It is common practice to generate messages of different kinds while the application is running. They can be categorised according to their purpose and three levels are usually defined:
- Information: Status information, for example, or confirmation that some operation has been completed.
- Warning: Notification that a potentially undesirable event has occurred, but which can be ignored if appropriate.
- Error: Notification that an error situation has arisen which needs to be resolved. This type of event usually means that processing cannot proceed.
The messages are written to a logfile so that they are preserved. During development with PyCharm, messages are typically written to the IDE console, but they are not preserved if the application is restarted. It is often useful to be able to examine historical messages in a logfile.
In a Flask application uses standard Python logging which can be customised in several ways. The steps below describe one simple way of implementing logging. For other options, please consult the documentation.
Logging is intialised in the Flask factory function and in this example, uses an environment variable
that is specific to the selected environment. After that, a
logger object must be initialised in
each view file where you want to generate log messages. When you create a
logger object, it picks
up the configuration settings defined in the factory function, so effectively you are simply making
a link to the same logger each time.
Defining the logging level
It is usually desirable to set different logging levels in different contexts. For example, you may
want to see all messages (errors, warnings and information) while you are developing your
application, but you may only want to lof errors in production. The logging level is defined by
the environment variable
LOGLEVEL which is defined in the root-level
config.py file. The example
below defines the logging level differently for each of the three configurations.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
Line 1: Import the
Line 5: Enable debug mode by default
Line 6: Disable tracking messages from Flask-SQLAlchemy. This saves on resources.
Line 10: Enable maximum output from SQLAlchemy during development
Line 11: Set the logging level to the lowest value (all messages) during development
Line 12: Configure the in-built SQLAlchemy logger to output all messages during development
Lines 17 & 23: Compare with line 11
Line2 18 & 24: Compare with line 12.
Although the Python logging system is highly customisable, it also provides a simple way to
create a basic setup. The example below uses the
basciConfig() method to configure the
essential parameters for simple but effective logging. This code should be inserted into the
Flask factory function in
1 2 3
Line 1: Call the
<path to logfile>with the path to the destination of log messages on your system. Be sure to choose a location that is writable.
Line 2: Set the logging level by referencing the configuration variable
Line 3: Define the format for your log messages. If you want to use a format that is different from this example, please refer to the documentation.
Generating log messages
With logging configured as described above, you can now generate arbitrary messages from your code. The outline below illustrates a few examples of how this could be done. All messages will appear in the specified logfile.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17