|Source to source translator. email@example.com (1995-08-30)|
|Re: Source to source translator. johnm@soda.CSUA.Berkeley.EDU (1995-09-04)|
|Re: Source to source translator. firstname.lastname@example.org (1995-09-05)|
|Re: Source to source translator. email@example.com (1995-09-07)|
|Re: Source to source translator. firstname.lastname@example.org (1995-09-16)|
|From:||email@example.com (Antoun Kanawati)|
|Date:||Thu, 7 Sep 1995 13:05:15 GMT|
firstname.lastname@example.org (Mokrane Amzal) wrote:
> Do anybody have ever heard about a system which allows abstract tree
> rewriting based on rules given in high level language ?
At work, we use cback to do C->C transformations on cfront output.
These result in smaller object files, and more easily debuggable
binaries (mangling is minimized for automatic variables, so you
don't have to guess about __<n> prefixes).
Cback is driven by a rule-set; I am not sure whether some parts of
it are hardcoded. For more info try sending email to newcode.com.
They should have an email@example.com, or something of that sort.
If you just want to deal with syntax trees, you can easily develop
the rewriting system yourself in LISP, using lists to represent
trees. The exercise is a standard introductory course homework
for LISP-related topics (AI and compilers/interpreters). For really
elaborate works, you may wish to look at the sources for extend-syntax,
syntax-case, and the other macro proposals for Scheme. I don't have the
URL for the Scheme repository handy, but you can find it by scanning
Antoun (Tony) Kanawati
Return to the
Search the comp.compilers archives again.