In the grammar for match, i.e. for the nonterminal MatchExpression, in MatchArms the right-hand side of the non-last arms only allows skipping the , for BlockExpression. However it seems to be the case that any ExpressionWithBlock works without the ,. Is this correct?
Furthermore, currently the case of a BlockExpression followed by , matches both alternatives of the BlockExpression ,? | Expression ,. Perhaps a nicer rule would use ExpressionWithoutBlock , | ExpressionWithBlock ,?, in effect being similar to my proposed rule for ExpressionStatement in #773, but with comma instead of semicolon.
Finally, in the last arm, the BlockExpression | Expression part seems a bit redundant as a BlockExpression is an Expression.
In the grammar for
match, i.e. for the nonterminalMatchExpression, inMatchArmsthe right-hand side of the non-last arms only allows skipping the,forBlockExpression. However it seems to be the case that anyExpressionWithBlockworks without the,. Is this correct?Furthermore, currently the case of a
BlockExpressionfollowed by,matches both alternatives of theBlockExpression ,? | Expression ,. Perhaps a nicer rule would useExpressionWithoutBlock , | ExpressionWithBlock ,?, in effect being similar to my proposed rule forExpressionStatementin #773, but with comma instead of semicolon.Finally, in the last arm, the
BlockExpression | Expressionpart seems a bit redundant as aBlockExpressionis anExpression.