no-unsafe-any time frame?

[johnknoop]

Hi

no-unsafe-any has been listed as not implemented on the roadmap for a long time. Are there any plans to ever implement it?

It's the single reason we can't move from TSLint.

[bradzacher]

roadmap is a bit of a misnomer for that file... It's less of a roadmap, and more a document to help people migrate from tslint.

When we started this project, we were planning on feature parity with TSLint, but we realised that not every rule in tslint should be migrated across for various reasons (such as too many unused options, features that are handled by other rules better, or rules that just aren't used).

We now are in more of a holding-pattern in regards to tslint rules - we don't implement any, and ask the community to submit PRs for rules they think they need. This creates a higher bar for entry, and ensures that we only have rules that the community actually will want to use.

In this case, nobody has asked for it, so nobody has implemented it. There has been an idea I've been kicking around for a little while (#791) to do something stricter than no-unsafe-any, but I haven't had time to work on that.

It's gotten a fair bit of traffic, so it might be a decent idea for me to prioritise it.
We've been a little bit swamped with the 2.0 release, and the TS3.7 release, so we haven't had a lot of time spare.

[bradzacher]

I'm going to close this in favour of #791.
I'm more than happy if someone has a go at a minimal implementation of the rule described in the issue, so that it has parity with tslint's no-unsafe-any rule.