Re: Bison/flex compatibility revisited

[Bruce Lilly]

John wrote:

> Can you elaborate? The current plan is to go ahead and generate
> the prefixed linkable symbols directly, eliminating many of the flex
> macros. Do you see a reason why bison macros will interfere with this
> scheme?

No, I don't think there will be a problem with use with bison.

However, while the current method (with the flex-generated header
and YY_HEADER_NO_UNDEFS) should also work with yacc, byacc, etc.,
the planned change will put the onus on the user to provide the
equivalent renaming when using with yacc (that could be via macros
or via a separate processing step with sed, etc.).

In other words, while the current method can work for a package
that includes a lexical analyzer and parser with renamed prefixes,
regardless of whether bison, yacc, byacc, etc. are used to build
the parser, the planned change will make things more complicated
for package developers (since the user of the package might build
with bison, or with yacc, ...).