Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > comp.lang.c > #27665 > unrolled thread
| Started by | gnuist007@hotmail.com |
|---|---|
| First post | 2012-10-24 20:13 -0700 |
| Last post | 2012-10-26 10:24 -0700 |
| Articles | 4 — 2 participants |
Back to article view | Back to comp.lang.c
Using lisp code in emacs inside a C program gnuist007@hotmail.com - 2012-10-24 20:13 -0700
Re: Using lisp code in emacs inside a C program gnuist007@hotmail.com - 2012-10-25 08:46 -0700
Re: Using lisp code in emacs inside a C program "Pascal J. Bourguignon" <pjb@informatimago.com> - 2012-10-28 18:42 +0100
Re: Using lisp code in emacs inside a C program gnuist007@hotmail.com - 2012-10-26 10:24 -0700
| From | gnuist007@hotmail.com |
|---|---|
| Date | 2012-10-24 20:13 -0700 |
| Subject | Using lisp code in emacs inside a C program |
| Message-ID | <38c1b402-deb0-497a-adc3-640c525ab8c2@g4g2000yqk.googlegroups.com> |
Dear Lispers, I wrote a humble file parsing/structure-searching program in emacs lisp. Essentially a collection of elisp functions called by a main function. I may need the function to grow and to maintain it. I think its easiest in the elisp in which I started the project. I want to convert it to an executable or a form that runs without the need for emacs. Hence, I want to be able to take whatever lisp interpreter code from emacs and link to it or incorporate inside my C executable. The main functions from the emacs that my program uses are essentially navigation and regexp and non-greedy wildcards, ie "*?" and multiline searches. I dont know if any of the standard regexp library in C or C++ supports such extended regexp. Even sed does not support it and I avoid perl. The only other option would be javascript but I want to stay loyal to lisp if you are able to give me enough ideas on this matter. I would be even willing to put together a small lisp interpreter in C with your help and then bootstrap it using Lisp etc and then dump the binary image after it has computed rest of the higher lisp definitions in primitive lisp and link it to my code in elisp. There is really no gui facility that it uses. I am sure many of you have done this kind of thing as I keep on hearing but have to go through all the steps myself. Gnuist
[toc] | [next] | [standalone]
| From | gnuist007@hotmail.com |
|---|---|
| Date | 2012-10-25 08:46 -0700 |
| Message-ID | <8281076c-13c5-4e5f-be0b-1f06dd9a82a5@g4g2000yqk.googlegroups.com> |
| In reply to | #27665 |
On Oct 25, 8:17 am, Sohail Somani <soh...@taggedtype.net> wrote: > On 24/10/2012 11:13 PM, gnuist...@hotmail.com wrote: > > > I would > > be even willing to put together a small lisp interpreter in C with > > your help and then bootstrap it using Lisp etc and then dump the > > binary image after it has computed rest of the higher lisp definitions > > in primitive lisp and link it to my code in elisp. > > Try http://ecls.sourceforge.net/ which is an embeddable Common Lisp. > It's pretty handy and good at what it says it does. Thanks for the link and the terminology of embedded. I want to know more about the authors, but I could not find any clue. Also, if anyone has used it and its compatibility with the emacs. I found a file called emacs.el in it and I am pasting it here so people can guess what it does and offer comments. What might be the structure of the program like? Certainly, would be interesting to read the source and if anyone has done so and if it is commented well or obfuscated? Here the personality of the authors comes in handy to know their motivations for the project. Maybe, I also try to find the core file where he has written the interpreter in C. Can anyone point it out and the general structure of the program? emacs.el 167 lines (153 with data), 4.3 kB (require 'cl) (defvar ecl-search-string "") (defun query-replace-ecl-doc (from-string to-string &optional delimited start end) (interactive (query-replace-read-args "Query replace" nil)) (let ((remaining (member (buffer-file-name (current-buffer)) ecl-doc- files))) (dolist (i (or remaining ecl-doc-files)) (let ((b (find-buffer-visiting i))) (unless (equal b (current-buffer)) (switch-to-buffer b) (beginning-of-buffer))) (perform-replace from-string to-string t nil delimited nil nil start end)))) (defun query-replace-regexp-ecl-doc (from-string to-string &optional delimited start end) (interactive (query-replace-read-args "Query replace" nil)) (let ((remaining (member (buffer-file-name (current-buffer)) ecl-doc- files))) (dolist (i (or remaining ecl-doc-files)) (let ((b (find-buffer-visiting i))) (unless (equal b (current-buffer)) (switch-to-buffer b) (beginning-of-buffer))) (query-replace-regexp from-string to-string delimited start end)))) (defun search-ecl-doc (string) (interactive "sString: ") (setq ecl-search-string string) (let ((remaining (member (buffer-file-name (current-buffer)) ecl-doc- files))) (dolist (i (or remaining ecl-doc-files)) (let ((b (find-buffer-visiting i))) (unless (equal b (current-buffer)) (print b) (switch-to-buffer b) (beginning-of-buffer))) (print '*) (setq case-fold-search t) (if (search-forward string nil t) (return))))) (defun search-next-ecl-doc () (interactive) (search-ecl-doc ecl-search-string)) (defun back-to-emacs () (interactive) (switch-to-buffer "emacs.el")) (defun next-ecl-doc () (interactive) (let ((remaining (member (buffer-file-name (current-buffer)) ecl-doc- files))) (when (cdr remaining) (switch-to-buffer (find-buffer-visiting (cadr remaining)))))) (global-set-key [?\M-p ?\C-i] 'back-to-emacs) (global-set-key [?\M-p ?\C-s] 'search-ecl-doc ) (global-set-key [?\M-p ?\C-n] 'search-next-ecl-doc ) (global-set-key [?\M-p ?\C-m] 'next-ecl-doc ) (global-set-key [?\M-p ?\C-p] 'ecl-load-symbols) (setq auto-mode-alist (acons "\\.d\\'" 'c-mode auto-mode-alist)) (setq ecl-doc-files (mapcar (lambda (x) ;(set-buffer "emacs.el") (concat (subseq (buffer-file-name (current-buffer)) 0 -8) x)) '( "asdf.xmlf" "bibliography.xmlf" "clos.xmlf" "compiler.xmlf" "copyright.xmlf" "declarations.xmlf" "discarded.xml" "discarded.xmlf" "ecl.xml" "ecldev.xmlf" "embed.xmlf" "ffi.xmlf" "gc.xmlf" "internals.xmlf" "interpreter.xmlf" "intro.xmlf" "io.xmlf" "macros.xmlf" "memory.xmlf" "mop.xmlf" "mp.xmlf" "os.xmlf" "pde.xmlf" "preface.xmlf" "ref_c_arrays.xml" "ref_c_characters.xml" "ref_c_conditions.xml" "ref_c_conses.xml" "ref_c_data_flow.xml" "ref_c_environment.xml" "ref_c_evaluation.xml" "ref_c_filenames.xml" "ref_c_files.xml" "ref_c_hash_tables.xml" "ref_c_numbers.xml" "ref_c_objects.xml" "ref_c_packages.xml" "ref_c_printer.xml" "ref_c_reader.xml" "ref_c_sequences.xml" "ref_c_streams.xml" "ref_c_strings.xml" "ref_c_structures.xml" "ref_c_symbols.xml" "ref_c_system_construction.xml" "ref_c_types_and_classes.xml" "ref_embed.xmlf" "ref_memory.xmlf" "ref_mp.xmlf" "ref_os.xmlf" "ref_signals.xmlf" "schemas.xml" "signals.xmlf" "uffi/ref_aggregate.xml" "uffi/ref_declare.xml" "uffi/ref_func_libr.xml" "uffi/ref_object.xml" "uffi/ref_primitive.xml" "uffi/ref_string.xml" "uffi/schemas.xml" "ansi_arrays.xml" "ansi_characters.xml" "ansi_conses.xml" "ansi_data_flow.xml" "ansi_environment.xml" "ansi_evaluation.xml" "ansi_filenames.xml" "ansi_files.xml" "ansi_hash_tables.xml" "ansi_numbers.xml" "ansi_objects.xml" "ansi_overview.xml" "ansi_packages.xml" "ansi_printer.xml" "ansi_reader.xml" "ansi_sequences.xml" "ansi_streams.xml" "ansi_strings.xml" "ansi_structures.xml" "ansi_symbols.xml" "ansi_system_construction.xml" "ansi_types.xml" ))) (mapcar 'find-file ecl-doc-files) (defun ecl-doc-revert () (interactive) (mapcar '(lambda (x) (let ((a (find-buffer-visiting x))) (and a (switch-to-buffer a) (revert-buffer t t)))) ecl-doc-files)) (defun ecl-doc-save () (interactive) (mapcar '(lambda (x) (let ((a (find-buffer-visiting x))) (and a (switch-to-buffer a) (save-buffer 0)))) ecl-doc-files))
[toc] | [prev] | [next] | [standalone]
| From | "Pascal J. Bourguignon" <pjb@informatimago.com> |
|---|---|
| Date | 2012-10-28 18:42 +0100 |
| Message-ID | <87vcduijcb.fsf@informatimago.com> |
| In reply to | #27676 |
gnuist007@hotmail.com writes: > On Oct 25, 8:17 am, Sohail Somani <soh...@taggedtype.net> wrote: >> On 24/10/2012 11:13 PM, gnuist...@hotmail.com wrote: >> >> > I would >> > be even willing to put together a small lisp interpreter in C with >> > your help and then bootstrap it using Lisp etc and then dump the >> > binary image after it has computed rest of the higher lisp definitions >> > in primitive lisp and link it to my code in elisp. >> >> Try http://ecls.sourceforge.net/ which is an embeddable Common Lisp. >> It's pretty handy and good at what it says it does. > > Thanks for the link and the terminology of embedded. > I want to know more about the authors, but I could not find any clue. The current maintainer is Juan Jose Garcia Ripoll. http://lisp-univ-etc.blogspot.fr/2012/06/juan-jose-garcia-ripoll-is-physicist.html http://ecls.sourceforge.net/ http://ecls.sourceforge.net/resources.html > Also, if anyone has used it and its compatibility with the emacs. All CL implementation is compatible with emacs, in the sense that you can use emacs to edit CL programs, and you can run it with M-x inferior-lisp RET, or even for most of them (including ecl) with M-x slime RET. > I found a file called emacs.el in it and I am pasting it here so people > can guess what it does and offer comments. What might be the structure > of the program like? Certainly, would be interesting to read the > source and if anyone has done so and if it is commented well or > obfuscated? Almost all projects written by programmers using emacs will contain some emacs lisp file containing some utility emacs commands the authors use to edit the project files. > Here the personality of the authors comes in handy to know > their motivations for the project. > > Maybe, I also try to find the core file where he has written the > interpreter in C. Can anyone point it out and the general structure of > the program? The ecl subdirectory named "src" contains: [pjb@triton :0 src]$ ls ./ aclocal.m4 clx/ config.sub* gc/ lsp/ ../ bare.lsp.in cmp/ configure* gmp/ new-cmp/ CHANGELOG* c/ compile.lsp.in* configure.in h/ util/ Makefile.in clos/ config.guess* doc/ install.sh* I would say cmp/ and new-cmp/ contain compilers. The c/ contains some C sources, including a file named "interpreter.d". Concerning personalities, I would say yours doesn't bode well, given that you didn't fetch the sources and see that yourself. One wonders about your motivations... ;-) -- __Pascal Bourguignon__ http://www.informatimago.com
[toc] | [prev] | [next] | [standalone]
| From | gnuist007@hotmail.com |
|---|---|
| Date | 2012-10-26 10:24 -0700 |
| Message-ID | <615275b8-7ddb-47ff-8836-814a2c631cc2@s12g2000vbw.googlegroups.com> |
| In reply to | #27665 |
On Oct 25, 11:00 pm, William Gardella <gardell...@gmail.com> wrote:
> Rivka Miller <rivkaumil...@gmail.com> writes:
> > \begin{quotation}
> > ECL (ECL for short) uses standard C calling conventions for Lisp
> > compiled functions, which allows C programs to easily call Lisp
> > functions and vice versa. No foreign function interface is required:
> > data can be exchanged between C and Lisp with no need for conversion.
> > \end{quotation}
>
> > How did ECL achieve this?
>
> > R
>
> Not to oversimplify too much, but a compiled ECL function *is* a C
> function. The Debian description of the package lists among its
> features:
>
> > ECL stands for Embeddable Common-Lisp. The ECL project is an effort to
> > modernize Giuseppe Attardi's ECL environment to produce an implementation of
> > the Common-Lisp language which complies to the ANSI X3J13 definition of the
> > language.
> > The current ECL implementation features:
> > * A bytecodes compiler and interpreter.
> > * A translator to C.
> > ...
>
> So it's like similar initiatives for Pascal, FORTRAN, etc.--the end
> result is a C program that you can poke at with tools designed to work
> with C (binutils, gdb, etc.) and which can communicate natively with C
> functions. I've not used it and have heard mixed reviews (mostly on
> #emacs) about its performance relative to other compiled CL
> implementations. But if what you need is Lisp inside C, it seems this
> is your bet--it can even be used to make C shared libraries, it looks
> like.
>
> Chicken, being an R5RS-to-C compiler, has similar aspirations for the
> world of Scheme.
>
You didnt answer her question on the computer science of translation,
code flow graphing, or any useful aspect than advertising your Chicken
and she posted only on this newsgroup, while I wanted it to be in the
related newsgroups. Maybe someone can give the pertinent referenecs to
key useful papers.
G
[toc] | [prev] | [standalone]
Back to top | Article view | comp.lang.c
csiph-web