[antlr-interest] Antlr C++ lexer and EBCDIC

John Lambert johnl at jbase.com
Wed Jul 17 11:44:28 PDT 2002


I am trying to resolve an issue on IBM Z series with an antlr lexer/parser in C++. 
It is mostly working with the exception of the bitsets for the lexer.
Antlr.tool lays down bitsets that represent the ASCII values of the characters because Java uses UTF8 no matter what the platform.
When the resulting C++ code is compiled the character literals are all compiled as EBCDIC but the bitsets are initialized from the serialized Java ones that represent ASCII values. 

Any suggestions as to where the correct place to solve this would be? 
I am thinking of a runtime option on the antlr.tool to force bitset serialization for lexer grammars to map values thru an AtoE conversion table.

John Lambert

 

Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/ 



More information about the antlr-interest mailing list