Debugging Tools

Baserow development dependencies include some useful tools for debugging that you can use.

snoop

snoop is a powerful set of Python debugging tools.

Automatic tracing

One of the common things to do is to use the @snoop decorator or the snoop context manager to trace the execution of a piece of Python code and show how variable values change over time:

@snoop
def test():
    for i in range(5):
        a = i*2

# or

with snoop:
    for i in range(5):
        a = i*2

The depth can be controlled with depth parameter, e.g. @snoop(depth=2) for tracing functions that go deep.

Objects and dictionaries can be expanded automatically to show all their items or attributes using the watch_explode parameter taking a list of watched variable names:

@snoop(watch_explode=['d'])
def test():
    d = {'key1': 0, 'key2': 1}
    for i in range(5):
        d["key1"] += 1

Pretty printing

Besides automatic tracing, variables can be pretty printed manually with pp function:

d = {'key1': 0, 'key2': 1}
pp(d)

Note that import snoop or from snoop import pp is not necessary as snoop is installed and available automatically.

django-extensions

django-extensions is available to provide a variety of features like a shell with auto-imported Django models or a command to list all registered urls.

You can use django-extensions commands inside backend docker containers:

• django-admin shell_plus starts an interactive Python shell with loaded Django contexts and imported models.
• django-admin show_urls lists all registered urls in the Baserow.

django-silk

django-silk is a live profiling and inspection tool for executed requests and database queries.

The interface can be accessed at http://localhost:8000/silk/ after Baserow is started in the debug mode. Every request is logged and can be analyzed, including the list of performed database queries.

django-silk can be also configured and used for profiling using the Python’s built-in profiler, see the official documentation for details.

Silk adds profiling overhead to every development request. To disable it when measuring request or OpenTelemetry performance, set BASEROW_ENABLE_SILK=off in .env.docker-dev and recreate the backend container with just dc-dev up -d backend. A plain container restart does not reload its environment. Direct Docker Compose usage reads the same value from its selected environment or .env file.

flower

Flower is an open source web application for monitoring and managing Celery clusters. It provides real-time information about the status of Celery workers and tasks.

The interface can be accessed at http://localhost:5555/ after the Baserow development environment has started.