This is a 3 character change

FWIW, the cost of this change isn't measured in characters or lines modified. It's a question of how much ongoing maintenance cost there is keeping this a stable interface for people to use long-term and whether that's even the API we want people using.

I understand the trade-offs and for my case think it is worth it (obviously). It seems like it follows in the spirit of the module, just at a lower level and the API for that function is very direct to work with when dealing with ASTs. Thank you for considering it.

If deemed a good change I'll be happy to submit a pull request with the change to help things along. I plan on maintaining this as a fork in the meantime, so I'll have it ready. Thanks again.

Went ahead and created the Pull Request to facilitate the decision.

This issue seems related to #12696. Although I don't know how exporting only fixImports helps.

The PR doesn't seem to help facilitate the decision, because it doesn't communicate why fixImports being exported makes sense and how it helps. It just implements it, but this issue is still in NeedsDecision state.

It would help if you provided more information about how it'd be used, etc.

The use case is when you want to update the imports in the AST without the text. For example I'm writing a custom formatter, to format code in a presentation like style, and wanted the import fix. I considered forking or just copying out the needed code, but thought it might be generally useful and it'd be better to not duplicate the functionality if unnecessary.

We have exposed an internal interface to the imports package that might suit your needs, but I don't think we're going to make it public. Feel free to fork x/tools to expose it, at your risk of course.