How to improve a large snapshot instance and make it more readable

I parsed an XML file and obtained a very large instance which contain many objects and lists. I want to make the parsed instance more readable and testable with snapshot data.

Here are my questions:

• What’s the best practice to make a snapshot of a large instance?
• How to make the snapshot file more readable (code highlighting, line breaks)?
• My IDE (VSCode) doesn’t recognize .ambr file, and will not have code highlighting automatically for that snapshot file. Any solution?

Snapshot Approach 1: stringify self.__dict__

def __repr__(self) -> str:
    return f"{type(self).__name__}({self.__dict__})"

output:

  Context({xml_id': 1234, 'links': [...], ...})  # a very long line

Pro: The type in the format Context({ ... }) would be highlighted with different color using Ruby/Python code highlighter.

Con: The class Context has many list properties and the line of snapshot would be very long (> 10,000 columns). It’s hard to read and compare.

Snapshot Approach 2: JSON serialization

def __repr__(self) -> str:
    return json.dumps(self.__dict__, indent=2)

output:

  {
    "xml_id": 1234, 
    "links": [
      ...
    ], 
    ...
  }

Pro: The line breaks make the large instance more readable.

Con: It’s hard to add types to custom objects like the format Context({ ... }) with json format. I want add the types to object strings and make them distinguished with string values by different highlighting color.

Any suggestion?

[iamogbz]

@xareelee also you can look into the matchers option, which I think can implement these cases without the need for monkey patching, as it allows you to replace the object that will be serialized.

[xareelee]

@iamogbz Thanks for the info. I think that it should be documented in docs/ for other users.