Thank you for the bug report! One question to help me understand the desired behavior in your use case.
Congress does support international characters within strings in a policy rule. For example:
'error("あ") :- p("あ", val1), p("あ", val2), not eq(val1, val2)' is legal and accepted by Congress.
However, with the attempted: 'error(?) :- p(?, val1), p(?, val2), not eq(val1, val2)', the international character is used to construct a variable identifier. But the characters allowed in identifiers are restricted according to the following grammar: https://github.com/openstack/congress/blob/master/congress/datalog/Congress.g#L224
So the question is, is your use case calling for international characters in strings or international characters in identifiers?
Hi Sonu,
Thank you for the bug report! One question to help me understand the desired behavior in your use case.
Congress does support international characters within strings in a policy rule. For example:
'error("あ") :- p("あ", val1), p("あ", val2), not eq(val1, val2)' is legal and accepted by Congress.
However, with the attempted: 'error(?) :- p(?, val1), p(?, val2), not eq(val1, val2)', the international character is used to construct a variable identifier. But the characters allowed in identifiers are restricted according to the following grammar: /github. com/openstack/ congress/ blob/master/ congress/ datalog/ Congress. g#L224
https:/
So the question is, is your use case calling for international characters in strings or international characters in identifiers?