Possible macro hygiene interference with ..k match clause expansion

[zyrolasting]

What version of Racket are you using?
7.7.0.5

What program did you run?

(let ([a 1]) (match (list a a a) [(list a a ..1) #t] [_ #f]))

What should have happened?

I expected #t, which would be consistent with:

(let ([a 1]) (match (list a a a) [(list a a a) #t] [_ #f]))

If you got an error message, please include it here.

a247: unbound identifier;, where 247 is circumstantial.

Please include any other relevant details

• Fedora 30.
• When reproducing in the REPL, the error will also mention that #%top is not bound.
• The error reproduces for both __k and ..k patterns
• The error does not reproduce if only one bound identifier appears in the match clause. This is what makes me suspect macro hygiene is related.

  (let ([a 1]) (match (list a a a) [(list a ..1) #t] [_ #f]))

[samth]

See also #3425

[samth]

This isn't a macro hygiene issue -- it's a bug in how match generates the equality comparison in the presence of repetition.

[zyrolasting]

Thank you for clarifying! Please feel free to close if this thread does not contribute any new information.