affirm

This small library improves Python assert error messages to contain more useful information.

I like to use assert's liberaly thoughout my code to document my assumptions and when one of them fails, I really like to know as much as possible about what failed and why.

Installation

$ pip install affirm

Using the assert statement

Before affirm, if you run a simple script in Python:

a = 1
b = 2
assert a > b

The result will be:

Traceback (most recent call last):
  File "test.py", line 3, in <module>
    assert a > b
AssertionError

We can see the traceback, but not the actuall values that caused the assert to fail, which is not very useful for debugging.

If you import affirm at the top of the script, like so:

import affirm
a = 1
b = 2
assert a > b

You'll get:

Traceback (most recent call last):
  File "test.py", line 4, in <module>
    assert a > b
AssertionError: assertion (a > b) failed with a=1, b=2

Which is much more useful.

Note that the standard behaviour of supplying a message with the assert statement still works:

import affirm
a = 1
b = 2
assert a > b, "something went wrong"

Will give you:

Traceback (most recent call last):
  File "test.py", line 4, in <module>
    assert a > b
AssertionError: something went wrong

Using the affirm function

There's only one problem with using the standard assert statement. If you catch the exception and print it yourself:

import affirm
a = 1
b = 2
try:
    assert a > b
except Exception as e:
    print(str(e))

The result will be:

Yes, absolutely nothing. Calling str on AssertionError results in an empty string.

If you want to be able to catch the assertion errors and print the messages into e.g. log, you'll need to use the affirm function instead of the assert statement, like so:

from affirm import affirm
a = 1
b = 2
try:
    affirm(a > b)
except Exception as e:
    print(str(e))

Now we get the expected:

assertion (a > b) failed with a=1, b=2