Feature/normalize output

[itraviv]

normalizing output - removing machine specific data:

absolute paths are replaced with the basename of the path.
time stamps are removed.
default (CPython) reprs of the form <.... at 0x[0-9a-fA-F] > will not contain the 'at 0x...' ending.

tests:

normalization does not affect the original test results so every test now runs with normalize=True
and normalize=False to preserve compatability.
assert_output can now verify the output of a normalized trace.

readme:

• updated accordingly

[cool-RR]

This looks excellent to me.

Two small notes:

1. [False, True, ] should be (False, True)
2. Where you're using except NotImplementedError you should use pytest.raises.

[cool-RR]

@alexmojaki if you feel like taking a look, feel free, if you're busy that's cool.

[cool-RR]

Wait, another note. Looking at the normalized output, I don't like how the code is now aligned with the variable changes. It's confusing. So when there's no datetime, instead of '', please use ' ' * 15. That way the alignment is preserved. I guess you'll need to change the tests a bit too. Please do all these changes in separate commits than what you've submitted so far, so I could easily compare them.

[cool-RR]

Good, the only thing that's left is that [False, True, ] should be (False, True)

[itraviv]

i swear it was there! 😒

[cool-RR]

Can you change it from (True, False, ) to (True, False)?

[itraviv]

Can you change it from (True, False, ) to (True, False)?

sure

[cool-RR]

@alexmojaki Let me know whether you have time to give it a look before I merge. (I'll squash Itamar's commits.)

[alexmojaki]

Go ahead.

[cool-RR]

I squashed these commits separately and made a release, adding Itamar to the authors file. Congrats on your first contribution!

[itraviv]

thanks!