txscript: Tighten standardness pubkey checks.

[davecgh]

This tightens the multisig and pay-to-pubkey standard script identification functions to use the same strict pubkey requirements as the consensus rules since standardness rules are generally intended to be more restrictive than the consensus rules which implies they are at a minimum at least as restrictive.

The tests are also updated to deal with the additional restriction accordingly.

[matheusd]

Just a small comment: other places in the code using the same naming style (Is*()/Check*()) usually have the Check() function return a more specific error while the Is() function just checks whether there is an error (eg IsSSTx vs CheckSSTX).

Do you think it's worth following the same style here or are the possible errors for pubkey so simple and obvious that it doesn't really make a difference?

[davecgh]

It's a fair point, although in this particular case, I'd rather not do that for a couple of reasons:

• This existing error has already been there for years and has tests in the reference data which specifically look for it
• This change is really also part of a much bigger change that I've been working on which significantly optimizes the entire script engine and returning an error that gets thrown away is counter to that goal and even creates an allocation due to the fact it's an interface

As an aside, I wouldn't use the stake package as an example of code representative of dcrd in general since it, unfortunately generally does not follow the style used in the rest of code and ultimately needs to be significantly cleaned up as a result.

[matheusd]

Ok, thanks for the response!