependent on the
level of granularity you want to use in logging an application, it could
be hard to manage if the number of <code class="xref py py-class docutils literal notranslate">Logger</code> instances becomes
effectively unbounded.
<div class="section" id="using-loggeradapters-to-impart-contextual-information">
<h3>Using LoggerAdapters to impart contextual information<a class="headerlink" href="#using-loggeradapters-to-impart-contextual-information" title="Permalink to this headline">¶</a></h3>
An easy way in which you can pass contextual information to be output along
with logging event information is to use the <code class="xref py py-class docutils literal notranslate">LoggerAdapter</code> class.
This class is designed to look like a <code class="xref py py-class docutils literal notranslate">Logger</code>, so that you can call
<code class="xref py py-meth docutils literal notranslate">debug()</code>, <code class="xref py py-meth docutils literal notranslate">info()</code>, <code class="xref py py-meth docutils literal notranslate">warning()</code>, <code class="xref py py-meth docutils literal notranslate">error()</code>,
<code class="xref py py-meth docutils literal notranslate">exception()</code>, <code class="xref py py-meth docutils literal notranslate">critical()</code> and <code class="xref py py-meth docutils literal notranslate">log()</code>. These methods have the
same signatures as their counterparts in <code class="xref py py-class docutils literal notranslate">Logger</code>, so you can use the
two types of instances interchangeably.
When you create an instance of <code class="xref py py-class docutils literal notranslate">LoggerAdapter</code>, you pass it a
<code class="xref py py-class docutils literal notranslate">Logger</code> instance and a dict-like object which contains your contextual
information. When you call one of the logging methods on an instance of
<code class="xref py py-class docutils literal notranslate">LoggerAdapter</code>, it delegates the call to the underlying instance of
<code class="xref py py-class docutils literal notranslate">Logger</code> passed to its constructor, and arranges to pass the contextual
information in the delegated call. Here’s a snippet from the code of
<code class="xref py py-class docutils literal notranslate">LoggerAdapter</code>:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>def debug(self, msg, *args, **kwargs):
 &quot;&quot;&quot;
 Delegate a debug call to the underlying logger, after adding
 contextual information from this adapter instance.
 &quot;&quot;&quot;
 msg, kwargs = self.process(msg, kwargs)
 self.logger.debug(msg, *args, **kwargs)
</pre></div>
</div>
The <code class="xref py py-meth docutils literal notranslate">process()</code> method of <code class="xref py py-class docutils literal notranslate">LoggerAdapter</code> is where the
contextual information is added to the logging output. It’s passed the message
and keyword arguments of the logging call, and it passes back (potentially)
modified versions of these to use in the call to the underlying logger. The
default implementation of this method leaves the message alone, but inserts
an ‘extra’ key in the keyword argument whose value is the dict-like object
passed to the constructor. Of course, if you had passed an ‘extra’ keyword
argument in the call to the adapter, it will be silently overwritten.
The advantage of using ‘extra’ is that the values in the dict-like object are
merged into the <code class="xref py py-class docutils literal notranslate">LogRecord</code> instance’s __dict__, allowing you to use
customized strings with your <code class="xref py py-class docutils literal notranslate">Formatter</code> instances which know about
the keys of the dict-like object. If you need a different method, e.g. if you
want to prepend or append the contextual information to the message string,
you just need to subclass <code class="xref py py-class docutils literal notranslate">LoggerAdapter</code> and override
<code class="xref py py-meth docutils literal notranslate">process()</code> to do what you need. Here is a simple example:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>class CustomAdapter(logging.LoggerAdapter):
 &quot;&quot;&quot;
 This example adapter expects the passed in dict-like object to have a
 &#39;connid&#39; key, whose value in brackets is prepended to the log message.
 &quot;&quot;&quot;
 def process(self, msg, kwargs):
 return &#39;[%s] %s&#39; % (self.extra[&#39;connid&#39;], msg), kwargs
</pre></div>
</div>
which you can use like this:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>logger = logging.getLogger(__name__)
adapter = CustomAdapter(logger, {&#39;connid&#39;: some_conn_id})
</pre></div>
</div>
Then any events that you log to the adapter will have the value of
<code class="docutils literal notranslate">some_conn_id</code> prepended to the log messages.
<div class="section" id="using-objects-other-than-dicts-to-pass-contextual-information">
<h4>Using objects other than dicts to pass contextual information<a class="headerlink" href="#using-objects-other-than-dicts-to-pass-contextual-information" title="Permalink to this headline">¶</a></h4>
You don’t need to pass an actual dict to a <code class="xref py py-class docutils literal notranslate">LoggerAdapter</code> - you could
pass an instance of a class which implements <code class="docutils literal notranslate">__getitem__</code> and <code class="docutils literal notranslate">__iter__</code> so
that it looks like a dict to logging. This would be useful if you want to
generate values dynamically (whereas the values in a dict would be constant).
</div>
</div>
<div class="section" id="using-filters-to-impart-contextual-information">
<h3>Using Filters to impart contextual information<a class="headerlink" href="#using-filters-to-impart-contextual-information" title="Permalink to this headline">¶</a></h3>
You can also add contextual information to log output using a user-defined
<code class="xref py py-class docutils literal notranslate">Filter</code>. <code class="docutils literal notranslate">Filter</code> instances are allowed to modify the <code class="docutils literal notranslate">LogRecords</code>
passed to them, including adding additional attributes which can then be output
using a suitable format string, or if needed a custom <code class="xref py py-class docutils literal notranslate">Formatter</code>.
For example in a web application, the request being processed (or at least,
the interesting parts of it) can be stored in a threadlocal
(<a class="reference internal" href="../library/threading.html#threading.local" title="threading.local"><code class="xref py py-class docutils literal notranslate">threading.local</code></a>) variable, and then accessed from a <code class="docutils literal notranslate">Filter</code> to
add, say, information from the request - say, the remote IP address and remote
user’s username - to the <code class="docutils literal notranslate">LogRecord</code>, using the attribute names ‘ip’ and
‘user’ as in the <code class="docutils literal notranslate">LoggerAdapter</code> example above. In that case, the same format
string can be used to get similar output to that shown above. Here’s an example
script:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>import logging
from random import choice

class ContextFilter(logging.Filter):
 &quot;&quot;&quot;
 This is a filter which injects contextual information into the log.

Rather than use actual contextual information, we just use random
 data in this demo.
 &quot;&quot;&quot;

USERS = [&#39;jim&#39;, &#39;fred&#39;, &#39;sheila&#39;]
 IPS = [&#39;123.231.231.123&#39;, &#39;127.0.0.1&#39;, &#39;192.168.0.1&#39;]

def filter(self, record):

record.ip = choice(ContextFilter.IPS)
 record.user = choice(ContextFilter.USERS)
 return True

if __name__ == &#39;__main__&#39;:
 levels = (logging.DEBUG, logging.INFO, logging.WARNING, logging.ERROR, logging.CRITICAL)
 logging.basicConfig(level=logging.DEBUG,
 format=&#39;%(asctime)-15s %(name)-5s %(levelname)-8s IP: %(ip)-15s User: %(user)-8s %(message)s&#39;)
 a1 = logging.getLogger(&#39;a.b.c&#39;)
 a2 = logging.getLogger(&#39;d.e.f&#39;)

f = ContextFilter()
 a1.addFilter(f)
 a2.addFilter(f)
 a1.debug(&#39;A debug message&#39;)
 a1.info(&#39;An info message with %s&#39;, &#39;some parameters&#39;)
 for x in range(10):
 lvl = choice(levels)
 lvlname = logging.getLevelName(lvl)
 a2.log(lvl, &#39;A message at %s level with %d %s&#39;, lvlname, 2, &#39;parameters&#39;)
</pre></div>
</div>
which, when run, produces something like:
<div class="highlight-none notranslate"><div class="highlight"><pre>2010-09-06 22:38:15,292 a.b.c DEBUG IP: 123.231.231.123 User: fred A debug message
2010-09-06 22:38:15,300 a.b.c INFO IP: 192.168.0.1 User: sheila An info message with some parameters
2010-09-06 22:38:15,300 d.e.f CRITICAL IP: 127.0.0.1 User: sheila A message at CRITICAL level with 2 parameters
2010-09-06 22:38:15,300 d.e.f ERROR IP: 127.0.0.1 User: jim A message at ERROR level with 2 parameters
2010-09-06 22:38:15,300 d.e.f DEBUG IP: 127.0.0.1 User: sheila A message at DEBUG level with 2 parameters
2010-09-06 22:38:15,300 d.e.f ERROR IP: 123.231.231.123 User: fred A message at ERROR level with 2 parameters
2010-09-06 22:38:15,300 d.e.f CRITICAL IP: 192.168.0.1 User: jim A message at CRITICAL level with 2 parameters
2010-09-06 22:38:15,300 d.e.f CRITICAL IP: 127.0.0.1 User: sheila A message at CRITICAL level with 2 parameters
2010-09-06 22:38:15,300 d.e.f DEBUG IP: 192.168.0.1 User: jim A message at DEBUG level with 2 parameters
2010-09-06 22:38:15,301 d.e.f ERROR IP: 127.0.0.1 User: sheila A message at ERROR level with 2 parameters
2010-09-06 22:38:15,301 d.e.f DEBUG IP: 123.231.231.123 User: fred A message at DEBUG level with 2 parameters
2010-09-06 22:38:15,301 d.e.f INFO IP: 123.231.231.123 User: fred A message at INFO level with 2 parameters
</pre></div>
</div>
</div>
</div>
<div class="section" id="logging-to-a-single-file-from-multiple-processes">
<h2>Logging to a single file from multiple processes<a class="headerlink" href="#logging-to-a-single-file-from-multiple-processes" title="Permalink to this headline">¶</a></h2>
Although logging is thread-safe, and logging to a single file from multiple
threads in a single process is supported, logging to a single file from
multiple processes is not supported, because there is no standard way to
serialize access to a single file across multiple processes in Python. If you
need to log to a single file from multiple processes, one way of doing this is
to have all the processes log to a <code class="xref py py-class docutils literal notranslate">SocketHandler</code>, and have a
separate process which implements a socket server which reads from the socket
and logs to file. (If you prefer, you can dedicate one thread in one of the
existing processes to perform this function.)
<a class="reference internal" href="#network-logging">This section</a> documents this approach in more detail and
includes a working socket receiver which can be used as a starting point for you
to adapt in your own applications.
If you are using a recent version of Python which includes the
<a class="reference internal" href="../library/multiprocessing.html#module-multiprocessing" title="multiprocessing: Process-based parallelism."><code class="xref py py-mod docutils literal notranslate">multiprocessing</code></a> module, you could write your own handler which uses the
<a class="reference internal" href="../library/multiprocessing.html#multiprocessing.Lock" title="multiprocessing.Lock"><code class="xref py py-class docutils literal notranslate">Lock</code></a> class from this module to serialize access to the
file from your processes. The existing <code class="xref py py-class docutils literal notranslate">FileHandler</code> and subclasses do
not make use of <a class="reference internal" href="../library/multiprocessing.html#module-multiprocessing" title="multiprocessing: Process-based parallelism."><code class="xref py py-mod docutils literal notranslate">multiprocessing</code></a> at present, though they may do so in the
future. Note that at present, the <a class="reference internal" href="../library/multiprocessing.html#module-multiprocessing" title="multiprocessing: Process-based parallelism."><code class="xref py py-mod docutils literal notranslate">multiprocessing</code></a> module does not provide
working lock functionality on all platforms (see
<a class="reference external" href="https://bugs.python.org/issue3770">https://bugs.python.org/issue3770</a>).
Alternatively, you can use a <code class="docutils literal notranslate">Queue</code> and a <a class="reference internal" href="../library/logging.handlers.html#logging.handlers.QueueHandler" title="logging.handlers.QueueHandler"><code class="xref py py-class docutils literal notranslate">QueueHandler</code></a> to send
all logging events to one of the processes in your multi-process application.
The following example script demonstrates how you can do this; in the example
a separate listener process listens for events sent by other processes and logs
them according to its own logging configuration. Although the example only
demonstrates one way of doing it (for example, you may want to use a listener
thread rather than a separate listener process – the implementation would be
analogous) it does allow for completely different logging configurations for
the listener and the other processes in your application, and can be used as
the basis for code meeting your own specific requirements:
<div class="highlight-python3 notranslate"><div class="highlight"><pre># You&#39;ll need these imports in your own code
import logging
import logging.handlers
import multiprocessing

# Next two import lines for this demo only
from random import choice, random
import time

#
# Because you&#39;ll want to define the logging configurations for listener and workers, the
# listener and worker process functions take a configurer parameter which is a callable
# for configuring logging for that process. These functions are also passed the queue,
# which they use for communication.
#
# In practice, you can configure the listener however you want, but note that in this
# simple example, the listener does not apply level or filter logic to received records.
# In practice, you would probably want to do this logic in the worker processes, to avoid
# sending events which would be filtered out between processes.
#
# The size of the rotated files is made small so you can see the results easily.
def listener_configurer():
 root = logging.getLogger()
 h = logging.handlers.RotatingFileHandler(&#39;mptest.log&#39;, &#39;a&#39;, 300, 10)
 f = logging.Formatter(&#39;%(asctime)s %(processName)-10s %(name)s %(levelname)-8s %(message)s&#39;)
 h.setFormatter(f)
 root.addHandler(h)

# This is the listener process top-level loop: wait for logging events
# (LogRecords)on the queue and handle them, quit when you get a None for a
# LogRecord.
def listener_process(queue, configurer):
 configurer()
 while True:
 try:
 record = queue.get()
 if record is None: # We send this as a sentinel to tell the listener to quit.
 break
 logger = logging.getLogger(record.name)
 logger.handle(record) # No level or filter logic applied - just do it!
 except Exception:
 import sys, traceback
 print(&#39;Whoops! Problem:&#39;, file=sys.stderr)
 traceback.print_exc(file=sys.stderr)

# Arrays used for random selections in this demo

LEVELS = [logging.DEBUG, logging.INFO, logging.WARNING,
 logging.ERROR, logging.CRITICAL]

LOGGERS = [&#39;a.b.c&#39;, &#39;d.e.f&#39;]

MESSAGES = [
 &#39;Random message #1&#39;,
 &#39;Random message #2&#39;,
 &#39;Random message #3&#39;,
]

# The worker configuration is done at the start of the worker process run.
# Note that on Windows you can&#39;t rely on fork semantics, so each process
# will run the logging configuration code when it starts.
def worker_configurer(queue):
 h = logging.handlers.QueueHandler(queue) # Just the one handler needed
 root = logging.getLogger()
 root.addHandler(h)
 # send all messages, for demo; no other level or filter logic applied.
 root.setLevel(logging.DEBUG)

# This is the worker process top-level loop, which just logs ten events with
# random intervening delays before terminating.
# The print messages are just so you know it&#39;s doing something!
def worker_process(queue, configurer):
 configurer(queue)
 name = multiprocessing.current_process().name
 print(&#39;Worker started: %s&#39; % name)
 for i in range(10):
 time.sleep(random())
 logger = logging.getLogger(choice(LOGGERS))
 level = choice(LEVELS)
 message = choice(MESSAGES)
 logger.log(level, message)
 print(&#39;Worker finished: %s&#39; % name)

# Here&#39;s where the demo gets orchestrated. Create the queue, create and start
# the listener, create ten workers and start them, wait for them to finish,
# then send a None to the queue to tell the listener to finish.
def main():
 queue = multiprocessing.Queue(-1)
 listener = multiprocessing.Process(target=listener_process,
 args=(queue, listener_configurer))
 listener.start()
 workers = []
 for i in range(10):
 worker = multiprocessing.Process(target=worker_process,
 args=(queue, worker_configurer))
 workers.append(worker)
 worker.start()
 for w in workers:
 w.join()
 queue.put_nowait(None)
 listener.join()

def logger_thread(q):
 while True:
 record = q.get()
 if record is None:
 break
 logger = logging.getLogger(record.name)
 logger.handle(record)

def worker_process(q):
 qh = logging.handlers.QueueHandler(q)
 root = logging.getLogger()
 root.setLevel(logging.DEBUG)
 root.addHandler(qh)
 levels = [logging.DEBUG, logging.INFO, logging.WARNING, logging.ERROR,
 logging.CRITICAL]
 loggers = [&#39;foo&#39;, &#39;foo.bar&#39;, &#39;foo.bar.baz&#39;,
 &#39;spam&#39;, &#39;spam.ham&#39;, &#39;spam.ham.eggs&#39;]
 for i in range(100):
 lvl = random.choice(levels)
 logger = logging.getLogger(random.choice(loggers))
 logger.log(lvl, &#39;Message no. %d&#39;, i)

if __name__ == &#39;__main__&#39;:
 q = Queue()
 d = {
 &#39;version&#39;: 1,
 &#39;formatters&#39;: {
 &#39;detailed&#39;: {
 &#39;class&#39;: &#39;logging.Formatter&#39;,
 &#39;format&#39;: &#39;%(asctime)s %(name)-15s %(levelname)-8s %(processName)-10s %(message)s&#39;
 }
 },
 &#39;handlers&#39;: {
 &#39;console&#39;: {
 &#39;class&#39;: &#39;logging.StreamHandler&#39;,
 &#39;level&#39;: &#39;INFO&#39;,
 },
 &#39;file&#39;: {
 &#39;class&#39;: &#39;logging.FileHandler&#39;,
 &#39;filename&#39;: &#39;mplog.log&#39;,
 &#39;mode&#39;: &#39;w&#39;,
 &#39;formatter&#39;: &#39;detailed&#39;,
 },
 &#39;foofile&#39;: {
 &#39;class&#39;: &#39;logging.FileHandler&#39;,
 &#39;filename&#39;: &#39;mplog-foo.log&#39;,
 &#39;mode&#39;: &#39;w&#39;,
 &#39;formatter&#39;: &#39;detailed&#39;,
 },
 &#39;errors&#39;: {
 &#39;class&#39;: &#39;logging.FileHandler&#39;,
 &#39;filename&#39;: &#39;mplog-errors.log&#39;,
 &#39;mode&#39;: &#39;w&#39;,
 &#39;level&#39;: &#39;ERROR&#39;,
 &#39;formatter&#39;: &#39;detailed&#39;,
 },
 },
 &#39;loggers&#39;: {
 &#39;foo&#39;: {
 &#39;handlers&#39;: [&#39;foofile&#39;]
 }
 },
 &#39;root&#39;: {
 &#39;level&#39;: &#39;DEBUG&#39;,
 &#39;handlers&#39;: [&#39;console&#39;, &#39;file&#39;, &#39;errors&#39;]
 },
 }
 workers = []
 for i in range(5):
 wp = Process(target=worker_process, name=&#39;worker %d&#39; % (i + 1), args=(q,))
 workers.append(wp)
 wp.start()
 logging.config.dictConfig(d)
 lp = threading.Thread(target=logger_thread, args=(q,))
 lp.start()
 # At this point, the main process could do some useful work of its own
 # Once it&#39;s done that, it can wait for the workers to terminate...
 for wp in workers:
 wp.join()
 # And now tell the logging thread to finish up, too
 q.put(None)
 lp.join()
</pre></div>
</div>
This variant shows how you can e.g. apply configuration for particular loggers
- e.g. the <code class="docutils literal notranslate">foo</code> logger has a special handler which stores all events in the
<code class="docutils literal notranslate">foo</code> subsystem in a file <code class="docutils literal notranslate">mplog-foo.log</code>. This will be used by the logging
machinery in the main process (even though the logging events are generated in
the worker processes) to direct the messages to the appropriate destinations.
</div>
<div class="section" id="using-file-rotation">
<h2>Using file rotation<a class="headerlink" href="#using-file-rotation" title="Permalink to this headline">¶</a></h2>
Sometimes you want to let a log file grow to a certain size, then open a new
file and log to that. You may want to keep a certain number of these files, and
when that many files have been created, rotate the files so that the number of
files and the size of the files both remain bounded. For this usage pattern, the
logging package provides a <code class="xref py py-class docutils literal notranslate">RotatingFileHandler</code>:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>import glob
import logging
import logging.handlers

LOG_FILENAME = &#39;logging_rotatingfile_example.out&#39;

# Set up a specific logger with our desired output level
my_logger = logging.getLogger(&#39;MyLogger&#39;)
my_logger.setLevel(logging.DEBUG)

# Add the log message handler to the logger
handler = logging.handlers.RotatingFileHandler(
 LOG_FILENAME, maxBytes=20, backupCount=5)

my_logger.addHandler(handler)

# Log some messages
for i in range(20):
 my_logger.debug(&#39;i = %d&#39; % i)

# See what files are created
logfiles = glob.glob(&#39;%s*&#39; % LOG_FILENAME)

for filename in logfiles:
 print(filename)
</pre></div>
</div>
The result should be 6 separate files, each with part of the log history for the
application:
<div class="highlight-none notranslate"><div class="highlight"><pre>logging_rotatingfile_example.out
logging_rotatingfile_example.out.1
logging_rotatingfile_example.out.2
logging_rotatingfile_example.out.3
logging_rotatingfile_example.out.4
logging_rotatingfile_example.out.5
</pre></div>
</div>
The most current file is always <code class="file docutils literal notranslate">logging_rotatingfile_example.out</code>,
and each time it reaches the size limit it is renamed with the suffix
<code class="docutils literal notranslate">.1</code>. Each of the existing backup files is renamed to increment the suffix
(<code class="docutils literal notranslate">.1</code> becomes <code class="docutils literal notranslate">.2</code>, etc.) and the <code class="docutils literal notranslate">.6</code> file is erased.
Obviously this example sets the log length much too small as an extreme
example. You would want to set maxBytes to an appropriate value.
</div>
<div class="section" id="use-of-alternative-formatting-styles">
<h2>Use of alternative formatting styles<a class="headerlink" href="#use-of-alternative-formatting-styles" title="Permalink to this headline">¶</a></h2>
When logging was added to the Python standard library, the only way of
formatting messages with variable content was to use the %-formatting
method. Since then, Python has gained two new formatting approaches:
<a class="reference internal" href="../library/string.html#string.Template" title="string.Template"><code class="xref py py-class docutils literal notranslate">string.Template</code></a> (added in Python 2.4) and <a class="reference internal" href="../library/stdtypes.html#str.format" title="str.format"><code class="xref py py-meth docutils literal notranslate">str.format()</code></a>
(added in Python 2.6).
Logging (as of 3.2) provides improved support for these two additional
formatting styles. The <code class="xref py py-class docutils literal notranslate">Formatter</code> class been enhanced to take an
additional, optional keyword parameter named <code class="docutils literal notranslate">style</code>. This defaults to
<code class="docutils literal notranslate">'%'</code>, but other possible values are <code class="docutils literal notranslate">'{'</code> and <code class="docutils literal notranslate">'$'</code>, which correspond
to the other two formatting styles. Backwards compatibility is maintained by
default (as you would expect), but by explicitly specifying a style parameter,
you get the ability to specify format strings which work with
<a class="reference internal" href="../library/stdtypes.html#str.format" title="str.format"><code class="xref py py-meth docutils literal notranslate">str.format()</code></a> or <a class="reference internal" href="../library/string.html#string.Template" title="string.Template"><code class="xref py py-class docutils literal notranslate">string.Template</code></a>. Here’s an example console
session to show the possibilities:
<div class="highlight-pycon notranslate"><div class="highlight"><pre>&gt;&gt;&gt; import logging
&gt;&gt;&gt; root = logging.getLogger()
&gt;&gt;&gt; root.setLevel(logging.DEBUG)
&gt;&gt;&gt; handler = logging.StreamHandler()
&gt;&gt;&gt; bf = logging.Formatter(&#39;{asctime} {name} {levelname:8s} {message}&#39;,
... style=&#39;{&#39;)
&gt;&gt;&gt; handler.setFormatter(bf)
&gt;&gt;&gt; root.addHandler(handler)
&gt;&gt;&gt; logger = logging.getLogger(&#39;foo.bar&#39;)
&gt;&gt;&gt; logger.debug(&#39;This is a DEBUG message&#39;)
2010-10-28 15:11:55,341 foo.bar DEBUG This is a DEBUG message
&gt;&gt;&gt; logger.critical(&#39;This is a CRITICAL message&#39;)
2010-10-28 15:12:11,526 foo.bar CRITICAL This is a CRITICAL message
&gt;&gt;&gt; df = logging.Formatter(&#39;$asctime $name ${levelname} $message&#39;,
... style=&#39;$&#39;)
&gt;&gt;&gt; handler.setFormatter(df)
&gt;&gt;&gt; logger.debug(&#39;This is a DEBUG message&#39;)
2010-10-28 15:13:06,924 foo.bar DEBUG This is a DEBUG message
&gt;&gt;&gt; logger.critical(&#39;This is a CRITICAL message&#39;)
2010-10-28 15:13:11,494 foo.bar CRITICAL This is a CRITICAL message
&gt;&gt;&gt;
</pre></div>
</div>
Note that the formatting of logging messages for final output to logs is
completely independent of how an individual logging message is constructed.
That can still use %-formatting, as shown here:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; logger.error(&#39;This is an%s %s %s&#39;, &#39;other,&#39;, &#39;ERROR,&#39;, &#39;message&#39;)
2010-10-28 15:19:29,833 foo.bar ERROR This is another, ERROR, message
&gt;&gt;&gt;
</pre></div>
</div>
Logging calls (<code class="docutils literal notranslate">logger.debug()</code>, <code class="docutils literal notranslate">logger.info()</code> etc.) only take
positional parameters for the actual logging message itself, with keyword
parameters used only for determining options for how to handle the actual
logging call (e.g. the <code class="docutils literal notranslate">exc_info</code> keyword parameter to indicate that
traceback information should be logged, or the <code class="docutils literal notranslate">extra</code> keyword parameter
to indicate additional contextual information to be added to the log). So
you cannot directly make logging calls using <a class="reference internal" href="../library/stdtypes.html#str.format" title="str.format"><code class="xref py py-meth docutils literal notranslate">str.format()</code></a> or
<a class="reference internal" href="../library/string.html#string.Template" title="string.Template"><code class="xref py py-class docutils literal notranslate">string.Template</code></a> syntax, because internally the logging package
uses %-formatting to merge the format string and the variable arguments.
There would no changing this while preserving backward compatibility, since
all logging calls which are out there in existing code will be using %-format
strings.
There is, however, a way that you can use {}- and $- formatting to construct
your individual log messages. Recall that for a message you can use an
arbitrary object as a message format string, and that the logging package will
call <code class="docutils literal notranslate">str()</code> on that object to get the actual format string. Consider the
following two classes:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>class BraceMessage:
 def __init__(self, fmt, *args, **kwargs):
 self.fmt = fmt
 self.args = args
 self.kwargs = kwargs

def __str__(self):
 return self.fmt.format(*self.args, **self.kwargs)

class DollarMessage:
 def __init__(self, fmt, **kwargs):
 self.fmt = fmt
 self.kwargs = kwargs

def __str__(self):
 from string import Template
 return Template(self.fmt).substitute(**self.kwargs)
</pre></div>
</div>
Either of these can be used in place of a format string, to allow {}- or
$-formatting to be used to build the actual “message” part which appears in the
formatted log output in place of “%(message)s” or “{message}” or “$message”.
It’s a little unwieldy to use the class names whenever you want to log
something, but it’s quite palatable if you use an alias such as __ (double
underscore — not to be confused with _, the single underscore used as a
synonym/alias for <a class="reference internal" href="../library/gettext.html#gettext.gettext" title="gettext.gettext"><code class="xref py py-func docutils literal notranslate">gettext.gettext()</code></a> or its brethren).
The above classes are not included in Python, though they’re easy enough to
copy and paste into your own code. They can be used as follows (assuming that
they’re declared in a module called <code class="docutils literal notranslate">wherever</code>):
<div class="highlight-pycon notranslate"><div class="highlight"><pre>&gt;&gt;&gt; from wherever import BraceMessage as __
&gt;&gt;&gt; print(__(&#39;Message with {0} {name}&#39;, 2, name=&#39;placeholders&#39;))
Message with 2 placeholders
&gt;&gt;&gt; class Point: pass
...
&gt;&gt;&gt; p = Point()
&gt;&gt;&gt; p.x = 0.5
&gt;&gt;&gt; p.y = 0.5
&gt;&gt;&gt; print(__(&#39;Message with coordinates: ({point.x:.2f}, {point.y:.2f})&#39;,
... point=p))
Message with coordinates: (0.50, 0.50)
&gt;&gt;&gt; from wherever import DollarMessage as __
&gt;&gt;&gt; print(__(&#39;Message with $num $what&#39;, num=2, what=&#39;placeholders&#39;))
Message with 2 placeholders
&gt;&gt;&gt;
</pre></div>
</div>
While the above examples use <code class="docutils literal notranslate">print()</code> to show how the formatting works, you
would of course use <code class="docutils literal notranslate">logger.debug()</code> or similar to actually log using this
approach.
One thing to note is that you pay no significant performance penalty with this
approach: the actual formatting happens not when you make the logging call, but
when (and if) the logged message is actually about to be output to a log by a
handler. So the only slightly unusual thing which might trip you up is that the
parentheses go around the format string and the arguments, not just the format
string. That’s because the __ notation is just syntax sugar for a constructor
call to one of the XXXMessage classes.
If you prefer, you can use a <code class="xref py py-class docutils literal notranslate">LoggerAdapter</code> to achieve a similar effect
to the above, as in the following example:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>import logging

class Message(object):
 def __init__(self, fmt, args):
 self.fmt = fmt
 self.args = args

class StyleAdapter(logging.LoggerAdapter):
 def __init__(self, logger, extra=None):
 super(StyleAdapter, self).__init__(logger, extra or {})

def log(self, level, msg, *args, **kwargs):
 if self.isEnabledFor(level):
 msg, kwargs = self.process(msg, kwargs)
 self.logger._log(level, Message(msg, args), (), **kwargs)

logger = StyleAdapter(logging.getLogger(__name__))

def main():
 logger.debug(&#39;Hello, {}&#39;, &#39;world!&#39;)

if __name__ == &#39;__main__&#39;:
 logging.basicConfig(level=logging.DEBUG)
 main()
</pre></div>
</div>
The above script should log the message <code class="docutils literal notranslate">Hello, world!</code> when run with
Python 3.2 or later.
</div>
<div class="section" id="customizing-logrecord">
<h2>Customizing <code class="docutils literal notranslate">LogRecord</code><a class="headerlink" href="#customizing-logrecord" title="Permalink to this headline">¶</a></h2>
Every logging event is represented by a <a class="reference internal" href="../library/logging.html#logging.LogRecord" title="logging.LogRecord"><code class="xref py py-class docutils literal notranslate">LogRecord</code></a> instance.
When an event is logged and not filtered out by a logger’s level, a
<a class="reference internal" href="../library/logging.html#logging.LogRecord" title="logging.LogRecord"><code class="xref py py-class docutils literal notranslate">LogRecord</code></a> is created, populated with information about the event and
then passed to the handlers for that logger (and its ancestors, up to and
including the logger where further propagation up the hierarchy is disabled).
Before Python 3.2, there were only two places where this creation was done:
<ul class="simple">
<li><a class="reference internal" href="../library/logging.html#logging.Logger.makeRecord" title="logging.Logger.makeRecord"><code class="xref py py-meth docutils literal notranslate">Logger.makeRecord()</code></a>, which is called in the normal process of
logging an event. This invoked <a class="reference internal" href="../library/logging.html#logging.LogRecord" title="logging.LogRecord"><code class="xref py py-class docutils literal notranslate">LogRecord</code></a> directly to create an
instance.</li>
<li><a class="reference internal" href="../library/logging.html#logging.makeLogRecord" title="logging.makeLogRecord"><code class="xref py py-func docutils literal notranslate">makeLogRecord()</code></a>, which is called with a dictionary containing
attributes to be added to the LogRecord. This is typically invoked when a
suitable dictionary has been received over the network (e.g. in pickle form
via a <a class="reference internal" href="../library/logging.handlers.html#logging.handlers.SocketHandler" title="logging.handlers.SocketHandler"><code class="xref py py-class docutils literal notranslate">SocketHandler</code></a>, or in JSON form via an
<a class="reference internal" href="../library/logging.handlers.html#logging.handlers.HTTPHandler" title="logging.handlers.HTTPHandler"><code class="xref py py-class docutils literal notranslate">HTTPHandler</code></a>).</li>
</ul>
This has usually meant that if you need to do anything special with a
<a class="reference internal" href="../library/logging.html#logging.LogRecord" title="logging.LogRecord"><code class="xref py py-class docutils literal notranslate">LogRecord</code></a>, you’ve had to do one of the following.
<ul class="simple">
<li>Create your own <a class="reference internal" href="../library/logging.html#logging.Logger" title="logging.Logger"><code class="xref py py-class docutils literal notranslate">Logger</code></a> subclass, which overrides
<a class="reference internal" href="../library/logging.html#logging.Logger.makeRecord" title="logging.Logger.makeRecord"><code class="xref py py-meth docutils literal notranslate">Logger.makeRecord()</code></a>, and set it using <a class="reference internal" href="../library/logging.html#logging.setLoggerClass" title="logging.setLoggerClass"><code class="xref py py-func docutils literal notranslate">setLoggerClass()</code></a>
before any loggers that you care about are instantiated.</li>
<li>Add a <a class="reference internal" href="../library/logging.html#logging.Filter" title="logging.Filter"><code class="xref py py-class docutils literal notranslate">Filter</code></a> to a logger or handler, which does the
necessary special manipulation you need when its
<a class="reference internal" href="../library/logging.html#logging.Filter.filter" title="logging.Filter.filter"><code class="xref py py-meth docutils literal notranslate">filter()</code></a> method is called.</li>
</ul>
The first approach would be a little unwieldy in the scenario where (say)
several different libraries wanted to do different things. Each would attempt
to set its own <a class="reference internal" href="../library/logging.html#logging.Logger" title="logging.Logger"><code class="xref py py-class docutils literal notranslate">Logger</code></a> subclass, and the one which did this last would
win.
The second approach works reasonably well for many cases, but does not allow
you to e.g. use a specialized subclass of <a class="reference internal" href="../library/logging.html#logging.LogRecord" title="logging.LogRecord"><code class="xref py py-class docutils literal notranslate">LogRecord</code></a>. Library
developers can set a suitable filter on their loggers, but they would have to
remember to do this every time they introduced a new logger (which they would
do simply by adding new packages or modules and doing
<div class="highlight-python3 notranslate"><div class="highlight"><pre>logger = logging.getLogger(__name__)
</pre></div>
</div>
at module level). It’s probably one too many things to think about. Developers
could also add the filter to a <a class="reference internal" href="../library/logging.handlers.html#logging.NullHandler" title="logging.NullHandler"><code class="xref py py-class docutils literal notranslate">NullHandler</code></a> attached to their
top-level logger, but this would not be invoked if an application developer
attached a handler to a lower-level library logger — so output from that
handler would not reflect the intentions of the library developer.
In Python 3.2 and later, <a class="reference internal" href="../library/logging.html#logging.LogRecord" title="logging.LogRecord"><code class="xref py py-class docutils literal notranslate">LogRecord</code></a> creation is done through a
factory, which you can specify. The factory is just a callable you can set with
<a class="reference internal" href="../library/logging.html#logging.setLogRecordFactory" title="logging.setLogRecordFactory"><code class="xref py py-func docutils literal notranslate">setLogRecordFactory()</code></a>, and interrogate with
<a class="reference internal" href="../library/logging.html#logging.getLogRecordFactory" title="logging.getLogRecordFactory"><code class="xref py py-func docutils literal notranslate">getLogRecordFactory()</code></a>. The factory is invoked with the same
signature as the <a class="reference internal" href="../library/logging.html#logging.LogRecord" title="logging.LogRecord"><code class="xref py py-class docutils literal notranslate">LogRecord</code></a> constructor, as <a class="reference internal" href="../library/logging.html#logging.LogRecord" title="logging.LogRecord"><code class="xref py py-class docutils literal notranslate">LogRecord</code></a>
is the default setting for the factory.
This approach allows a custom factory to control all aspects of LogRecord
creation. For example, you could return a subclass, or just add some additional
attributes to the record once created, using a pattern similar to this:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>old_factory = logging.getLogRecordFactory()

def record_factory(*args, **kwargs):
 record = old_factory(*args, **kwargs)
 record.custom_attribute = 0xdecafbad
 return record

logging.setLogRecordFactory(record_factory)
</pre></div>
</div>
This pattern allows different libraries to chain factories together, and as
long as they don’t overwrite each other’s attributes or unintentionally
overwrite the attributes provided as standard, there should be no surprises.
However, it should be borne in mind that each link in the chain adds run-time
overhead to all logging operations, and the technique should only be used when
the use of a <a class="reference internal" href="../library/logging.html#logging.Filter" title="logging.Filter"><code class="xref py py-class docutils literal notranslate">Filter</code></a> does not provide the desired result.
</div>
<div class="section" id="subclassing-queuehandler-a-zeromq-example">
<h2>Subclassing QueueHandler - a ZeroMQ example<a class="headerlink" href="#subclassing-queuehandler-a-zeromq-example" title="Permalink to this headline">¶</a></h2>
You can use a <code class="xref py py-class docutils literal notranslate">QueueHandler</code> subclass to send messages to other kinds
of queues, for example a ZeroMQ ‘publish’ socket. In the example below,the
socket is created separately and passed to the handler (as its ‘queue’):
<div class="highlight-python3 notranslate"><div class="highlight"><pre>import zmq # using pyzmq, the Python binding for ZeroMQ
import json # for serializing records portably

ctx = zmq.Context()
sock = zmq.Socket(ctx, zmq.PUB) # or zmq.PUSH, or other suitable value
sock.bind(&#39;tcp://*:5556&#39;) # or wherever

class ZeroMQSocketHandler(QueueHandler):
 def enqueue(self, record):
 self.queue.send_json(record.__dict__)

handler = ZeroMQSocketHandler(sock)
</pre></div>
</div>
Of course there are other ways of organizing this, for example passing in the
data needed by the handler to create the socket:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>class ZeroMQSocketHandler(QueueHandler):
 def __init__(self, uri, socktype=zmq.PUB, ctx=None):
 self.ctx = ctx or zmq.Context()
 socket = zmq.Socket(self.ctx, socktype)
 socket.bind(uri)
 super().__init__(socket)

def enqueue(self, record):
 self.queue.send_json(record.__dict__)

def close(self):
 self.queue.close()
</pre></div>
</div>
</div>
<div class="section" id="subclassing-queuelistener-a-zeromq-example">
<h2>Subclassing QueueListener - a ZeroMQ example<a class="headerlink" href="#subclassing-queuelistener-a-zeromq-example" title="Permalink to this headline">¶</a></h2>
You can also subclass <code class="xref py py-class docutils literal notranslate">QueueListener</code> to get messages from other kinds
of queues, for example a ZeroMQ ‘subscribe’ socket. Here’s an example:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>class ZeroMQSocketListener(QueueListener):
 def __init__(self, uri, *handlers, **kwargs):
 self.ctx = kwargs.get(&#39;ctx&#39;) or zmq.Context()
 socket = zmq.Socket(self.ctx, zmq.SUB)
 socket.setsockopt_string(zmq.SUBSCRIBE, &#39;&#39;) # subscribe to everything
 socket.connect(uri)
 super().__init__(socket, *handlers, **kwargs)

def dequeue(self):
 msg = self.queue.recv_json()
 return logging.makeLogRecord(msg)
</pre></div>
</div>
<div class="admonition seealso">
See also
<dl class="docutils">
<dt>Module <a class="reference internal" href="../library/logging.html#module-logging" title="logging: Flexible event logging system for applications."><code class="xref py py-mod docutils literal notranslate">logging</code></a></dt>
<dd>API reference for the logging module.</dd>
<dt>Module <a class="reference internal" href="../library/logging.config.html#module-logging.config" title="logging.config: Configuration of the logging module."><code class="xref py py-mod docutils literal notranslate">logging.config</code></a></dt>
<dd>Configuration API for the logging module.</dd>
<dt>Module <a class="reference internal" href="../library/logging.handlers.html#module-logging.handlers" title="logging.handlers: Handlers for the logging module."><code class="xref py py-mod docutils literal notranslate">logging.handlers</code></a></dt>
<dd>Useful handlers included with the logging module.</dd>
</dl>
<a class="reference internal" href="logging.html#logging-basic-tutorial">A basic logging tutorial</a>
<a class="reference internal" href="logging.html#logging-advanced-tutorial">A more advanced logging tutorial</a>
</div>
</div>
<div class="section" id="an-example-dictionary-based-configuration">
<h2>An example dictionary-based configuration<a class="headerlink" href="#an-example-dictionary-based-configuration" title="Permalink to this headline">¶</a></h2>
Below is an example of a logging configuration dictionary - it’s taken from
the <a class="reference external" href="https://docs.djangoproject.com/en/1.9/topics/logging/#configuring-logging">documentation on the Django project</a>.
This dictionary is passed to <a class="reference internal" href="../library/logging.config.html#logging.config.dictConfig" title="logging.config.dictConfig"><code class="xref py py-func docutils literal notranslate">dictConfig()</code></a> to put the configuration into effect:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>LOGGING = {
 &#39;version&#39;: 1,
 &#39;disable_existing_loggers&#39;: True,
 &#39;formatters&#39;: {
 &#39;verbose&#39;: {
 &#39;format&#39;: &#39;%(levelname)s %(asctime)s %(module)s %(process)d %(thread)d %(message)s&#39;
 },
 &#39;simple&#39;: {
 &#39;format&#39;: &#39;%(levelname)s %(message)s&#39;
 },
 },
 &#39;filters&#39;: {
 &#39;special&#39;: {
 &#39;()&#39;: &#39;project.logging.SpecialFilter&#39;,
 &#39;foo&#39;: &#39;bar&#39;,
 }
 },
 &#39;handlers&#39;: {
 &#39;null&#39;: {
 &#39;level&#39;:&#39;DEBUG&#39;,
 &#39;class&#39;:&#39;django.utils.log.NullHandler&#39;,
 },
 &#39;console&#39;:{
 &#39;level&#39;:&#39;DEBUG&#39;,
 &#39;class&#39;:&#39;logging.StreamHandler&#39;,
 &#39;formatter&#39;: &#39;simple&#39;
 },
 &#39;mail_admins&#39;: {
 &#39;level&#39;: &#39;ERROR&#39;,
 &#39;class&#39;: &#39;django.utils.log.AdminEmailHandler&#39;,
 &#39;filters&#39;: [&#39;special&#39;]
 }
 },
 &#39;loggers&#39;: {
 &#39;django&#39;: {
 &#39;handlers&#39;:[&#39;null&#39;],
 &#39;propagate&#39;: True,
 &#39;level&#39;:&#39;INFO&#39;,
 },
 &#39;django.request&#39;: {
 &#39;handlers&#39;: [&#39;mail_admins&#39;],
 &#39;level&#39;: &#39;ERROR&#39;,
 &#39;propagate&#39;: False,
 },
 &#39;myproject.custom&#39;: {
 &#39;handlers&#39;: [&#39;console&#39;, &#39;mail_admins&#39;],
 &#39;level&#39;: &#39;INFO&#39;,
 &#39;filters&#39;: [&#39;special&#39;]
 }
 }
}
</pre></div>
</div>
For more information about this configuration, you can see the <a class="reference external" href="https://docs.djangoproject.com/en/1.9/topics/logging/#configuring-logging">relevant
section</a>
of the Django documentation.
</div>
<div class="section" id="using-a-rotator-and-namer-to-customize-log-rotation-processing">
<h2>Using a rotator and namer to customize log rotation processing<a class="headerlink" href="#using-a-rotator-and-namer-to-customize-log-rotation-processing" title="Permalink to this headline">¶</a></h2>
An example of how you can define a namer and rotator is given in the following
snippet, which shows zlib-based compression of the log file:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>def namer(name):
 return name + &quot;.gz&quot;

def rotator(source, dest):
 with open(source, &quot;rb&quot;) as sf:
 data = sf.read()
 compressed = zlib.compress(data, 9)
 with open(dest, &quot;wb&quot;) as df:
 df.write(compressed)
 os.remove(source)

rh = logging.handlers.RotatingFileHandler(...)
rh.rotator = rotator
rh.namer = namer
</pre></div>
</div>
These are not “true” .gz files, as they are bare compressed data, with no
“container” such as you’d find in an actual gzip file. This snippet is just
for illustration purposes.
</div>
<div class="section" id="a-more-elaborate-multiprocessing-example">
<h2>A more elaborate multiprocessing example<a class="headerlink" href="#a-more-elaborate-multiprocessing-example" title="Permalink to this headline">¶</a></h2>
The following working example shows how logging can be used with multiprocessing
using configuration files. The configurations are fairly simple, but serve to
illustrate how more complex ones could be implemented in a real multiprocessing
scenario.
In the example, the main process spawns a listener process and some worker
processes. Each of the main process, the listener and the workers have three
separate configurations (the workers all share the same configuration). We can
see logging in the main process, how the workers log to a QueueHandler and how
the listener implements a QueueListener and a more complex logging
configuration, and arranges to dispatch events received via the queue to the
handlers specified in the configuration. Note that these configurations are
purely illustrative, but you should be able to adapt this example to your own
scenario.
Here’s the script - the docstrings and the comments hopefully explain how it
works:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>import logging
import logging.config
import logging.handlers
from multiprocessing import Process, Queue, Event, current_process
import os
import random
import time

class MyHandler:
 &quot;&quot;&quot;
 A simple handler for logging events. It runs in the listener process and
 dispatches events to loggers based on the name in the received record,
 which then get dispatched, by the logging system, to the handlers
 configured for those loggers.
 &quot;&quot;&quot;
 def handle(self, record):
 logger = logging.getLogger(record.name)
 # The process name is transformed just to show that it&#39;s the listener
 # doing the logging to files and console
 record.processName = &#39;%s (for %s)&#39; % (current_process().name, record.processName)
 logger.handle(record)

def listener_process(q, stop_event, config):
 &quot;&quot;&quot;
 This could be done in the main process, but is just done in a separate
 process for illustrative purposes.

This initialises logging according to the specified configuration,
 starts the listener and waits for the main process to signal completion
 via the event. The listener is then stopped, and the process exits.
 &quot;&quot;&quot;
 logging.config.dictConfig(config)
 listener = logging.handlers.QueueListener(q, MyHandler())
 listener.start()
 if os.name == &#39;posix&#39;:
 # On POSIX, the setup logger will have been configured in the
 # parent process, but should have been disabled following the
 # dictConfig call.
 # On Windows, since fork isn&#39;t used, the setup logger won&#39;t
 # exist in the child, so it would be created and the message
 # would appear - hence the &quot;if posix&quot; clause.
 logger = logging.getLogger(&#39;setup&#39;)
 logger.critical(&#39;Should not appear, because of disabled logger ...&#39;)
 stop_event.wait()
 listener.stop()

def worker_process(config):
 &quot;&quot;&quot;
 A number of these are spawned for the purpose of illustration. In
 practice, they could be a heterogeneous bunch of processes rather than
 ones which are identical to each other.

This initialises logging according to the specified configuration,
 and logs a hundred messages with random levels to randomly selected
 loggers.

A small sleep is added to allow other processes a chance to run. This
 is not strictly needed, but it mixes the output from the different
 processes a bit more than if it&#39;s left out.
 &quot;&quot;&quot;
 logging.config.dictConfig(config)
 levels = [logging.DEBUG, logging.INFO, logging.WARNING, logging.ERROR,
 logging.CRITICAL]
 loggers = [&#39;foo&#39;, &#39;foo.bar&#39;, &#39;foo.bar.baz&#39;,
 &#39;spam&#39;, &#39;spam.ham&#39;, &#39;spam.ham.eggs&#39;]
 if os.name == &#39;posix&#39;:
 # On POSIX, the setup logger will have been configured in the
 # parent process, but should have been disabled following the
 # dictConfig call.
 # On Windows, since fork isn&#39;t used, the setup logger won&#39;t
 # exist in the child, so it would be created and the message
 # would appear - hence the &quot;if posix&quot; clause.
 logger = logging.getLogger(&#39;setup&#39;)
 logger.critical(&#39;Should not appear, because of disabled logger ...&#39;)
 for i in range(100):
 lvl = random.choice(levels)
 logger = logging.getLogger(random.choice(loggers))
 logger.log(lvl, &#39;Message no. %d&#39;, i)
 time.sleep(0.01)

def main():
 q = Queue()
 # The main process gets a simple configuration which prints to the console.
 config_initial = {
 &#39;version&#39;: 1,
 &#39;formatters&#39;: {
 &#39;detailed&#39;: {
 &#39;class&#39;: &#39;logging.Formatter&#39;,
 &#39;format&#39;: &#39;%(asctime)s %(name)-15s %(levelname)-8s %(processName)-10s %(message)s&#39;
 }
 },
 &#39;handlers&#39;: {
 &#39;console&#39;: {
 &#39;class&#39;: &#39;logging.StreamHandler&#39;,
 &#39;level&#39;: &#39;INFO&#39;,
 },
 },
 &#39;root&#39;: {
 &#39;level&#39;: &#39;DEBUG&#39;,
 &#39;handlers&#39;: [&#39;console&#39;]
 },
 }
 # The worker process configuration is just a QueueHandler attached to the
 # root logger, which allows all messages to be sent to the queue.
 # We disable existing loggers to disable the &quot;setup&quot; logger used in the
 # parent process. This is needed on POSIX because the logger will
 # be there in the child following a fork().
 config_worker = {
 &#39;version&#39;: 1,
 &#39;disable_existing_loggers&#39;: True,
 &#39;handlers&#39;: {
 &#39;queue&#39;: {
 &#39;class&#39;: &#39;logging.handlers.QueueHandler&#39;,
 &#39;queue&#39;: q,
 },
 },
 &#39;root&#39;: {
 &#39;level&#39;: &#39;DEBUG&#39;,
 &#39;handlers&#39;: [&#39;queue&#39;]
 },
 }
 # The listener process configuration shows that the full flexibility of
 # logging configuration is available to dispatch events to handlers however
 # you want.
 # We disable existing loggers to disable the &quot;setup&quot; logger used in the
 # parent process. This is needed on POSIX because the logger will
 # be there in the child following a fork().
 config_listener = {
 &#39;version&#39;: 1,
 &#39;disable_existing_loggers&#39;: True,
 &#39;formatters&#39;: {
 &#39;detailed&#39;: {
 &#39;class&#39;: &#39;logging.Formatter&#39;,
 &#39;format&#39;: &#39;%(asctime)s %(name)-15s %(levelname)-8s %(processName)-10s %(message)s&#39;
 },
 &#39;simple&#39;: {
 &#39;class&#39;: &#39;logging.Formatter&#39;,
 &#39;format&#39;: &#39;%(name)-15s %(levelname)-8s %(processName)-10s %(message)s&#39;
 }
 },
 &#39;handlers&#39;: {
 &#39;console&#39;: {
 &#39;class&#39;: &#39;logging.StreamHandler&#39;,
 &#39;level&#39;: &#39;INFO&#39;,
 &#39;formatter&#39;: &#39;simple&#39;,
 },
 &#39;file&#39;: {
 &#39;class&#39;: &#39;logging.FileHandler&#39;,
 &#39;filename&#39;: &#39;mplog.log&#39;,
 &#39;mode&#39;: &#39;w&#39;,
 &#39;formatter&#39;: &#39;detailed&#39;,
 },
 &#39;foofile&#39;: {
 &#39;class&#39;: &#39;logging.FileHandler&#39;,
 &#39;filename&#39;: &#39;mplog-foo.log&#39;,
 &#39;mode&#39;: &#39;w&#39;,
 &#39;formatter&#39;: &#39;detailed&#39;,
 },
 &#39;errors&#39;: {
 &#39;class&#39;: &#39;logging.FileHandler&#39;,
 &#39;filename&#39;: &#39;mplog-errors.log&#39;,
 &#39;mode&#39;: &#39;w&#39;,
 &#39;level&#39;: &#39;ERROR&#39;,
 &#39;formatter&#39;: &#39;detailed&#39;,
 },
 },
 &#39;loggers&#39;: {
 &#39;foo&#39;: {
 &#39;handlers&#39;: [&#39;foofile&#39;]
 }
 },
 &#39;root&#39;: {
 &#39;level&#39;: &#39;DEBUG&#39;,
 &#39;handlers&#39;: [&#39;console&#39;, &#39;file&#39;, &#39;errors&#39;]
 },
 }
 # Log some initial events, just to show that logging in the parent works
 # normally.
 logging.config.dictConfig(config_initial)
 logger = logging.getLogger(&#39;setup&#39;)
 logger.info(&#39;About to create workers ...&#39;)
 workers = []
 for i in range(5):
 wp = Process(target=worker_process, name=&#39;worker %d&#39; % (i + 1),
 args=(config_worker,))
 workers.append(wp)
 wp.start()
 logger.info(&#39;Started worker: %s&#39;, wp.name)
 logger.info(&#39;About to create listener ...&#39;)
 stop_event = Event()
 lp = Process(target=listener_process, name=&#39;listener&#39;,
 args=(q, stop_event, config_listener))
 lp.start()
 logger.info(&#39;Started listener&#39;)
 # We now hang around for the workers to finish their work.
 for wp in workers:
 wp.join()
 # Workers all done, listening can now stop.
 # Logging in the parent still works normally.
 logger.info(&#39;Telling listener to stop ...&#39;)
 stop_event.set()
 lp.join()
 logger.info(&#39;All done.&#39;)

class StructuredMessage(object):
 def __init__(self, message, **kwargs):
 self.message = message
 self.kwargs = kwargs

def __str__(self):
 return &#39;%s &gt;&gt;&gt; %s&#39; % (self.message, json.dumps(self.kwargs))

_ = StructuredMessage # optional, to improve readability

logging.basicConfig(level=logging.INFO, format=&#39;%(message)s&#39;)
logging.info(_(&#39;message 1&#39;, foo=&#39;bar&#39;, bar=&#39;baz&#39;, num=123, fnum=123.456))
</pre></div>
</div>
If the above script is run, it prints:
<div class="highlight-none notranslate"><div class="highlight"><pre>message 1 &gt;&gt;&gt; {&quot;fnum&quot;: 123.456, &quot;num&quot;: 123, &quot;bar&quot;: &quot;baz&quot;, &quot;foo&quot;: &quot;bar&quot;}
</pre></div>
</div>
Note that the order of items might be different according to the version of
Python used.
If you need more specialised processing, you can use a custom JSON encoder,
as in the following complete example:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>from __future__ import unicode_literals

import json
import logging

# This next bit is to ensure the script runs unchanged on 2.x and 3.x
try:
 unicode
except NameError:
 unicode = str

class Encoder(json.JSONEncoder):
 def default(self, o):
 if isinstance(o, set):
 return tuple(o)
 elif isinstance(o, unicode):
 return o.encode(&#39;unicode_escape&#39;).decode(&#39;ascii&#39;)
 return super(Encoder, self).default(o)

def __str__(self):
 s = Encoder().encode(self.kwargs)
 return &#39;%s &gt;&gt;&gt; %s&#39; % (self.message, s)

_ = StructuredMessage # optional, to improve readability

def owned_file_handler(filename, mode=&#39;a&#39;, encoding=None, owner=None):
 if owner:
 if not os.path.exists(filename):
 open(filename, &#39;a&#39;).close()
 shutil.chown(filename, *owner)
 return logging.FileHandler(filename, mode, encoding)

LOGGING = {
 &#39;version&#39;: 1,
 &#39;disable_existing_loggers&#39;: False,
 &#39;formatters&#39;: {
 &#39;default&#39;: {
 &#39;format&#39;: &#39;%(asctime)s %(levelname)s %(name)s %(message)s&#39;
 },
 },
 &#39;handlers&#39;: {
 &#39;file&#39;:{
 # The values below are popped from this dictionary and
 # used to create the handler, set the handler&#39;s level and
 # its formatter.
 &#39;()&#39;: owned_file_handler,
 &#39;level&#39;:&#39;DEBUG&#39;,
 &#39;formatter&#39;: &#39;default&#39;,
 # The values below are passed to the handler creator callable
 # as keyword arguments.
 &#39;owner&#39;: [&#39;pulse&#39;, &#39;pulse&#39;],
 &#39;filename&#39;: &#39;chowntest.log&#39;,
 &#39;mode&#39;: &#39;w&#39;,
 &#39;encoding&#39;: &#39;utf-8&#39;,
 },
 },
 &#39;root&#39;: {
 &#39;handlers&#39;: [&#39;file&#39;],
 &#39;level&#39;: &#39;DEBUG&#39;,
 },
}

logging.config.dictConfig(LOGGING)
logger = logging.getLogger(&#39;mylogger&#39;)
logger.debug(&#39;A debug message&#39;)
</pre></div>
</div>
To run this, you will probably need to run as <code class="docutils literal notranslate">root</code>:
<div class="highlight-shell-session notranslate"><div class="highlight"><pre>$ sudo python3.3 chowntest.py
$ cat chowntest.log
2013-11-05 09:34:51,128 DEBUG mylogger A debug message
$ ls -l chowntest.log
-rw-r--r-- 1 pulse pulse 55 2013-11-05 09:34 chowntest.log
</pre></div>
</div>
Note that this example uses Python 3.3 because that’s where <a class="reference internal" href="../library/shutil.html#shutil.chown" title="shutil.chown"><code class="xref py py-func docutils literal notranslate">shutil.chown()</code></a>
makes an appearance. This approach should work with any Python version that
supports <a class="reference internal" href="../library/logging.config.html#logging.config.dictConfig" title="logging.config.dictConfig"><code class="xref py py-func docutils literal notranslate">dictConfig()</code></a> - namely, Python 2.7, 3.2 or later. With pre-3.3
versions, you would need to implement the actual ownership change using e.g.
<a class="reference internal" href="../library/os.html#os.chown" title="os.chown"><code class="xref py py-func docutils literal notranslate">os.chown()</code></a>.
In practice, the handler-creating function may be in a utility module somewhere
in your project. Instead of the line in the configuration:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&#39;()&#39;: owned_file_handler,
</pre></div>
</div>
you could use e.g.:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&#39;()&#39;: &#39;ext://project.util.owned_file_handler&#39;,
</pre></div>
</div>
where <code class="docutils literal notranslate">project.util</code> can be replaced with the actual name of the package
where the function resides. In the above working script, using
<code class="docutils literal notranslate">'ext://__main__.owned_file_handler'</code> should work. Here, the actual callable
is resolved by <a class="reference internal" href="../library/logging.config.html#logging.config.dictConfig" title="logging.config.dictConfig"><code class="xref py py-func docutils literal notranslate">dictConfig()</code></a> from the <code class="docutils literal notranslate">ext://</code> specification.
This example hopefully also points the way to how you could implement other
types of file change - e.g. setting specific POSIX permission bits - in the
same way, using <a class="reference internal" href="../library/os.html#os.chmod" title="os.chmod"><code class="xref py py-func docutils literal notranslate">os.chmod()</code></a>.
Of course, the approach could also be extended to types of handler other than a
<a class="reference internal" href="../library/logging.handlers.html#logging.FileHandler" title="logging.FileHandler"><code class="xref py py-class docutils literal notranslate">FileHandler</code></a> - for example, one of the rotating file handlers,
or a different type of handler altogether.
</div>
<div class="section" id="using-particular-formatting-styles-throughout-your-application">
<h2>Using particular formatting styles throughout your application<a class="headerlink" href="#using-particular-formatting-styles-throughout-your-application" title="Permalink to this headline">¶</a></h2>
In Python 3.2, the <a class="reference internal" href="../library/logging.html#logging.Formatter" title="logging.Formatter"><code class="xref py py-class docutils literal notranslate">Formatter</code></a> gained a <code class="docutils literal notranslate">style</code> keyword
parameter which, while defaulting to <code class="docutils literal notranslate">%</code> for backward compatibility, allowed
the specification of <code class="docutils literal notranslate">{</code> or <code class="docutils literal notranslate">$</code> to support the formatting approaches
supported by <a class="reference internal" href="../library/stdtypes.html#str.format" title="str.format"><code class="xref py py-meth docutils literal notranslate">str.format()</code></a> and <a class="reference internal" href="../library/string.html#string.Template" title="string.Template"><code class="xref py py-class docutils literal notranslate">string.Template</code></a>. Note that this
governs the formatting of logging messages for final output to logs, and is
completely orthogonal to how an individual logging message is constructed.
Logging calls (<a class="reference internal" href="../library/logging.html#logging.Logger.debug" title="logging.Logger.debug"><code class="xref py py-meth docutils literal notranslate">debug()</code></a>, <a class="reference internal" href="../library/logging.html#logging.Logger.info" title="logging.Logger.info"><code class="xref py py-meth docutils literal notranslate">info()</code></a> etc.) only take
positional parameters for the actual logging message itself, with keyword
parameters used only for determining options for how to handle the logging call
(e.g. the <code class="docutils literal notranslate">exc_info</code> keyword parameter to indicate that traceback information
should be logged, or the <code class="docutils literal notranslate">extra</code> keyword parameter to indicate additional
contextual information to be added to the log). So you cannot directly make
logging calls using <a class="reference internal" href="../library/stdtypes.html#str.format" title="str.format"><code class="xref py py-meth docutils literal notranslate">str.format()</code></a> or <a class="reference internal" href="../library/string.html#string.Template" title="string.Template"><code class="xref py py-class docutils literal notranslate">string.Template</code></a> syntax,
because internally the logging package uses %-formatting to merge the format
string and the variable arguments. There would no changing this while preserving
backward compatibility, since all logging calls which are out there in existing
code will be using %-format strings.
There have been suggestions to associate format styles with specific loggers,
but that approach also runs into backward compatibility problems because any
existing code could be using a given logger name and using %-formatting.
For logging to work interoperably between any third-party libraries and your
code, decisions about formatting need to be made at the level of the
individual logging call. This opens up a couple of ways in which alternative
formatting styles can be accommodated.
<div class="section" id="using-logrecord-factories">
<h3>Using LogRecord factories<a class="headerlink" href="#using-logrecord-factories" title="Permalink to this headline">¶</a></h3>
In Python 3.2, along with the <a class="reference internal" href="../library/logging.html#logging.Formatter" title="logging.Formatter"><code class="xref py py-class docutils literal notranslate">Formatter</code></a> changes mentioned
above, the logging package gained the ability to allow users to set their own
<a class="reference internal" href="../library/logging.html#logging.LogRecord" title="logging.LogRecord"><code class="xref py py-class docutils literal notranslate">LogRecord</code></a> subclasses, using the <a class="reference internal" href="../library/logging.html#logging.setLogRecordFactory" title="logging.setLogRecordFactory"><code class="xref py py-func docutils literal notranslate">setLogRecordFactory()</code></a> function.
You can use this to set your own subclass of <a class="reference internal" href="../library/logging.html#logging.LogRecord" title="logging.LogRecord"><code class="xref py py-class docutils literal notranslate">LogRecord</code></a>, which does the
Right Thing by overriding the <a class="reference internal" href="../library/logging.html#logging.LogRecord.getMessage" title="logging.LogRecord.getMessage"><code class="xref py py-meth docutils literal notranslate">getMessage()</code></a> method. The base
class implementation of this method is where the <code class="docutils literal notranslate">msg % args</code> formatting
happens, and where you can substitute your alternate formatting; however, you
should be careful to support all formatting styles and allow %-formatting as
the default, to ensure interoperability with other code. Care should also be
taken to call <code class="docutils literal notranslate">str(self.msg)</code>, just as the base implementation does.
Refer to the reference documentation on <a class="reference internal" href="../library/logging.html#logging.setLogRecordFactory" title="logging.setLogRecordFactory"><code class="xref py py-func docutils literal notranslate">setLogRecordFactory()</code></a> and
<a class="reference internal" href="../library/logging.html#logging.LogRecord" title="logging.LogRecord"><code class="xref py py-class docutils literal notranslate">LogRecord</code></a> for more information.
</div>
<div class="section" id="using-custom-message-objects">
<h3>Using custom message objects<a class="headerlink" href="#using-custom-message-objects" title="Permalink to this headline">¶</a></h3>
There is another, perhaps simpler way that you can use {}- and $- formatting to
construct your individual log messages. You may recall (from
<a class="reference internal" href="logging.html#arbitrary-object-messages">Using arbitrary objects as messages</a>) that when logging you can use an arbitrary
object as a message format string, and that the logging package will call
<a class="reference internal" href="../library/stdtypes.html#str" title="str"><code class="xref py py-func docutils literal notranslate">str()</code></a> on that object to get the actual format string. Consider the
following two classes:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>class BraceMessage(object):
 def __init__(self, fmt, *args, **kwargs):
 self.fmt = fmt
 self.args = args
 self.kwargs = kwargs

class DollarMessage(object):
 def __init__(self, fmt, **kwargs):
 self.fmt = fmt
 self.kwargs = kwargs

def __str__(self):
 from string import Template
 return Template(self.fmt).substitute(**self.kwargs)
</pre></div>
</div>
Either of these can be used in place of a format string, to allow {}- or
$-formatting to be used to build the actual “message” part which appears in the
formatted log output in place of “%(message)s” or “{message}” or “$message”.
If you find it a little unwieldy to use the class names whenever you want to log
something, you can make it more palatable if you use an alias such as <code class="docutils literal notranslate">M</code> or
<code class="docutils literal notranslate">_</code> for the message (or perhaps <code class="docutils literal notranslate">__</code>, if you are using <code class="docutils literal notranslate">_</code> for
localization).
Examples of this approach are given below. Firstly, formatting with
<a class="reference internal" href="../library/stdtypes.html#str.format" title="str.format"><code class="xref py py-meth docutils literal notranslate">str.format()</code></a>:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; __ = BraceMessage
&gt;&gt;&gt; print(__(&#39;Message with {0} {1}&#39;, 2, &#39;placeholders&#39;))
Message with 2 placeholders
&gt;&gt;&gt; class Point: pass
...
&gt;&gt;&gt; p = Point()
&gt;&gt;&gt; p.x = 0.5
&gt;&gt;&gt; p.y = 0.5
&gt;&gt;&gt; print(__(&#39;Message with coordinates: ({point.x:.2f}, {point.y:.2f})&#39;, point=p))
Message with coordinates: (0.50, 0.50)
</pre></div>
</div>
Secondly, formatting with <a class="reference internal" href="../library/string.html#string.Template" title="string.Template"><code class="xref py py-class docutils literal notranslate">string.Template</code></a>:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; __ = DollarMessage
&gt;&gt;&gt; print(__(&#39;Message with $num $what&#39;, num=2, what=&#39;placeholders&#39;))
Message with 2 placeholders
&gt;&gt;&gt;
</pre></div>
</div>
One thing to note is that you pay no significant performance penalty with this
approach: the actual formatting happens not when you make the logging call, but
when (and if) the logged message is actually about to be output to a log by a
handler. So the only slightly unusual thing which might trip you up is that the
parentheses go around the format string and the arguments, not just the format
string. That’s because the __ notation is just syntax sugar for a constructor
call to one of the <code class="docutils literal notranslate">XXXMessage</code> classes shown above.
</div>
</div>
<div class="section" id="configuring-filters-with-dictconfig">
<h2>Configuring filters with <a class="reference internal" href="../library/logging.config.html#logging.config.dictConfig" title="logging.config.dictConfig"><code class="xref py py-func docutils literal notranslate">dictConfig()</code></a><a class="headerlink" href="#configuring-filters-with-dictconfig" title="Permalink to this headline">¶</a></h2>
You can configure filters using <a class="reference internal" href="../library/logging.config.html#logging.config.dictConfig" title="logging.config.dictConfig"><code class="xref py py-func docutils literal notranslate">dictConfig()</code></a>, though it
might not be obvious at first glance how to do it (hence this recipe). Since
<a class="reference internal" href="../library/logging.html#logging.Filter" title="logging.Filter"><code class="xref py py-class docutils literal notranslate">Filter</code></a> is the only filter class included in the standard
library, and it is unlikely to cater to many requirements (it’s only there as a
base class), you will typically need to define your own <a class="reference internal" href="../library/logging.html#logging.Filter" title="logging.Filter"><code class="xref py py-class docutils literal notranslate">Filter</code></a>
subclass with an overridden <a class="reference internal" href="../library/logging.html#logging.Filter.filter" title="logging.Filter.filter"><code class="xref py py-meth docutils literal notranslate">filter()</code></a> method. To do this,
specify the <code class="docutils literal notranslate">()</code> key in the configuration dictionary for the filter,
specifying a callable which will be used to create the filter (a class is the
most obvious, but you can provide any callable which returns a
<a class="reference internal" href="../library/logging.html#logging.Filter" title="logging.Filter"><code class="xref py py-class docutils literal notranslate">Filter</code></a> instance). Here is a complete example:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>import logging
import logging.config
import sys

class MyFilter(logging.Filter):
 def __init__(self, param=None):
 self.param = param

def filter(self, record):
 if self.param is None:
 allow = True
 else:
 allow = self.param not in record.msg
 if allow:
 record.msg = &#39;changed: &#39; + record.msg
 return allow

if __name__ == &#39;__main__&#39;:
 logging.config.dictConfig(LOGGING)
 logging.debug(&#39;hello&#39;)
 logging.debug(&#39;hello - noshow&#39;)
</pre></div>
</div>
This example shows how you can pass configuration data to the callable which
constructs the instance, in the form of keyword parameters. When run, the above
script will print:
<div class="highlight-none notranslate"><div class="highlight"><pre>changed: hello
</pre></div>
</div>
which shows that the filter is working as configured.
A couple of extra points to note:
<ul class="simple">
<li>If you can’t refer to the callable directly in the configuration (e.g. if it
lives in a different module, and you can’t import it directly where the
configuration dictionary is), you can use the form <code class="docutils literal notranslate">ext://...</code> as described
in <a class="reference internal" href="../library/logging.config.html#logging-config-dict-externalobj">Access to external objects</a>. For example, you could have used
the text <code class="docutils literal notranslate">'ext://__main__.MyFilter'</code> instead of <code class="docutils literal notranslate">MyFilter</code> in the above
example.</li>
<li>As well as for filters, this technique can also be used to configure custom
handlers and formatters. See <a class="reference internal" href="../library/logging.config.html#logging-config-dict-userdef">User-defined objects</a> for more
information on how logging supports using user-defined objects in its
configuration, and see the other cookbook recipe <a class="reference internal" href="#custom-handlers">Customizing handlers with dictConfig()</a> above.</li>
</ul>
</div>
<div class="section" id="customized-exception-formatting">
<h2>Customized exception formatting<a class="headerlink" href="#customized-exception-formatting" title="Permalink to this headline">¶</a></h2>
There might be times when you want to do customized exception formatting - for
argument’s sake, let’s say you want exactly one line per logged event, even
when exception information is present. You can do this with a custom formatter
class, as shown in the following example:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>import logging

class OneLineExceptionFormatter(logging.Formatter):
 def formatException(self, exc_info):
 &quot;&quot;&quot;
 Format an exception so that it prints on a single line.
 &quot;&quot;&quot;
 result = super(OneLineExceptionFormatter, self).formatException(exc_info)
 return repr(result) # or format into one line however you want to

def format(self, record):
 s = super(OneLineExceptionFormatter, self).format(record)
 if record.exc_text:
 s = s.replace(&#39;\n&#39;, &#39;&#39;) + &#39;|&#39;
 return s

def configure_logging():
 fh = logging.FileHandler(&#39;output.txt&#39;, &#39;w&#39;)
 f = OneLineExceptionFormatter(&#39;%(asctime)s|%(levelname)s|%(message)s|&#39;,
 &#39;%d/%m/%Y %H:%M:%S&#39;)
 fh.setFormatter(f)
 root = logging.getLogger()
 root.setLevel(logging.DEBUG)
 root.addHandler(fh)

def main():
 configure_logging()
 logging.info(&#39;Sample message&#39;)
 try:
 x = 1 / 0
 except ZeroDivisionError as e:
 logging.exception(&#39;ZeroDivisionError: %s&#39;, e)

class TTSHandler(logging.Handler):
 def emit(self, record):
 msg = self.format(record)
 # Speak slowly in a female English voice
 cmd = [&#39;espeak&#39;, &#39;-s150&#39;, &#39;-ven+f3&#39;, msg]
 p = subprocess.Popen(cmd, stdout=subprocess.PIPE,
 stderr=subprocess.STDOUT)
 # wait for the program to finish
 p.communicate()

def configure_logging():
 h = TTSHandler()
 root = logging.getLogger()
 root.addHandler(h)
 # the default formatter just returns the message
 root.setLevel(logging.DEBUG)

def main():
 logging.info(&#39;Hello&#39;)
 logging.debug(&#39;Goodbye&#39;)

if __name__ == &#39;__main__&#39;:
 configure_logging()
 sys.exit(main())
</pre></div>
</div>
When run, this script should say “Hello” and then “Goodbye” in a female voice.
The above approach can, of course, be adapted to other TTS systems and even
other systems altogether which can process messages via external programs run
from a command line.
</div>
<div class="section" id="buffering-logging-messages-and-outputting-them-conditionally">
<h2>Buffering logging messages and outputting them conditionally<a class="headerlink" href="#buffering-logging-messages-and-outputting-them-conditionally" title="Permalink to this headline">¶</a></h2>
There might be situations where you want to log messages in a temporary area
and only output them if a certain condition occurs. For example, you may want to
start logging debug events in a function, and if the function completes without
errors, you don’t want to clutter the log with the collected debug information,
but if there is an error, you want all the debug information to be output as well
as the error.
Here is an example which shows how you could do this using a decorator for your
functions where you want logging to behave this way. It makes use of the
<a class="reference internal" href="../library/logging.handlers.html#logging.handlers.MemoryHandler" title="logging.handlers.MemoryHandler"><code class="xref py py-class docutils literal notranslate">logging.handlers.MemoryHandler</code></a>, which allows buffering of logged events
until some condition occurs, at which point the buffered events are <code class="docutils literal notranslate">flushed</code>
- passed to another handler (the <code class="docutils literal notranslate">target</code> handler) for processing. By default,
the <code class="docutils literal notranslate">MemoryHandler</code> flushed when its buffer gets filled up or an event whose
level is greater than or equal to a specified threshold is seen. You can use this
recipe with a more specialised subclass of <code class="docutils literal notranslate">MemoryHandler</code> if you want custom
flushing behavior.
The example script has a simple function, <code class="docutils literal notranslate">foo</code>, which just cycles through
all the logging levels, writing to <code class="docutils literal notranslate">sys.stderr</code> to say what level it’s about
to log at, and then actually logging a message at that level. You can pass a
parameter to <code class="docutils literal notranslate">foo</code> which, if true, will log at ERROR and CRITICAL levels -
otherwise, it only logs at DEBUG, INFO and WARNING levels.
The script just arranges to decorate <code class="docutils literal notranslate">foo</code> with a decorator which will do the
conditional logging that’s required. The decorator takes a logger as a parameter
and attaches a memory handler for the duration of the call to the decorated
function. The decorator can be additionally parameterised using a target handler,
a level at which flushing should occur, and a capacity for the buffer. These
default to a <a class="reference internal" href="../library/logging.handlers.html#logging.StreamHandler" title="logging.StreamHandler"><code class="xref py py-class docutils literal notranslate">StreamHandler</code></a> which writes to <code class="docutils literal notranslate">sys.stderr</code>,
<code class="docutils literal notranslate">logging.ERROR</code> and <code class="docutils literal notranslate">100</code> respectively.
Here’s the script:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>import logging
from logging.handlers import MemoryHandler
import sys

def log_if_errors(logger, target_handler=None, flush_level=None, capacity=None):
 if target_handler is None:
 target_handler = logging.StreamHandler()
 if flush_level is None:
 flush_level = logging.ERROR
 if capacity is None:
 capacity = 100
 handler = MemoryHandler(capacity, flushLevel=flush_level, target=target_handler)

def decorator(fn):
 def wrapper(*args, **kwargs):
 logger.addHandler(handler)
 try:
 return fn(*args, **kwargs)
 except Exception:
 logger.exception(&#39;call failed&#39;)
 raise
 finally:
 super(MemoryHandler, handler).flush()
 logger.removeHandler(handler)
 return wrapper

return decorator

def write_line(s):
 sys.stderr.write(&#39;%s\n&#39; % s)

def foo(fail=False):
 write_line(&#39;about to log at DEBUG ...&#39;)
 logger.debug(&#39;Actually logged at DEBUG&#39;)
 write_line(&#39;about to log at INFO ...&#39;)
 logger.info(&#39;Actually logged at INFO&#39;)
 write_line(&#39;about to log at WARNING ...&#39;)
 logger.warning(&#39;Actually logged at WARNING&#39;)
 if fail:
 write_line(&#39;about to log at ERROR ...&#39;)
 logger.error(&#39;Actually logged at ERROR&#39;)
 write_line(&#39;about to log at CRITICAL ...&#39;)
 logger.critical(&#39;Actually logged at CRITICAL&#39;)
 return fail

decorated_foo = log_if_errors(logger)(foo)

if __name__ == &#39;__main__&#39;:
 logger.setLevel(logging.DEBUG)
 write_line(&#39;Calling undecorated foo with False&#39;)
 assert not foo(False)
 write_line(&#39;Calling undecorated foo with True&#39;)
 assert foo(True)
 write_line(&#39;Calling decorated foo with False&#39;)
 assert not decorated_foo(False)
 write_line(&#39;Calling decorated foo with True&#39;)
 assert decorated_foo(True)
</pre></div>
</div>
When this script is run, the following output should be observed:
<div class="highlight-none notranslate"><div class="highlight"><pre>Calling undecorated foo with False
about to log at DEBUG ...
about to log at INFO ...
about to log at WARNING ...
Calling undecorated foo with True
about to log at DEBUG ...
about to log at INFO ...
about to log at WARNING ...
about to log at ERROR ...
about to log at CRITICAL ...
Calling decorated foo with False
about to log at DEBUG ...
about to log at INFO ...
about to log at WARNING ...
Calling decorated foo with True
about to log at DEBUG ...
about to log at INFO ...
about to log at WARNING ...
about to log at ERROR ...
Actually logged at DEBUG
Actually logged at INFO
Actually logged at WARNING
Actually logged at ERROR
about to log at CRITICAL ...
Actually logged at CRITICAL
</pre></div>
</div>
As you can see, actual logging output only occurs when an event is logged whose
severity is ERROR or greater, but in that case, any previous events at lower
severities are also logged.
You can of course use the conventional means of decoration:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>@log_if_errors(logger)
def foo(fail=False):
 ...
</pre></div>
</div>
</div>
<div class="section" id="formatting-times-using-utc-gmt-via-configuration">
<h2>Formatting times using UTC (GMT) via configuration<a class="headerlink" href="#formatting-times-using-utc-gmt-via-configuration" title="Permalink to this headline">¶</a></h2>
Sometimes you want to format times using UTC, which can be done using a class
such as <cite>UTCFormatter</cite>, shown below:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>import logging
import time

class UTCFormatter(logging.Formatter):
 converter = time.gmtime
</pre></div>
</div>
and you can then use the <code class="docutils literal notranslate">UTCFormatter</code> in your code instead of
<a class="reference internal" href="../library/logging.html#logging.Formatter" title="logging.Formatter"><code class="xref py py-class docutils literal notranslate">Formatter</code></a>. If you want to do that via configuration, you can
use the <a class="reference internal" href="../library/logging.config.html#logging.config.dictConfig" title="logging.config.dictConfig"><code class="xref py py-func docutils literal notranslate">dictConfig()</code></a> API with an approach illustrated by
the following complete example:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>import logging
import logging.config
import time

LOGGING = {
 &#39;version&#39;: 1,
 &#39;disable_existing_loggers&#39;: False,
 &#39;formatters&#39;: {
 &#39;utc&#39;: {
 &#39;()&#39;: UTCFormatter,
 &#39;format&#39;: &#39;%(asctime)s %(message)s&#39;,
 },
 &#39;local&#39;: {
 &#39;format&#39;: &#39;%(asctime)s %(message)s&#39;,
 }
 },
 &#39;handlers&#39;: {
 &#39;console1&#39;: {
 &#39;class&#39;: &#39;logging.StreamHandler&#39;,
 &#39;formatter&#39;: &#39;utc&#39;,
 },
 &#39;console2&#39;: {
 &#39;class&#39;: &#39;logging.StreamHandler&#39;,
 &#39;formatter&#39;: &#39;local&#39;,
 },
 },
 &#39;root&#39;: {
 &#39;handlers&#39;: [&#39;console1&#39;, &#39;console2&#39;],
 }
}

if __name__ == &#39;__main__&#39;:
 logging.config.dictConfig(LOGGING)
 logging.warning(&#39;The local time is %s&#39;, time.asctime())
</pre></div>
</div>
When this script is run, it should print something like:
<div class="highlight-none notranslate"><div class="highlight"><pre>2015-10-17 12:53:29,501 The local time is Sat Oct 17 13:53:29 2015
2015-10-17 13:53:29,501 The local time is Sat Oct 17 13:53:29 2015
</pre></div>
</div>
showing how the time is formatted both as local time and UTC, one for each
handler.
</div>
<div class="section" id="using-a-context-manager-for-selective-logging">
<h2>Using a context manager for selective logging<a class="headerlink" href="#using-a-context-manager-for-selective-logging" title="Permalink to this headline">¶</a></h2>
There are times when it would be useful to temporarily change the logging
configuration and revert it back after doing something. For this, a context
manager is the most obvious way of saving and restoring the logging context.
Here is a simple example of such a context manager, which allows you to
optionally change the logging level and add a logging handler purely in the
scope of the context manager:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>import logging
import sys

class LoggingContext(object):
 def __init__(self, logger, level=None, handler=None, close=True):
 self.logger = logger
 self.level = level
 self.handler = handler
 self.close = close

def __enter__(self):
 if self.level is not None:
 self.old_level = self.logger.level
 self.logger.setLevel(self.level)
 if self.handler:
 self.logger.addHandler(self.handler)

def __exit__(self, et, ev, tb):
 if self.level is not None:
 self.logger.setLevel(self.old_level)
 if self.handler:
 self.logger.removeHandler(self.handler)
 if self.handler and self.close:
 self.handler.close()
 # implicit return of None =&gt; don&#39;t swallow exceptions
</pre></div>
</div>
If you specify a level value, the logger’s level is set to that value in the
scope of the with block covered by the context manager. If you specify a
handler, it is added to the logger on entry to the block and removed on exit
from the block. You can also ask the manager to close the handler for you on
block exit - you could do this if you don’t need the handler any more.
To illustrate how it works, we can add the following block of code to the
above:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>if __name__ == &#39;__main__&#39;:
 logger = logging.getLogger(&#39;foo&#39;)
 logger.addHandler(logging.StreamHandler())
 logger.setLevel(logging.INFO)
 logger.info(&#39;1. This should appear just once on stderr.&#39;)
 logger.debug(&#39;2. This should not appear.&#39;)
 with LoggingContext(logger, level=logging.DEBUG):
 logger.debug(&#39;3. This should appear once on stderr.&#39;)
 logger.debug(&#39;4. This should not appear.&#39;)
 h = logging.StreamHandler(sys.stdout)
 with LoggingContext(logger, level=logging.DEBUG, handler=h, close=True):
 logger.debug(&#39;5. This should appear twice - once on stderr and once on stdout.&#39;)
 logger.info(&#39;6. This should appear just once on stderr.&#39;)
 logger.debug(&#39;7. This should not appear.&#39;)
</pre></div>
</div>
We initially set the logger’s level to <code class="docutils literal notranslate">INFO</code>, so message #1 appears and
message #2 doesn’t. We then change the level to <code class="docutils literal notranslate">DEBUG</code> temporarily in the
following <code class="docutils literal notranslate">with</code> block, and so message #3 appears. After the block exits, the
logger’s level is restored to <code class="docutils literal notranslate">INFO</code> and so message #4 doesn’t appear. In the
next <code class="docutils literal notranslate">with</code> block, we set the level to <code class="docutils literal notranslate">DEBUG</code> again but also add a handler
writing to <code class="docutils literal notranslate">sys.stdout</code>. Thus, message #5 appears twice on the console (once
via <code class="docutils literal notranslate">stderr</code> and once via <code class="docutils literal notranslate">stdout</code>). After the <code class="docutils literal notranslate">with</code> statement’s
completion, the status is as it was before so message #6 appears (like message
#1) whereas message #7 doesn’t (just like message #2).
If we run the resulting script, the result is as follows:
<div class="highlight-shell-session notranslate"><div class="highlight"><pre>$ python logctx.py
1. This should appear just once on stderr.
3. This should appear once on stderr.
5. This should appear twice - once on stderr and once on stdout.
5. This should appear twice - once on stderr and once on stdout.
6. This should appear just once on stderr.
</pre></div>
</div>
If we run it again, but pipe <code class="docutils literal notranslate">stderr</code> to <code class="docutils literal notranslate">/dev/null</code>, we see the following,
which is the only message written to <code class="docutils literal notranslate">stdout</code>:
<div class="highlight-shell-session notranslate"><div class="highlight"><pre>$ python logctx.py 2&gt;/dev/null
5. This should appear twice - once on stderr and once on stdout.
</pre></div>
</div>
Once again, but piping <code class="docutils literal notranslate">stdout</code> to <code class="docutils literal notranslate">/dev/null</code>, we get:
<div class="highlight-shell-session notranslate"><div class="highlight"><pre>$ python logctx.py &gt;/dev/null
1. This should appear just once on stderr.
3. This should appear once on stderr.
5. This should appear twice - once on stderr and once on stdout.
6. This should appear just once on stderr.
</pre></div>
</div>
In this case, the message #5 printed to <code class="docutils literal notranslate">stdout</code> doesn’t appear, as expected.
Of course, the approach described here can be generalised, for example to attach
logging filters temporarily. Note that the above code works in Python 2 as well
as Python 3.
</div>
</div>

</div>
 </div>
 </div>
 <div class="sphinxsidebar" role="navigation" aria-label="main navigation">
 <div class="sphinxsidebarwrapper">
 <h3><a href="../contents.html">Table Of Contents</a></h3>
 <ul>
<li><a class="reference internal" href="#">Logging Cookbook</a><ul>
<li><a class="reference internal" href="#using-logging-in-multiple-modules">Using logging in multiple modules</a></li>
<li><a class="reference internal" href="#logging-from-multiple-threads">Logging from multiple threads</a></li>
<li><a class="reference internal" href="#multiple-handlers-and-formatters">Multiple handlers and formatters</a></li>
<li><a class="reference internal" href="#logging-to-multiple-destinations">Logging to multiple destinations</a></li>
<li><a class="reference internal" href="#configuration-server-example">Configuration server example</a></li>
<li><a class="reference internal" href="#dealing-with-handlers-that-block">Dealing with handlers that block</a></li>
<li><a class="reference internal" href="#sending-and-receiving-logging-events-across-a-network">Sending and receiving logging events across a network</a></li>
<li><a class="reference internal" href="#adding-contextual-information-to-your-logging-output">Adding contextual information to your logging output</a><ul>
<li><a class="reference internal" href="#using-loggeradapters-to-impart-contextual-information">Using LoggerAdapters to impart contextual information</a><ul>
<li><a class="reference internal" href="#using-objects-other-than-dicts-to-pass-contextual-information">Using objects other than dicts to pass contextual information</a></li>
</ul>
</li>
<li><a class="reference internal" href="#using-filters-to-impart-contextual-information">Using Filters to impart contextual information</a></li>
</ul>
</li>
<li><a class="reference internal" href="#logging-to-a-single-file-from-multiple-processes">Logging to a single file from multiple processes</a></li>
<li><a class="reference internal" href="#using-file-rotation">Using file rotation</a></li>
<li><a class="reference internal" href="#use-of-alternative-formatting-styles">Use of alternative formatting styles</a></li>
<li><a class="reference internal" href="#customizing-logrecord">Customizing <code class="docutils literal notranslate">LogRecord</code></a></li>
<li><a class="reference internal" href="#subclassing-queuehandler-a-zeromq-example">Subclassing QueueHandler - a ZeroMQ example</a></li>
<li><a class="reference internal" href="#subclassing-queuelistener-a-zeromq-example">Subclassing QueueListener - a ZeroMQ example</a></li>
<li><a class="reference internal" href="#an-example-dictionary-based-configuration">An example dictionary-based configuration</a></li>
<li><a class="reference internal" href="#using-a-rotator-and-namer-to-customize-log-rotation-processing">Using a rotator and namer to customize log rotation processing</a></li>
<li><a class="reference internal" href="#a-more-elaborate-multiprocessing-example">A more elaborate multiprocessing example</a></li>
<li><a class="reference internal" href="#inserting-a-bom-into-messages-sent-to-a-sysloghandler">Inserting a BOM into messages sent to a SysLogHandler</a></li>
<li><a class="reference internal" href="#implementing-structured-logging">Implementing structured logging</a></li>
<li><a class="reference internal" href="#customizing-handlers-with-dictconfig">Customizing handlers with <code class="docutils literal notranslate">dictConfig()</code></a></li>
<li><a class="reference internal" href="#using-particular-formatting-styles-throughout-your-application">Using particular formatting styles throughout your application</a><ul>
<li><a class="reference internal" href="#using-logrecord-factories">Using LogRecord factories</a></li>
<li><a class="reference internal" href="#using-custom-message-objects">Using custom message objects</a></li>
</ul>
</li>
<li><a class="reference internal" href="#configuring-filters-with-dictconfig">Configuring filters with <code class="docutils literal notranslate">dictConfig()</code></a></li>
<li><a class="reference internal" href="#customized-exception-formatting">Customized exception formatting</a></li>
<li><a class="reference internal" href="#speaking-logging-messages">Speaking logging messages</a></li>
<li><a class="reference internal" href="#buffering-logging-messages-and-outputting-them-conditionally">Buffering logging messages and outputting them conditionally</a></li>
<li><a class="reference internal" href="#formatting-times-using-utc-gmt-via-configuration">Formatting times using UTC (GMT) via configuration</a></li>
<li><a class="reference internal" href="#using-a-context-manager-for-selective-logging">Using a context manager for selective logging</a></li>
</ul>
</li>
</ul>

<h4>Previous topic</h4>
 <a href="logging.html"
 title="previous chapter">Logging HOWTO</a>
 <h4>Next topic</h4>
 <a href="regex.html"
 title="next chapter">Regular Expression HOWTO</a>
 <div role="note" aria-label="source link">
 <h3>This Page</h3>
 <ul class="this-page-menu">
 <li><a href="../bugs.html">Report a Bug</a></li>
 <li>
 <a href="https://github.com/python/cpython/blob/3.6/Doc/howto/logging-cookbook.rst"
 rel="nofollow">Show Source
 </a>
 </li>
 </ul>
 </div>
 </div>
 </div>
 <div class="clearer"></div>
 </div> 
 <div class="related" role="navigation" aria-label="related navigation">
 <h3>Navigation</h3>
 <ul>
 <li class="right" style="margin-right: 10px">
 <a href="../genindex.html" title="General Index"
 >index</a></li>
 <li class="right" >
 <a href="../py-modindex.html" title="Python Module Index"
 >modules</a> |</li>
 <li class="right" >
 <a href="regex.html" title="Regular Expression HOWTO"
 >next</a> |</li>
 <li class="right" >
 <a href="logging.html" title="Logging HOWTO"
 >previous</a> |</li>
 <li><img src="../_static/py.png" alt=""
 style="vertical-align: middle; margin-top: -1px"/></li>
 <li><a href="https://www.python.org/">Python</a> &#187;</li>
 <li>
 <a href="../index.html">3.6.7 Documentation</a> &#187;
 </li>

<li class="nav-item nav-item-1"><a href="index.html" >Python HOWTOs</a> &#187;</li>
 <li class="right">

</ul>
 </div> 
 <div class="footer">
 &copy; <a href="../copyright.html">Copyright</a> 2001-2023, Python Software Foundation.
 
 The Python Software Foundation is a non-profit corporation.
 <a href="https://www.python.org/psf/donations/">Please donate.</a>
 
 Last updated on Dec 18, 2023.
 <a href="../bugs.html">Found a bug</a>?
 
 Created using <a href="http://sphinx.pocoo.org/">Sphinx</a> 1.7.6.
 </div>

</body>
</html>

MR SKB