Related articles |
---|
Help ! lex and yacc POSIX extensions amin@rsvl.unisys.com (1993-01-18) |
Newsgroups: | comp.compilers,comp.unix.programmer |
From: | amin@rsvl.unisys.com (Amin Kassem) |
Organization: | Compilers Central |
Date: | Mon, 18 Jan 1993 20:55:42 GMT |
Keywords: | yacc, lex, testing, comment |
In an effort to test our implementation of lex and yacc without having a
biased test, we are soliciting the net and the netters for tests. If you
have a good test case that you think will exercise (or break :-)) the
following features we would like to hear from you. Our objective is to
run small yet good tests to verify the implementation of these
functionalities.
Lex :- 1) Exclusive start conditions (%x).
2) %array vs. %pointer.
3) collating symbols and collation related bracket symbols in EREs.
Yacc :- 1) -b option.
2) -p option.
What other important functionalities (POSIX extensions) do you think should
also be tested?
Our knowledge of lex and yacc is VERY limited. Any helpful hints that
would help us ensure a good test strategy would be greatly appreciated.
Private e-mail is preferred.
Thank you much.
Regards,
--
Amin Kassem - Unisys Corporation
System Software & Interface - R&D
e-mail: amin@rsvl.unisys.com
[I have seen no widely available implementation of lex yet that has all
the Posix extensions. Flex has %x, which is the most useful of the
extensions (to English speakers, at least) so there's lots of code with
%x. The new options to yacc merely change the file names and symbols used
in the generated parser, which I think would be pretty easy to test. I
shamelessly recommend Levine (that's me), Mason, and Brown, "lex&yacc",
published by O'Reilly which is the most complete book on the topic that I
know of. -John]
--
Return to the
comp.compilers page.
Search the
comp.compilers archives again.