The main objection was that other people might then think about
language-dependent parsing of the Bison skeleton (for example, as in Perl,
because of clashes between the "$N"/"@N" variables, and the internal syntax of
the language).
What's wrong with that? Couldn't it prove to be necessary in order to
support some languages? If you would prefer to point me to a previous
post, that's fine.
http://www.mail-archive.com/[email protected]/msg01308.html
http://www.mail-archive.com/[email protected]/msg01325.html
http://www.mail-archive.com/[email protected]/msg01336.html
If I understood the first two comments well, Akim doesn't want
language-dependent code to pollute the Bison parser. And I tend to agree.
The comments in Akim's last message are addressed by this revision of
the patch. The %skeleton mechanism is not touched at all. People
contributing new kinds of parsers will have to patch "c-lang.m4" and
friends if they want the new skeleton to be selected automatically; it
is not necessary however (and hence the patch is backwards compatible)
if they prefer their users to use %skeleton in order to invoke the new
feature.
Paolo
