From 0648a261d3fa9c00daa28da209fb9f84385c7e39 Mon Sep 17 00:00:00 2001 From: Melody Horn Date: Sun, 25 Oct 2020 11:20:16 -0600 Subject: drop ZWJ et al from identifiers --- crowbar_reference_compiler/scanner.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/crowbar_reference_compiler/scanner.py b/crowbar_reference_compiler/scanner.py index fff8c35..eeca318 100644 --- a/crowbar_reference_compiler/scanner.py +++ b/crowbar_reference_compiler/scanner.py @@ -25,7 +25,7 @@ class GenerousTokenList(List[Token]): KEYWORD = re.compile("bool|break|case|char|const|continue|default|do|double|else|enum|extern|float|for|function|if|include|int|long|return|short|signed|sizeof|struct|switch|typedef|unsigned|void|while") -IDENTIFIER = re.compile(r"[\p{L}\p{Pc}\p{Cf}\p{Sk}\p{Mn}][\p{L}\p{Pc}\p{Cf}\p{Sk}\p{Mn}\p{N}]*") +IDENTIFIER = re.compile(r"[\p{L}\p{Pc}\p{Sk}\p{Mn}][\p{L}\p{Pc}\p{Sk}\p{Mn}\p{N}]*") CONSTANT = re.compile(r"""([0-9_]+)|(0[bB][01_]+)|(0[xX][0-9a-fA-F_]+)|([0-9_]+(\.[0-9_]+|[eE][0-9_]+|\.[0-9_]+[eE][0-9_]+))|('([^\'\\]|\\'|\\"|\\\\|\\r|\\n|\\t|\\0|\\x[0-9a-fA-F]{2}|\\u[0-9a-fA-F]{4}|\\U[0-9a-fA-F]{8})')""") STRING_LITERAL = re.compile(r'''"([^\\"]|\\'|\\"|\\\\|\\r|\\n|\\t|\\0|\\x[0-9a-fA-F]{2}|\\u[0-9a-fA-F]{4}|\\U[0-9a-fA-F]{8})*"''') PUNCTUATOR = re.compile(r"->|\+\+|--|>>|<<|<=|>=|&&|\|\||[=!+\-*/%&|^]=|[\[\](){}.,+\-*/%;!&|^~><=]") -- cgit v1.2.3