While parsing a large JSON, it’s possible that some keys may only exist in certain circumstances, such as when an error is present. It’s not uncommon to get a 200 OK from the server’s API, but then the response you got contains errors that should be checked for.

What’s the best way to handle this?

I know that using things like get() are one way to handle KeyErrors.

if json.get('errors') is not None:
    do_something

But what if we need to keep going? This feels horrible:

if json.get('errors') is not None:
   if json.get('errors').get('code') is not None:
      if json.get('errors').get('code') != 0:
         this_is_bad_throw_error

Is there a better way? For example, ruby has a dig() method that basically condenses that whole line into x = dig(:errors, :code) and then you just check on x if it’s a nil object or not; so 2 lines, instead of 4 (or more if you have to keep going deeper).

I have used try-except before to catch the KeyError, but this isn’t an approach I necessarily care for.

I’m not against external libraries, but due to the location of where the code is being run it’s extremely beneficial to stick to built-ins.