trace-obfuscation: Add is_card_number func to identify credit cards

[ajgajg1134]

What does this PR do?

A brief description of the change being made with this pull request.

Motivation

As part of R&D week I've been playing with libdatadog and FFI in go, I realized some of the obfuscation the agent does was missing from libdatadog so I decided to try and port the datadog-agent's "IsCreditCard" function. For now this won't be used, but can serve as a starting point for additional obfuscation in libdatadog

Additional Notes

My rust is a bit... rusty... 😎 . So please make sure what I'm doing is idiomatic and I'm not making any silly mistakes / missing any good refactors here. The original go code it pretty tough to read and I believe this new implementation is far more readable, however I avoided changing the design too much to reduce the likelihood I introduce any bugs in the port.

How to test the change?

All the unit tests from datadog-agent were brought over (And a few new tests for my own benefit were added)

[pawelchcki]

Overall looks good to me.
I'd suggest setting up a criterion benchmark for is_card_number too - just so we can track the 🚤 in the future

[ajgajg1134]

Overall looks good to me. I'd suggest setting up a criterion benchmark for is_card_number too - just so we can track the 🚤 in the future

I added some criterion benchmarks which are hopefully about what you expected? I've never written criterion benchmarks before but these seemed to be "good enough". Let me know if I shouldn't be using a slice of data and grouping and just do a single test of 1 valid credit card.

Also I'm not sure if just adding an extra [[bench]] section to the cargo toml was the right way to add this as an extra file 🤔

[thedavl]

you're going to want to add the path to your new bench file here to fix that alpine docker build CI issue!

[pawelchcki]

One nit from me and one important comment from Levi.

Otherwise it looks good to me.

Thanks @ajgajg1134 🙇

[r1viollet]

Should we merge this one ?

[morrisonlevi]

@r1viollet, I think it's waiting on a review from the serverless team, since it's part of a folder they own.

[thedavl]

Aside from the clippy errors, everything looks good! Just fix those before merging

[ajgajg1134]

Aside from the clippy errors, everything looks good! Just fix those before merging

So it looks like those clippy failures are outside this PR? They seem to be in the trace-normalization code which isn't touched here? Do you still want me to fix those in this PR?

[thedavl]

Aside from the clippy errors, everything looks good! Just fix those before merging

So it looks like those clippy failures are outside this PR? They seem to be in the trace-normalization code which isn't touched here? Do you still want me to fix those in this PR?

Ah, sorry. I missed what files it was failing on. Must have been some clippy change, as it was green before.

That's fine then, you're good to merge!

[ajgajg1134]

Aside from the clippy errors, everything looks good! Just fix those before merging

So it looks like those clippy failures are outside this PR? They seem to be in the trace-normalization code which isn't touched here? Do you still want me to fix those in this PR?

Ah, sorry. I missed what files it was failing on. Must have been some clippy change, as it was green before.

That's fine then, you're good to merge!

Sweet! I don't have permissions to merge so hopefully someone else can do so :)

[ivoanjo]

Thanks for the great contribution y'all, pressing the magic button!