Extending the tokenizer for it should not be super hard. The advantage would be that without escape character, we do not have full csv support.
Basically:
- It is not really difficult or time consuming to add
- It might be a good new feature as an additional incentive that we also improve
loadtxt.