On 11/04/2011 01:59 PM, Jason Merrill wrote: > Right, my point is that the comment before > cp_parser_transaction_expression refers to compound_statement, which > seems wrong. > > ...because the compound-statement variant is handled in cp_parser_transaction.
Oh, duh. I was so focused on the attribute portion that I didn't notice the compound-statement mistake. Will fix. r~
