Is there a high-level way to read in lines from an output file and have the types recognized by the structure of the contents?

Question

Suppose I have an output file that I want to read and each line was created by joining several types together, prepending and appending the list braces,

[('tupleValueA','tupleValueB'), 'someString', ('anotherTupleA','anotherTupleB')]

I want to read the lines in. Now I can read them in, and operate on the string to assign values and types but I was wondering if Python had a higher level method for this.

After building a function to do this I tried to find a higher level approach but didn't find one.

Answer 1

What you are looking for is eval. But please keep in mind that this function will evaluate and execute the lines. So don't run it on untrusted input ever!

>>> print eval("[('tupleValueA', 1), 'someString']")
[('tupleValueA', 1), 'someString']

If you have control over the script that generate the output file, then I would suggest you use json encoding. JSON format is very similar to the python string representation of lists and dictionaries. And will be much more secure and robust to read from.

>>> import json
>>> json.dumps(['foo', {'bar': ('baz', None, 1.0, 2)}])
'["foo", {"bar": ["baz", null, 1.0, 2]}]'
>>> json.loads('["foo", {"bar": ["baz", null, 1.0, 2]}]')
["foo", {"bar": ["baz", null, 1.0, 2]}]

Answer 2

The problem you describe is conventionally referred to as serialization. JavaScript Object Notation (JSON) is one popular serialization protocol.

Answer 3

Probably it would be better to store the data with a module like pickle in the first place, instead of using normal strings. This way you avoid a lot of problems that come with eval and get a more robust solution.