You should pass this as a parameter. That makes the test easier as you can craft any tx/pattern you want and make sure you have coverage for all the detection methods.

This whole IFP match is totally useless. You're losing more time with this branch than it takes to just fail matching the IFP script.
Just loop until miner is found or you're running out of outpoint.

If that's not going to be reused anywhere it's not worth making it a function. It makes testing less efficient as you don't supply a list of the know miners

This is a copy paste of the same function in alias server.
What do you conclude ?

This isn't the only thing currently living in config.js that isn't really a config and should require a diff to change. Could move this and also > > the opReturn constants into their own files.

imo that's beyond the scope of this diff.

Agreed, and that should be done first so that this diff can be rebased on top.

I'd like to keep the test vectors in the same place as the miners, so it's clear that adding another miner requires adding associated test
blocks in this format.

That rational makes zero sense. The whole point of unit tests is to make sure the function does what it claims. The unit tests are not supposed to run on every single possible case ! You're testing some happy paths and the edges cases, then the function does always the same job so unless there is a new feature, you don't need more test. You're not testing block parsing on the whole blockchain, and update the test vectors for every new block right ?

That's wrong and that's my point. It's the second time already, and this is causing code duplication + test duplication. This is not OK.