SnakeByte [15]: Debugging With pdb In The Trenches

Dalle3 idea of debugging code from the view of the code itself.

If debugging is the process of removing software bugs, then programming must be the process of putting them in.

~ Edsger W. Dijkstra

Image Explanation: Above is the image showing the perspective of debugging code from the viewpoint of the code itself. The scene features lines of code on a large monitor, with sections highlighted to indicate errors. In the foreground, anthropomorphic code characters are working to fix these highlighted sections, set against a digital landscape of code lines forming a cityscape.

So meta and canonical.

In any event Dear Readers, it is Wednesday! That means as usual everyday is Wednesday in a startup, you actually work at a company where you enjoy the work or it is SnakeByte [ ] time!

i haven’t written a SnakeByte is quite some time. Also, recently, in a previous blog, I mentioned that I had a little oversite on my part, and my blog went down. i didn’t have alerting turned on ye ole AWS and those gosh darn binlogs where not deleting in ye ole WordPress as such we took the proverbial digger into downtime land. i re-provisioned with an additional sizing of the volume and changed the disc from magnetic to SSD, turned on alerts and we are back in business.

So per my usual routine i grab one of the python books from the book shelf randomly open then read about command or commands and attempt to write a blog as fast as possible.

Today’s random command from Python In A Nutshell is pdb, the Python Debugger. i’ll walk you through the basic of using pdb to debug your Python code, which, as it turns out is better than a bunch of print().

Getting Started with pdb

To leverage pdb, import it in your Python script. You can then set breakpoints manually with pdb.set_trace(). When the execution hits this line, the script pauses, allowing you to engage in an interactive debugging session.

Consider this straightforward example:

# example.py
import pdb

def add(a, b):
    result = a + b
    return result

def main():
    x = 10
    y = 20
    pdb.set_trace()  # Breakpoint set here
    total = add(x, y)
    print(f'Total: {total}')

if __name__ == '__main__':
    main()

Here, we have a simple add function and a main function that invokes add. The pdb.set_trace() in the main function sets a breakpoint where the debugger will halt execution.

Running the Code with pdb

Execute the script from the command line to initiate the debugger:

python example.py

When the execution reaches pdb.set_trace(), you will see the debugger prompt ((Pdb)):

> /path/to/example.py(11)main()
-> total = add(x, y)
(Pdb)

Essential pdb Commands

Once at the debugger prompt, several commands can help you navigate and inspect your code. Key commands include:

• l (list): Displays the surrounding source code.
• n (next): Moves to the next line within the same function.
• s (step): Steps into the next function call.
• c (continue): Resumes execution until the next breakpoint.
• p (print): Evaluates and prints an expression.
• q (quit): Exits the debugger and terminates the program.

Let’s walk through using these commands in our example:

List the surrounding code:

Pdb) 1
  6         def main():
  7             x = 10
  8             y = 20
  9             pdb.set_trace()  # Breakpoint here
 10  ->         total = add(x, y)
 11             print(f'Total: {total}')
 12     
 13         if __name__ == '__main__':
 14             main()

Print variable values:

(Pdb) p x
10
(Pdb) p y
20

Step into the add function:

(Pdb) s
> /path/to/example.py(3)add()
-> result = a + b
(Pdb)

Inspect parameters a and b:

(Pdb) p a
10
(Pdb) p b
20

Continue execution:

(Pdb) c
Total: 30

Advanced pdb Features

For more nuanced debugging needs, pdb offers advanced capabilities:

Conditional Breakpoints: Trigger breakpoints based on specific conditions:

if x == 10:
    pdb.set_trace()

Post-Mortem Debugging: Analyze code after an exception occurs:

import pdb

def faulty_function():
    return 1 / 0

try:
    faulty_function()
except ZeroDivisionError:
    pdb.post_mortem() #they need to add a pre-mortem what can go wrong will...

Command Line Invocation: Run a script under pdb control directly from the command line like the first simple example:

python -m pdb example.py

Effective debugging is crucial for building robust and maintainable software. pdb provides a powerful, interactive way to understand and fix your Python code. By mastering pdb, you can streamline your debugging process, saving time and enhancing your productivity.

pdb, the Python Debugger, is an indispensable tool for any developer. It allows us to set breakpoints, step through code, inspect variables, and evaluate expressions on the fly. This level of control is crucial for diagnosing and resolving issues efficiently. While i used cli in the examples it also works with Jupyter Notebooks.

We’ve covered the basics and advanced features of pdb, equipping you with the knowledge to tackle even the most challenging bugs. As you integrate these techniques into your workflow, you’ll find that debugging becomes less of a chore and more of a strategic advantage unless you create a perfect design which is no code at all!

Until then,

#iwishyouwater <- La Vaca Gigante Wipeouts 2024

𝕋𝕖𝕕 ℂ. 𝕋𝕒𝕟𝕟𝕖𝕣 𝕁𝕣. (@tctjr)

Muzak to Blog By: Apostrophe (‘) by Frank Zappa. i was part of something called the Zappa Ensemble in graduate school as one of the engineers. The musicians where amazing. Apostrohe (‘)is an amazing album. The solo on Uncle Remus is just astounding well as is most of his solos.