From: Dave <da...@ch...> - 2001-08-24 17:25:39
|
Sam Steingold wrote: > > * Honorable Dave <da...@ch...> writes: > > Sam Steingold wrote: > > > * Honorable Dave <da...@ch...> writes: > > > > Looking at the Tasks page at sf.net, I followed the JVM Compilation > > > > Is this still a wanted feature for CLISP? If so I am very interested. > > > this is a _very_ wanted feature. > > > would you like to work on it? > > Yes I would. Where can I get the specifics? > > How familiar with JVM are you? > How familiar with CLISP bytecodes are you? > start with reading the appropriate impnotes parts, bytecode.d and eval.d > get and build CVS CLISP. > subscribe to clisp-devel and discuss this matter there. I am fairly familiar with the java classfile and bytecode spec, having worked with a couple of the bytecode libs out there, as well as doing some quick-n-dirty bytecode mangling for custom classloaders. I also feel I have a good understanding of the JVM, classloading, class introspection/reflection, etc. As for CLISP bytecodes, I dont have any direct experience with them, but I will get started with the impnotes you mentioned. -- Dave |
From: Dave <da...@ch...> - 2001-08-30 01:06:01
|
Sam Steingold wrote: > start with reading the appropriate impnotes parts, bytecode.d and eval.d So I read the CLISP internals section of the impnotes and eval.d. I also picked up "Programming for the Java Virtual Machine" by Joshua Engel <http://www.amazon.com./exec/obidos/ASIN/0201309726> which may be useful for this project, it even has a chapter on implementing Scheme in Java, among others. So whats next? Are there pre-existing implementation ideas that have been talked about? I didnt see any in the clisp-devel logs. Thanks, Dave |
From: Sam S. <sd...@gn...> - 2001-08-30 13:28:49
|
> * In message <Pin...@ch...> > * On the subject of "Re: [clisp-list] JVM Compilation" > * Sent on Wed, 29 Aug 2001 19:11:15 -0600 (MDT) > * Honorable Dave <da...@ch...> writes: > > Sam Steingold wrote: > > start with reading the appropriate impnotes parts, bytecode.d and eval.d > > So I read the CLISP internals section of the impnotes and eval.d. I > also picked up "Programming for the Java Virtual Machine" by Joshua > Engel <http://www.amazon.com./exec/obidos/ASIN/0201309726> which may > be useful for this project, it even has a chapter on implementing > Scheme in Java, among others. > > So whats next? Are there pre-existing implementation ideas that have > been talked about? I didnt see any in the clisp-devel logs. none that I know of. maybe Bruno will share some of his ideas with us. you will probably have to start with a design spec and then post it here. -- Sam Steingold (http://www.podval.org/~sds) Support Israel's right to defend herself! <http://www.i-charity.com/go/israel> Read what the Arab leaders say to their people on <http://www.memri.org/> MS Windows: error: the operation completed successfully. |
From: Dave <da...@ch...> - 2001-09-01 00:32:50
|
Sam Steingold wrote: > * Honorable Dave <da...@ch...> writes: > > So whats next? Are there pre-existing implementation ideas that have > > been talked about? I didnt see any in the clisp-devel logs. > > none that I know of. > maybe Bruno will share some of his ideas with us. > you will probably have to start with a design spec and then post it > here. Sounds fair. How formal of a design spec are you wanting? Are you just wanting to find out how I plan to implement this? I have reviewed Kawa's design, which by the way, seems to be where that JVM book I mentioned got its info for Scheme -> Java bytecode, only the author neglects to mention Kawa. Do you know of any other Lisp to Java resources? I am also wondering if there are certain goals for a CLISP bytecode -> Java bytecode compile, such as, for example, an FFI to Java. This would add Swing and AWT as available GUI toolkits available to CLISP, at least if you are willing and able to run your Lisp code under the JVM. Another useful result of doing this compiler, is that I realized that by generating Java bytecode, it is then possible to run it through GCJ to produce a native binary. Maybe thats one of the reasons you guys had in mind for this compiler. Dave |
From: Sam S. <sd...@gn...> - 2001-09-01 15:38:41
|
> * In message <Pin...@ch...> > * On the subject of "Re: [clisp-list] JVM Compilation" > * Sent on Fri, 31 Aug 2001 18:38:25 -0600 (MDT) > * Honorable Dave <da...@ch...> writes: > > Sam Steingold wrote: > > * Honorable Dave <da...@ch...> writes: > > > So whats next? Are there pre-existing implementation ideas that have > > > been talked about? I didnt see any in the clisp-devel logs. > > > > none that I know of. > > maybe Bruno will share some of his ideas with us. > > you will probably have to start with a design spec and then post it > > here. > > Sounds fair. How formal of a design spec are you wanting? Are you > just wanting to find out how I plan to implement this? you have a free hand. no formalities. nevertheless, I would love to know what you are doing, so I would appreciate clarity of design and verbosity of comments. > I have reviewed Kawa's design, which by the way, seems to be where > that JVM book I mentioned got its info for Scheme -> Java bytecode, > only the author neglects to mention Kawa. Do you know of any other > Lisp to Java resources? nope. sorry. > I am also wondering if there are certain goals for a CLISP bytecode -> > Java bytecode compile, such as, for example, an FFI to Java. This > would add Swing and AWT as available GUI toolkits available to CLISP, this would be lovely! > at least if you are willing and able to run your Lisp code under the > JVM. Another useful result of doing this compiler, is that I realized > that by generating Java bytecode, it is then possible to run it > through GCJ to produce a native binary. Maybe thats one of the > reasons you guys had in mind for this compiler. yes, I thought about it, although it would probably be better to write a direct GCC frontend for the CLISP byte codes (e.g., as a bytecode->C pretty-printer), at least if the issue is speed. -- Sam Steingold (http://www.podval.org/~sds) Support Israel's right to defend herself! <http://www.i-charity.com/go/israel> Read what the Arab leaders say to their people on <http://www.memri.org/> Every day above ground is a good day. |