hackeron@athena.mit.edu (Harris L Gilliam) (11/27/89)
Here is a summary of all the responses I got to my question about public domain clu and clu libraries. CLU is copyrighted but no longer licensed so it is readily available. Since you're at MIT already, you should have no problem getting it. Paul Johnson (prj@pm-prj.lcs.mit.edu, x3-1945) can tell you how to get a copy via ftp and answer your other questions. The quick answer to your question about linking CLU code with object code from other languages is that you can't. The only other code with which CLU can be linked is assembler (using the CLU assembler that comes with the CLU system). BTW, CLU is not *just* used for teaching at MIT. Professor Guttag has an ongoing theorem proving/verification project for which the tools are all written in CLU. Jon Allen's group was (or is) writing some VLSI tools in CLU. There may be other uses as well. Sharon Perl sharon@lcs.mit.edu ------------------------------------------------------------------------- If someone wants to write a public domain CLU compiler, I can offer several useful modules, including a scanner, a parser, parts of the semantic analyzer and a prettyprinter. These modules were originally designed as abstract data types for program manipulation such as program synthesis, program trans- formation, partial evaluation, interpretation, and flow analysis. They may be slightly less efficient than more specialized modules would be, but I believe that they are easy to use. Matti Jokinen moj@utu.fi ------------------------------------------------------------------------- Not to my knowledge. I am working on something that may turn into a full-blown compiler in the fullness of time. If it does, I will try to make it public domain. But don't hold your breath. > If there are no full implementations then where could I find some info on > the inards of clu so that I (or some other brave soul) can implement it? Try the CLU experts at MIT. I believe that Joh Guttag has got someone doing CLU compiler development right now. They are supposed to be working on a compiler that emits C code ... so that they can port CLU to run on a DECStation. > Also, are there any ftp sites (or mail archives) with clu source to do > various neat things (like some cool data types or something)? Pass ... I'd like to hear about this myself. > Finally has anyone been able to mix clu with some other object code (like > C or LISP or FORTRAN, etc) Not to my knowledge. And before you ask, it would be a LOT of work to allow this. The MIT CLU implementation uses unconventional procedure calling sequences and representations for primitive types. The GC is very very touchy about what values can be safely stored where. In short it is a monstrous tarpit. The new work that John Guttag and Co are doing may solve this problem ... > and are there any libraries for doing X windows stuff in clu ? Not to my knowledge. At one point MIT had a CLU library for X10 (X10.4 I think), but to my knowledge they have not changed it for X11. -- Steve From: Stephen Crawley <scc%computer-lab.cambridge.ac.uk@NSFNET-RELAY.AC.UK> ---------------------------------------------------------------------------- It is really funny to ask such a question in news from a site so near to the CLU creation lab. Contact someone in the Laboratory of Computer Science at MIT in Barbara Liskov's group. The best way is to ask Stephen Garland (garland@larch.lcs.mit.edu). Best regards, -- Miki HERMANN Centre de Recherche en Informatique de Nancy -------------------------------------------- e-mail: hermann@loria.crin.fr post: CRIN, B.P. 239, F-54506 Vandoeuvre-les-Nancy Cedex, FRANCE phone: +33 83.91.20.00 ext(poste) 28.68 ---------------------------------------------------------------------------- Send mail to dcurtis@meson.lcs.mit.edu . She (Dorothy Curtis) is working now on building a portable compiler for Clu. She knows about getting other sources of Clu and X routines for Clu. -- Boaz. Thanks to all who replied by mail !!! |Harris L. Gilliam () 4 Ames St. Cambridge MA 02139 | |Internet : hackeron@athena.mit.edu () hgilliam@gertie.media.mit.edu | |UUCP : !bloom-beacon!mit-athena!hackeron | |GEnie : H.GILLIAM1 | |Harris L. Gilliam () 4 Ames St. Cambridge MA 02139 | |Internet : hackeron@athena.mit.edu () hgilliam@gertie.media.mit.edu | |UUCP : !bloom-beacon!mit-athena!hackeron | |GEnie : H.GILLIAM1 |