Rule suppression does not work for non builtin rules

[kapilmb]

The current implementation supports rule suppression only for builtin rules.

[charlieschmidt]

@kapilmb if it's low on the backlog and if you can point me to the best "testing" method - ie how does one build from source and the best/easiestly test those changes, I'm happy to take a stab at it here. I imagine probably using scriptanalyzer as the C# library for VS debugging is probably easiest, so write a little stub application that invokes the analyzer with the params I'd need on a module with my custom rules & functions? Or is there something more elegant for testing/debugging?

[charlieschmidt]

@kapilmb stab at it here: https://github.com/charlieschmidt/PSScriptAnalyzer

Problems:

1. ends up doing the external rules 1 by 1, instead of the runspacepool that was there already - i suspect theres a way around that, but the GetExternalRecord() method is a bit complicated at first glance, so wasn't sure the best way to do it.
2. the powershell methods that are the rules have to match the diagnosticrecord.rulename exactly - cause the initial ExternalRule.GetName() just has the function name, but the supressionrule list and diagnosticrecord list might have something else and it won't match any longer. which isn't awesome necessarily, but also isn't that horrible of a requirement. The alternative would probably be to modify the ExternalRule after you get the first diagnostic record for it so it has the "real" name maybe, or to modify the SupressRecord method so it does something fancier when matching external rules and supressed records - but then that ends up being a lame exceptional case.

Initial thoughts on either of those problems? Happy to keep working at it. I suspect (1) is probably just a matter of rejiggering the responses from GetExternalRecord so it can still run the rules in parallel and return the appropriate lists. (2) - dunno.

[kapilmb]

@charlieschmidt Thanks for the effort.

1. I have left some comments on your commit.
2. Yes, you are right - There is a potential for inconsistency which can render rule suppression useless. But, your proposal of getting the "real" rule name after examining diagnostic record appears to be promising. Basically, once we have the diagnostic records we can examine them for unique rule names and run suppression using those rule names.

[kapilmb]

@charlieschmidt Merged your contribution in PR #588. Thanks again!

I am leaving this open, as #588 fixes suppression for only AST based external rules. So that leaves the token based external rules not subject to suppression. Will add that through another PR then close this issue.

[kapilmb]

Closing this as the fix in #588 fixes token based external rule suppression too!