tag:old.nabble.com,2006:forum-1685Nabble - Gnu - Guile2009-12-09T15:28:55ZGuile is Project GNU's extension language. It is a library designed to help programmers create flexible applications. Using Guile in an application allows programmers to write plug-ins, or modules (there are many names, but the concept is essentially the same) and users to use them to have an application fit their needs. Gnu - Guile home is <a href="http://www.gnu.org/software/guile/" target="_top" rel="nofollow">here</a>.tag:old.nabble.com,2006:post-26719609Re: SRFI-9 on top of raw structs2009-12-09T15:28:55Z2009-12-09T15:28:55ZNeil Jerram<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26719609&i=0" target="_top" rel="nofollow">ludo@...</a> (Ludovic Courtès) writes:
<br><br>> The problem with records is that accessors are defined with
<br>> ‘record-accessor’, as in:
<br>>
<br>> (define module-name
<br>> (record-accessor module-type 'name))
<br>>
<br>> Here, ‘record-accessor’ returns a procedure. Until the compiler has a
<br>> smart inliner, each ‘module-name’ call is actually a procedure call.
<br><br>OK, I see now. Thanks for explaining!
<br><br>Regards,
<br> Neil
<br><br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Guile---Dev-f1686.html" embed="fixTarget[1686]" target="_top" >Gnu - Guile - Dev</a></p>tag:old.nabble.com,2006:post-26714639Re: ERROR: Stack overflow in guile-1.8.72009-12-09T09:57:04Z2009-12-09T09:57:04ZAntonio, Fabio Di Narzo-22009/12/9 Ludovic Courtès <<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26714639&i=0" target="_top" rel="nofollow">ludo@...</a>>:
<div class='shrinkable-quote'><br>> Hi,
<br>>
<br>> "Antonio, Fabio Di Narzo" <<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26714639&i=1" target="_top" rel="nofollow">antonio.dinarzo@...</a>> writes:
<br>>
<br>>> I'm on an AMD64 linux system. I succesfully compiled and installed
<br>>> guile-1.8.7, but when I try to run guile I get the:
<br>>> ERROR: Stack overflow
<br>>
<br>> Did you pass particular CFLAGS to the build system? Normally the
<br>> default stack limit works well when compiling with ‘-O2’.
</div>No. My config.log says:
<br>"
<br>CFLAGS='-I/home/dinarzo/local/include -Wall -Wmissing-prototypes -Werror'
<br>CPP='gcc -E'
<br>"
<br><br>>
<br>>> error message, and I can't reach the repl. "make check" however passes
<br>>> smoothly. I read about this:
<br>>> <a href="http://lists.gnu.org/archive/html/guile-devel/2009-03/msg00114.html" target="_top" rel="nofollow">http://lists.gnu.org/archive/html/guile-devel/2009-03/msg00114.html</a><br>>> and yes, for me too setting the stack limit to 60'000 in eval.c solves
<br>>> the issue. How can I circumvent/solve this problem without resorting
<br>>> to patching the sources?
<br>>
<br>> Putting ‘(debug-set! stack 60000)’ in ~/.guile should solve it.
<br>Sweet!
<br><br>Thanks,
<br>antonio.
<br><br>>
<br>> Thanks,
<br>> Ludo’.
<br>>
<br>>
<br>>
<br>>
<br>>
<br><br><br><br>--
<br>Antonio, Fabio Di Narzo
<br>Ph.D. student at
<br>Department of Statistical Sciences
<br>University of Bologna, Italy
<br><br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Guile---User-f1687.html" embed="fixTarget[1687]" target="_top" >Gnu - Guile - User</a></p>tag:old.nabble.com,2006:post-26712966Re: ERROR: Stack overflow in guile-1.8.72009-12-09T08:13:22Z2009-12-09T08:13:22ZLudovic Courtès-3Hi,
<br><br>"Antonio, Fabio Di Narzo" <<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26712966&i=0" target="_top" rel="nofollow">antonio.dinarzo@...</a>> writes:
<br><br>> I'm on an AMD64 linux system. I succesfully compiled and installed
<br>> guile-1.8.7, but when I try to run guile I get the:
<br>> ERROR: Stack overflow
<br><br>Did you pass particular CFLAGS to the build system? Normally the
<br>default stack limit works well when compiling with ‘-O2’.
<br><br>> error message, and I can't reach the repl. "make check" however passes
<br>> smoothly. I read about this:
<br>> <a href="http://lists.gnu.org/archive/html/guile-devel/2009-03/msg00114.html" target="_top" rel="nofollow">http://lists.gnu.org/archive/html/guile-devel/2009-03/msg00114.html</a><br>> and yes, for me too setting the stack limit to 60'000 in eval.c solves
<br>> the issue. How can I circumvent/solve this problem without resorting
<br>> to patching the sources?
<br><br>Putting ‘(debug-set! stack 60000)’ in ~/.guile should solve it.
<br><br>Thanks,
<br>Ludo’.
<br><br><br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Guile---User-f1687.html" embed="fixTarget[1687]" target="_top" >Gnu - Guile - User</a></p>tag:old.nabble.com,2006:post-26710527ERROR: Stack overflow in guile-1.8.72009-12-09T05:49:23Z2009-12-09T05:49:23ZAntonio, Fabio Di Narzo-2Hi all.
<br>I'm on an AMD64 linux system. I succesfully compiled and installed
<br>guile-1.8.7, but when I try to run guile I get the:
<br>ERROR: Stack overflow
<br>error message, and I can't reach the repl. "make check" however passes
<br>smoothly. I read about this:
<br><a href="http://lists.gnu.org/archive/html/guile-devel/2009-03/msg00114.html" target="_top" rel="nofollow">http://lists.gnu.org/archive/html/guile-devel/2009-03/msg00114.html</a><br>and yes, for me too setting the stack limit to 60'000 in eval.c solves
<br>the issue. How can I circumvent/solve this problem without resorting
<br>to patching the sources?
<br>Tell me if you need any more details about my system setup.
<br><br>I tried today branch_1.8 snapshot to see if the problem is still
<br>present, but here "make check" stops with an unrelated error (which
<br>probably deserves a different thread).
<br><br>Cheers,
<br>antonio.
<br>--
<br>Antonio, Fabio Di Narzo
<br>Ph.D. student at
<br>Department of Statistical Sciences
<br>University of Bologna, Italy
<br><br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Guile---User-f1687.html" embed="fixTarget[1687]" target="_top" >Gnu - Guile - User</a></p>tag:old.nabble.com,2006:post-26707913Re: [Guile-commits] GNU Guile branch, master, updated. release_1-9-5-129-g7230aaf2009-12-09T02:13:11Z2009-12-09T02:13:11ZLudovic Courtès-3Hey Andy!
<br><br>"Andy Wingo" <<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26707913&i=0" target="_top" rel="nofollow">wingo@...</a>> writes:
<br><br>> +scm_c_make_structv (SCM vtable, size_t n_tail, size_t n_init, scm_t_bits *init)
<br><br>Cool! I guess we can have a ‘make-struct’ opcode that would call this.
<br>WDYT? I’m interested because currently ‘make-struct’ is ~4 times slower
<br>than ‘cons’ (using the ‘cons’ opcode), which is a shameful
<br>discrimination.
<br><br>Besides, please please follow GCS. Also, make sure change log entries
<br>describe the changes, not the rationale, and list all the variables,
<br>functions, or macros that were changed.
<br><br>Thanks,
<br>Ludo’.
<br><br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Guile---Dev-f1686.html" embed="fixTarget[1686]" target="_top" >Gnu - Guile - Dev</a></p>tag:old.nabble.com,2006:post-26703091Re: SRFI-9 on top of raw structs2009-12-08T16:04:13Z2009-12-08T16:04:13ZLudovic Courtès-3Hi Neil,
<br><br>Neil Jerram <<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26703091&i=0" target="_top" rel="nofollow">neil@...</a>> writes:
<br><br>> <a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26703091&i=1" target="_top" rel="nofollow">ludo@...</a> (Ludovic Courtès) writes:
<br><br>[...]
<br><br>>> In the ‘wip-vlist’ branch (which is about implementing Bagwell’s vlists)
<br>>> SRFI-9 is reimplemented in terms of raw structs, instead of records.
<br>>
<br>> Well I enjoyed a bit of reading, and I don't see any problem with these
<br>> changes, but I don't see why "it makes it easy to write a
<br>> macro-generating macro akin to Dybvig’s ‘define-integrable’". Couldn't
<br>> that have been achieved just by rewriting the SRFI-9 define-macro as
<br>> define-syntax, but still using make-record-type etc.?
<br><br>The problem with records is that accessors are defined with
<br>‘record-accessor’, as in:
<br><br> (define module-name
<br> (record-accessor module-type 'name))
<br><br>Here, ‘record-accessor’ returns a procedure. Until the compiler has a
<br>smart inliner, each ‘module-name’ call is actually a procedure call.
<br><br>Conversely, the ‘define-inlineable’ macro in srfi-9.scm leads to
<br>accessor definitions along these lines:
<br><br> (define module-name-procedure
<br> (lambda (x) (struct-ref x 0)))
<br><br> (define-syntax module-name
<br> (lambda (x)
<br> (syntax-case x ()
<br> ((_ obj)
<br> #'(struct-ref obj 0))
<br> (_
<br> #'module-name-procedure))))
<br><br>Thus, an expression like ‘(module-name (current-module))’ is effectively
<br>expanded to ‘(struct-ref (current-module) 0)’, whereas
<br>‘(procedure? module-name)’ is expanded to ‘(procedure? module-name-procedure)’.
<br><br>IOW, this is a poor man’s inliner. But still, I find the approach
<br>pretty neat. :-)
<br><br>Thanks,
<br>Ludo’.
<br><br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Guile---Dev-f1686.html" embed="fixTarget[1686]" target="_top" >Gnu - Guile - Dev</a></p>tag:old.nabble.com,2006:post-26702724Re: Another posting to remove2009-12-08T15:27:11Z2009-12-08T15:27:11ZNeil JerramWendy Ann Tarzian <<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26702724&i=0" target="_top" rel="nofollow">wendyanntarzian@...</a>> writes:
<br><br>> Do you know who maintains the archive? I assumed it was you.
<br><br>Hi Wendy,
<br><br>People at gnu.org are only responsible for .gnu.org websites. I presume
<br>that there might also have been a copy of the posting at
<br><a href="http://lists.gnu.org/archive/html/guile-user/2005-04/index.html" target="_top" rel="nofollow">http://lists.gnu.org/archive/html/guile-user/2005-04/index.html</a>, but it
<br>is not there now - and I would guess that that is because the gnu.org
<br>webmasters have removed it according to your request.
<br><br>To remove the copies of the posting at www.archivum.org and
<br>www.mail-archive.com, you will need to contact the webmasters of those
<br>websites.
<br><br>Regards,
<br> Neil
<br><br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Guile---User-f1687.html" embed="fixTarget[1687]" target="_top" >Gnu - Guile - User</a></p>tag:old.nabble.com,2006:post-26702500Re: [Guile-commits] GNU Guile branch, master, updated. release_1-9-5-104-g8f79d4e2009-12-08T15:08:49Z2009-12-08T15:08:49ZNeil Jerram<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26702500&i=0" target="_top" rel="nofollow">ludo@...</a> (Ludovic Courtès) writes:
<br><br>> Ludo’, amazed by the pace at which things have been falling into place
<br>> over the last couple of weeks!
<br><br>Indeed!
<br><br> Neil
<br><br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Guile---Dev-f1686.html" embed="fixTarget[1686]" target="_top" >Gnu - Guile - Dev</a></p>tag:old.nabble.com,2006:post-26702496Re: SRFI-9 on top of raw structs2009-12-08T15:08:16Z2009-12-08T15:08:16ZNeil Jerram<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26702496&i=0" target="_top" rel="nofollow">ludo@...</a> (Ludovic Courtès) writes:
<br><br>> Hello,
<br><br>Hi Ludo,
<br><br>> In the ‘wip-vlist’ branch (which is about implementing Bagwell’s vlists)
<br>> SRFI-9 is reimplemented in terms of raw structs, instead of records.
<br><br>Well I enjoyed a bit of reading, and I don't see any problem with these
<br>changes, but I don't see why "it makes it easy to write a
<br>macro-generating macro akin to Dybvig’s ‘define-integrable’". Couldn't
<br>that have been achieved just by rewriting the SRFI-9 define-macro as
<br>define-syntax, but still using make-record-type etc.?
<br><br>Regards,
<br> Neil
<br><br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Guile---Dev-f1686.html" embed="fixTarget[1686]" target="_top" >Gnu - Guile - Dev</a></p>tag:old.nabble.com,2006:post-26682131SRFI-9 on top of raw structs2009-12-07T09:25:36Z2009-12-07T09:25:36ZLudovic Courtès-3Hello,
<br><br>In the ‘wip-vlist’ branch (which is about implementing Bagwell’s vlists)
<br>SRFI-9 is reimplemented in terms of raw structs, instead of records.
<br><br>The advantage is that it makes it easy to write a macro-generating macro
<br>akin to Dybvig’s ‘define-integrable’ [0] and use it to define record
<br>accessors such that direct calls to accessors are effectively inlined.
<br><br>On the attached use case, inlining reduces execution time by ~15%. It
<br>also has a noticeable impact on the vlist implementation itself.
<br><br>If there are no objections I’ll commit it to ‘master’.
<br><br>Thanks,
<br>Ludo’.
<br><br>[0] <a href="http://www.scheme.com/tspl3/syntax.html#./syntax:s57" target="_top" rel="nofollow">http://www.scheme.com/tspl3/syntax.html#./syntax:s57</a><br> Thanks to Andy for pointing it out!
<br><br><br />(use-modules (srfi srfi-9)
<br> (ice-9 time))
<br><br>(define-record-type <foo>
<br> (make-foo x)
<br> foo?
<br> (x get-x))
<br><br>(define s (make-foo 1))
<br><br>(define n 7000000)
<br><br>(time (let loop ((i n))
<br> (and (> i 0)
<br> (begin
<br> (get-x s)
<br> (loop (1- i))))))
<br><br>(time (let ((get-x get-x)) ;; prevent inlining
<br> (let loop ((i n))
<br> (and (> i 0)
<br> (begin
<br> (get-x s)
<br> (loop (1- i)))))))
<br><p>From forum: <a href="http://old.nabble.com/Gnu---Guile---Dev-f1686.html" embed="fixTarget[1686]" target="_top" >Gnu - Guile - Dev</a></p>tag:old.nabble.com,2006:post-26681858Re: [Guile-commits] GNU Guile branch, master, updated. release_1-9-5-104-g8f79d4e2009-12-07T09:09:09Z2009-12-07T09:09:09ZLudovic Courtès-3Hi Andy,
<br><br>"Andy Wingo" <<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26681858&i=0" target="_top" rel="nofollow">wingo@...</a>> writes:
<br><div class='shrinkable-quote'><br>> commit 9ea31741dad29ae123e468a203b72df6d190f6e1
<br>> Author: Andy Wingo <<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26681858&i=1" target="_top" rel="nofollow">wingo@...</a>>
<br>> Date: Sat Dec 5 10:52:18 2009 +0100
<br>>
<br>> fluids are tc7 objects
<br>>
<br>> If you're wondering what I'm doing, I'm trying to eventually reimplement
<br>> smobs in terms of structs, so that applicable smobs can just follow the
<br>> applicable struct dispatch path. But to do that I have to get structs
<br>> initialized before things that use smobs, which means transforming a
<br>> bunch of smobby things to tc7 things. But this transformation is good
<br>> for performance anyway, and we currently have a glut of unused tc7s,
<br>> so here we go...
</div><br>I hadn’t seen this commit series until now but I like it! :-)
<br><br>I’ve always thought that smobs-as-structs would be a good thing, not
<br>only to reduce code size and simplify procedure dispatch, but also to
<br>remove arbitrary limits like on the number of smob types.
<br><br>Keep up!
<br><br>Thanks,
<br>Ludo’, amazed by the pace at which things have been falling into place
<br>over the last couple of weeks!
<br><br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Guile---Dev-f1686.html" embed="fixTarget[1686]" target="_top" >Gnu - Guile - Dev</a></p>tag:old.nabble.com,2006:post-26674873Re: guile-aspell : spellchecking wth Guile2009-12-07T01:42:52Z2009-12-07T01:42:52ZLudovic Courtès-3Hi Mike,
<br><br>Mike Gran <<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26674873&i=0" target="_top" rel="nofollow">spk121@...</a>> writes:
<br><br>> I wrote a bit of glue code for Guile and Gnu Aspell, a spellchecking library.
<br>> I packaged it as a Guile library and saved it at
<br>> <a href="http://www.lonelycactus.com/aspell.html" target="_top" rel="nofollow">http://www.lonelycactus.com/aspell.html</a><br><br>Looks fun!
<br><br>One nitpick: I’d prefer ‘(aspell)’ instead of ‘(aspell aspell)’ as the
<br>module name. :-)
<br><br>IMO it’d a nice addition to submit upstream when you feel like it.
<br><br>Thanks,
<br>Ludo’.
<br><br><br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Guile---User-f1687.html" embed="fixTarget[1687]" target="_top" >Gnu - Guile - User</a></p>tag:old.nabble.com,2006:post-26674797Re: use-module not finding module in guile-1.9.52009-12-07T01:39:19Z2009-12-07T01:39:19ZLudovic Courtès-3Hi Linas,
<br><br>It could be that (dbi dbi) is found but that the ‘load-extension’ in
<br>there fails, leading to this (arguably suboptimal) error message.
<br><br>Can you first check whether ‘(search-path %load-path "dbi/dbi.scm")’
<br>yields the correct answer?
<br><br>Thanks,
<br>Ludo’.
<br><br><br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Guile---User-f1687.html" embed="fixTarget[1687]" target="_top" >Gnu - Guile - User</a></p>tag:old.nabble.com,2006:post-26674959Re: use-module not finding module in guile-1.9.52009-12-07T01:39:19Z2009-12-07T01:39:19ZLudovic Courtès-3Hi Linas,
<br><br>It could be that (dbi dbi) is found but that the ‘load-extension’ in
<br>there fails, leading to this (arguably suboptimal) error message.
<br><br>Can you first check whether ‘(search-path %load-path "dbi/dbi.scm")’
<br>yields the correct answer?
<br><br>Thanks,
<br>Ludo’.
<br><br><br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Guile---Dev-f1686.html" embed="fixTarget[1686]" target="_top" >Gnu - Guile - Dev</a></p>tag:old.nabble.com,2006:post-26669224Re: Another posting to remove2009-12-06T13:23:38Z2009-12-06T13:23:38ZWendy Ann Tarzian<html><head></head><body><div style='font-family: Arial; font-size: 12pt; color: #000000'>Do you know who maintains the archive? I assumed it was you.<BR>----- Original Message -----<BR>From: "Linas Vepstas" <<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26669224&i=0" target="_top" rel="nofollow">linasvepstas@...</a>><BR>To: "Wendy Ann Tarzian" <<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26669224&i=1" target="_top" rel="nofollow">wendyanntarzian@...</a>><BR>Cc: <a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26669224&i=2" target="_top" rel="nofollow">Guile-user@...</a>, <a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26669224&i=3" target="_top" rel="nofollow">webmasters@...</a><BR>Sent: Sunday, December 6, 2009 3:16:11 PM GMT -06:00 US/Canada Central<BR>Subject: Re: Another posting to remove<BR><BR>Hi Wendy,<BR><BR>The below is located on some mailing list archive. No one here<BR>has the ability to do a thing about this. You would have to contact<BR>the people who maintain the archive.<BR><BR>--linas<BR><BR>2009/12/6 Wendy Ann Tarzian <<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26669224&i=4" target="_top" rel="nofollow">wendyanntarzian@...</a>><BR>><BR>> Please remove this posting from internet access.<BR>><BR>> <a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26669224&i=5" target="_top" rel="nofollow">guile-user@...</a> (thread)<BR>> ... People identified -Seeking assistance regarding Bill of Rights and National Security Breaches by Illinois Leaders and others, Wendy Tarzian, 2005/04/16 ...<BR>> <a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26669224&i=6" target="_top" rel="nofollow">www.archivum.info/guile-user@...</a>/2005-04/threads.html<BR></div></body></html><p>From forum: <a href="http://old.nabble.com/Gnu---Guile---User-f1687.html" embed="fixTarget[1687]" target="_top" >Gnu - Guile - User</a></p>tag:old.nabble.com,2006:post-26669168Re: Another posting to remove2009-12-06T13:16:11Z2009-12-06T13:16:11ZLinas Vepstas-3Hi Wendy,
<br><br>The below is located on some mailing list archive. No one here
<br>has the ability to do a thing about this. You would have to contact
<br>the people who maintain the archive.
<br><br>--linas
<br><br>2009/12/6 Wendy Ann Tarzian <<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26669168&i=0" target="_top" rel="nofollow">wendyanntarzian@...</a>>
<br>>
<br>> Please remove this posting from internet access.
<br>>
<br>> <a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26669168&i=1" target="_top" rel="nofollow">guile-user@...</a> (thread)
<br>> ... People identified -Seeking assistance regarding Bill of Rights and National Security Breaches by Illinois Leaders and others, Wendy Tarzian, 2005/04/16 ...
<br>> <a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26669168&i=2" target="_top" rel="nofollow">www.archivum.info/guile-user@...</a>/2005-04/threads.html
<br><br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Guile---User-f1687.html" embed="fixTarget[1687]" target="_top" >Gnu - Guile - User</a></p>tag:old.nabble.com,2006:post-26669108use-module not finding module in guile-1.9.52009-12-06T13:09:53Z2009-12-06T13:09:53ZLinas Vepstas-3Hi,
<br><br>I'm experiencing trouble loading a module in guile-1.9.5
<br><br>I maintain the guile-dbi module at <a href="http://home.gna.org/guile-dbi/" target="_top" rel="nofollow">http://home.gna.org/guile-dbi/</a><br>and am having trouble getting guile-1.9.5 to find the module:
<br><br>$ /usr/local/bin/guile
<br>Guile Scheme interpreter 0.5 on Guile 1.9.5
<br>Copyright (C) 2001-2008 Free Software Foundation, Inc.
<br><br>Enter `,help' for help.
<br>scheme@(guile-user)> (use-modules (dbi dbi))
<br><br>Backtrace:
<br>In system/repl/common.scm:
<br> 85: 0* [repl-eval # #]
<br>In system/base/compile.scm:
<br> 239: 1 [compile (use-modules (dbi dbi)) #:key ...]
<br>In system/base/compile.scm:
<br> 185: 2 [lp # # # ...]
<br>In ice-9/boot-9.scm:
<br>1775: 3 [save-module-excursion #<program b31fa0 at
<br>language/scheme/compile-tree-il.scm:29:3 ()>]
<br>In unknown file:
<br> ?: 4* [dynamic-wind #<program b31f20 at ice-9/boot-9.scm:1774:18 ()> ...]
<br>In language/scheme/compile-tree-il.scm:
<br> 32: 5* [#<program b31fa0 at language/scheme/compile-tree-il.scm:29:3 ()>]
<br>In ice-9/psyntax-pp.scm:
<br>11057: 6 [#<program 9aad40 at ice-9/psyntax-pp.scm:11040:8 (#{\x\\\
<br>1648}# . #{\rest\\\ 1649}#)> # ...]
<br>In unknown file:
<br> ?: 7* [with-fluid* #<fluid 10> c ...]
<br>In ice-9/psyntax-pp.scm:
<br>6736: 8* [#{\chi-top-sequence\\\ 162}# (# *unspecified*) () (#) ...]
<br>In ice-9/psyntax-pp.scm:
<br>6720: 9 [#{\dobody\\\ 866}# ((process-use-modules (list #))
<br>*unspecified*) () ...]
<br>In ice-9/psyntax-pp.scm:
<br>3311: 10 [#{\chi-top\\\ 166}# (process-use-modules (list (list #))) () ...]
<br>In ice-9/psyntax-pp.scm:
<br>7854: 11 [#{\top-level-eval-hook\\\ 90}# # #]
<br>In unknown file:
<br> ?: 12* [primitive-eval ("noexpand" (process-use-modules (list #)))]
<br> ?: 13* [process-use-modules (((dbi dbi)))]
<br>In ice-9/boot-9.scm:
<br>2992: 14* [process-use-modules (((dbi dbi)))]
<br>In unknown file:
<br> ?: 15* [map #<program 9d8b80 at ice-9/boot-9.scm:2991:25 (mif-args)> ((#))]
<br>In ice-9/boot-9.scm:
<br>2993: 16* [#<program 9d8b80 at ice-9/boot-9.scm:2991:25 (mif-args)> ((dbi dbi))]
<br>In ice-9/boot-9.scm:
<br>2144: 17 [resolve-interface (dbi dbi)]
<br><br>ERROR: In procedure resolve-interface:
<br>ERROR: no code for module (dbi dbi)
<br>scheme@(guile-user)>
<br><br><br>The basic dbi.scm is located in /usr/local/share/guile/site/dbi
<br><br>Copying it to /usr/share/guile/site/dbi does not help.
<br><br>The module is found, and works, with guile1.8.6
<br>The total contents of dbi.scm is
<br><br>(define-module (dbi dbi))
<br><br>(load-extension "libguile-dbi" "init_dbi")
<br><br>; exported primitives
<br><br>(export dbi-open
<br> dbi-close
<br> dbi-query
<br> dbi-get_row
<br> dbi-get_status)
<br><br><br><br>--linas
<br><br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Guile---User-f1687.html" embed="fixTarget[1687]" target="_top" >Gnu - Guile - User</a></p>tag:old.nabble.com,2006:post-26669116use-module not finding module in guile-1.9.52009-12-06T13:09:53Z2009-12-06T13:09:53ZLinas Vepstas-3Hi,
<br><br>I'm experiencing trouble loading a module in guile-1.9.5
<br><br>I maintain the guile-dbi module at <a href="http://home.gna.org/guile-dbi/" target="_top" rel="nofollow">http://home.gna.org/guile-dbi/</a><br>and am having trouble getting guile-1.9.5 to find the module:
<br><br>$ /usr/local/bin/guile
<br>Guile Scheme interpreter 0.5 on Guile 1.9.5
<br>Copyright (C) 2001-2008 Free Software Foundation, Inc.
<br><br>Enter `,help' for help.
<br>scheme@(guile-user)> (use-modules (dbi dbi))
<br><br>Backtrace:
<br>In system/repl/common.scm:
<br> 85: 0* [repl-eval # #]
<br>In system/base/compile.scm:
<br> 239: 1 [compile (use-modules (dbi dbi)) #:key ...]
<br>In system/base/compile.scm:
<br> 185: 2 [lp # # # ...]
<br>In ice-9/boot-9.scm:
<br>1775: 3 [save-module-excursion #<program b31fa0 at
<br>language/scheme/compile-tree-il.scm:29:3 ()>]
<br>In unknown file:
<br> ?: 4* [dynamic-wind #<program b31f20 at ice-9/boot-9.scm:1774:18 ()> ...]
<br>In language/scheme/compile-tree-il.scm:
<br> 32: 5* [#<program b31fa0 at language/scheme/compile-tree-il.scm:29:3 ()>]
<br>In ice-9/psyntax-pp.scm:
<br>11057: 6 [#<program 9aad40 at ice-9/psyntax-pp.scm:11040:8 (#{\x\\\
<br>1648}# . #{\rest\\\ 1649}#)> # ...]
<br>In unknown file:
<br> ?: 7* [with-fluid* #<fluid 10> c ...]
<br>In ice-9/psyntax-pp.scm:
<br>6736: 8* [#{\chi-top-sequence\\\ 162}# (# *unspecified*) () (#) ...]
<br>In ice-9/psyntax-pp.scm:
<br>6720: 9 [#{\dobody\\\ 866}# ((process-use-modules (list #))
<br>*unspecified*) () ...]
<br>In ice-9/psyntax-pp.scm:
<br>3311: 10 [#{\chi-top\\\ 166}# (process-use-modules (list (list #))) () ...]
<br>In ice-9/psyntax-pp.scm:
<br>7854: 11 [#{\top-level-eval-hook\\\ 90}# # #]
<br>In unknown file:
<br> ?: 12* [primitive-eval ("noexpand" (process-use-modules (list #)))]
<br> ?: 13* [process-use-modules (((dbi dbi)))]
<br>In ice-9/boot-9.scm:
<br>2992: 14* [process-use-modules (((dbi dbi)))]
<br>In unknown file:
<br> ?: 15* [map #<program 9d8b80 at ice-9/boot-9.scm:2991:25 (mif-args)> ((#))]
<br>In ice-9/boot-9.scm:
<br>2993: 16* [#<program 9d8b80 at ice-9/boot-9.scm:2991:25 (mif-args)> ((dbi dbi))]
<br>In ice-9/boot-9.scm:
<br>2144: 17 [resolve-interface (dbi dbi)]
<br><br>ERROR: In procedure resolve-interface:
<br>ERROR: no code for module (dbi dbi)
<br>scheme@(guile-user)>
<br><br><br>The basic dbi.scm is located in /usr/local/share/guile/site/dbi
<br><br>Copying it to /usr/share/guile/site/dbi does not help.
<br><br>The module is found, and works, with guile1.8.6
<br>The total contents of dbi.scm is
<br><br>(define-module (dbi dbi))
<br><br>(load-extension "libguile-dbi" "init_dbi")
<br><br>; exported primitives
<br><br>(export dbi-open
<br> dbi-close
<br> dbi-query
<br> dbi-get_row
<br> dbi-get_status)
<br><br><br><br>--linas
<br><br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Guile---Dev-f1686.html" embed="fixTarget[1686]" target="_top" >Gnu - Guile - Dev</a></p>tag:old.nabble.com,2006:post-26668848Re: UTF-8 regression in guile 1.9.52009-12-06T12:43:46Z2009-12-06T12:43:46ZLinas Vepstas-32009/12/6 Mike Gran <<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26668848&i=0" target="_top" rel="nofollow">spk121@...</a>>:
<br>>
<br>>> > need to call (setlocale LC_ALL "")
<br>>
<br>> But for Guile to store characters as codepoints, declaring a locale
<br>> pretty much a requirement now.
<br><br>Would it make sense to add (setlocale LC_ALL "") to some default,
<br>e.g. boot-9.scm ?
<br><br>--linas
<br><br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Guile---Dev-f1686.html" embed="fixTarget[1686]" target="_top" >Gnu - Guile - Dev</a></p>tag:old.nabble.com,2006:post-26668811Re: UTF-8 regression in guile 1.9.52009-12-06T12:40:29Z2009-12-06T12:40:29ZMike Gran<br><div class='shrinkable-quote'><br>> > Hmm. The "ã" is a dead giveaway that you are printing a UTF-8 string
<br>> > that is being interpreted as a ISO-8859-1 string.
<br>> >
<br>> > You've already said that you're in a UTF-8 locale. It could be that you
<br>> > need to call (setlocale LC_ALL "")
<br>>
<br>> That cured it.
<br>>
<br>> > as well as having a setlocale call in your program.
<br>>
<br>> Doesn't seem to be required, after the above.
<br>>
<br>> Thanks!
<br>>
<br>> Why this happened is strange; I'm now investigating. Sorry to
<br>> have bothered you with something that is dohh .. basic.
</div><br>1.9.x does work fundamentally differently w.r.t. strings.
<br>The reason for that is because of how strings are now stored.
<br>In 1.8.x, a character was a byte. In 1.9.x a character is a
<br>codepoint.
<br><br>But for Guile to store characters as codepoints, declaring a locale
<br>pretty much a requirement now.
<br><br>-Mike
<br><br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Guile---Dev-f1686.html" embed="fixTarget[1686]" target="_top" >Gnu - Guile - Dev</a></p>tag:old.nabble.com,2006:post-26668127Re: UTF-8 regression in guile 1.9.52009-12-06T11:33:25Z2009-12-06T11:33:25ZLinas Vepstas-32009/12/6 Mike Gran <<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26668127&i=0" target="_top" rel="nofollow">spk121@...</a>>:
<div class='shrinkable-quote'><br>>> From: Linas Vepstas <<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26668127&i=1" target="_top" rel="nofollow">linasvepstas@...</a>>
<br>>
<br>>
<br>>> Then, from the guile prompt, I can evaluate the following:
<br>>>
<br>>> (new-node "てみました。")
<br>>>
<br>>> and get the output "The name is てみました。"
<br>>>
<br>>>
<br>>> However, in guile-1.9.5, the above gives me:
<br>>>
<br>>> "The name is ã¦ã¿ã¾ããã"
<br>>
<br>> Hmm. The "ã" is a dead giveaway that you are printing a UTF-8 string
<br>> that is being interpreted as a ISO-8859-1 string.
<br>>
<br>> You've already said that you're in a UTF-8 locale. It could be that you
<br>> need to call (setlocale LC_ALL "")
</div><br>That cured it.
<br><br>> as well as having a setlocale call in your program.
<br><br>Doesn't seem to be required, after the above.
<br><br>Thanks!
<br><br>Why this happened is strange; I'm now investigating. Sorry to
<br>have bothered you with something that is dohh .. basic.
<br><br>--linas
<br><br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Guile---Dev-f1686.html" embed="fixTarget[1686]" target="_top" >Gnu - Guile - Dev</a></p>tag:old.nabble.com,2006:post-26667971Re: UTF-8 regression in guile 1.9.52009-12-06T11:16:51Z2009-12-06T11:16:51ZMike Gran> From: Linas Vepstas <<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26667971&i=0" target="_top" rel="nofollow">linasvepstas@...</a>>
<br><br><div class='shrinkable-quote'><br>> Then, from the guile prompt, I can evaluate the following:
<br>>
<br>> (new-node "てみました。")
<br>>
<br>> and get the output "The name is てみました。"
<br>>
<br>>
<br>> However, in guile-1.9.5, the above gives me:
<br>>
<br>> "The name is ã¦ã¿ã¾ããã"
</div><br>Hmm. The "ã" is a dead giveaway that you are printing a UTF-8 string
<br>that is being interpreted as a ISO-8859-1 string.
<br><br>You've already said that you're in a UTF-8 locale. It could be that you
<br>need to call (setlocale LC_ALL "") from the command line before entering
<br>(new-node "てみました。") as well as having a setlocale call in your program.
<br><br>Thanks,
<br><br>Mike Gran
<br><br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Guile---Dev-f1686.html" embed="fixTarget[1686]" target="_top" >Gnu - Guile - Dev</a></p>tag:old.nabble.com,2006:post-26667684UTF-8 regression in guile 1.9.52009-12-06T10:43:03Z2009-12-06T10:43:03ZLinas Vepstas-3Hi,
<br><br>I seem to see either a regression in guile-1.9.5 with regard
<br>to UTF-8 strings, or at least some sort of incompatible change.
<br><br>In guile-1.8.6, I am able to do the following:
<br><br>SCM new_node (SCM sname)
<br>{
<br> char * cname = scm_to_locale_string(sname);
<br> printf ("The name is %s\n", cname);
<br> free (cname);
<br> return SCM_EOL;
<br>}
<br><br>scm_c_define_gsubr("new-node", 1, 0, 0, ss_name);
<br><br>Then, from the guile prompt, I can evaluate the following:
<br><br> (new-node "てみました。")
<br><br>and get the output "The name is てみました。"
<br><br><br>However, in guile-1.9.5, the above gives me:
<br><br> "The name is ã¦ã¿ã¾ããã"
<br><br>Now, it is very possible that I've forgotten to say
<br><br> (use-modules some-new-utf8-module)
<br><br>but I am unclear on what that module is (and why its not
<br>specified by default).
<br><br>In both cases, my shell has: LANG=en_US.UTF-8
<br><br>--linas
<br><br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Guile---Dev-f1686.html" embed="fixTarget[1686]" target="_top" >Gnu - Guile - Dev</a></p>tag:old.nabble.com,2006:post-26668901Another posting to remove2009-12-06T09:53:10Z2009-12-06T09:53:10ZWendy Ann Tarzian<html><head></head><body><div style='font-family: Arial; font-size: 12pt; color: #000000'><DIV style="FONT-SIZE: 12pt; COLOR: #000000; FONT-FAMILY: Arial">Please remove this posting from internet access.</DIV>
<DIV style="FONT-SIZE: 12pt; COLOR: #000000; FONT-FAMILY: Arial"> </DIV>
<DIV style="FONT-SIZE: 12pt; COLOR: #000000; FONT-FAMILY: Arial"><BR><A class=l href="http://www.archivum.info/guile-user@gnu.org/2005-04/threads.html" target="_top" rel="nofollow"><FONT color=#2200cc>guile-user@... (thread)</FONT></A> <A id=XPLSS_1040553995 href="#" target="_top" rel="nofollow"><IMG height=20 src="linkscanner://safe.gif/" width=20 border=0></A></DIV>
<DIV class=s style="FONT-SIZE: 12pt; COLOR: #000000; FONT-FAMILY: Arial"><B>...</B> People identified -Seeking assistance regarding Bill of Rights and National Security Breaches by Illinois Leaders and others, <STRONG>Wendy Tarzian</STRONG>, 2005/04/16 <B>...</B><BR><CITE><FONT color=#008000><a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26668901&i=0" target="_top" rel="nofollow">www.archivum.info/guile-user@...</a>/2005-04/threads.html</FONT></CITE></DIV></div></body></html><p>From forum: <a href="http://old.nabble.com/Gnu---Guile---User-f1687.html" embed="fixTarget[1687]" target="_top" >Gnu - Guile - User</a></p>tag:old.nabble.com,2006:post-26668899Request to remove postings2009-12-06T09:16:27Z2009-12-06T09:16:27ZWendy Ann Tarzian<html><head></head><body><div style='font-family: Arial; font-size: 12pt; color: #000000'><BR>
<P>Hello:</P>
<P> </P>
<P>I'm trying to find a job and can't because of this posting. I'm sending you this email to request that you remove from the internet the following posting: </P>
<P> </P>
<P><A class=l href="http://www.mail-archive.com/guile-user@gnu.org/msg00119.html" target="_top" rel="nofollow"><FONT color=#2200cc>Scientists and Chicago Police Involved -Krok, Gunn and Kennedy at <B>...</B></FONT></A> <A id=XPLSS_1730945853 href="#" target="_top" rel="nofollow"><IMG height=20 src="linkscanner://safe.gif/" width=20 border=0></A></P>
<DIV class=s><STRONG>Wendy Tarzian</STRONG> Sat, 16 Apr 2005 15:59:38 -0700. This is important to corporations that value their databases - data on clients, etc. <B>...</B><BR><CITE><FONT color=#008000><a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26668899&i=0" target="_top" rel="nofollow">www.mail-archive.com/guile-user@...</a>/msg00119.html - </FONT></CITE><SPAN class=gl><A href="http://74.125.95.132/search?q=cache:IGDzzd6P0dIJ:www.mail-archive.com/guile-user%40gnu.org/msg00119.html+%22wendy+tarzian%22&cd=11&hl=en&ct=clnk&gl=us" target="_top" rel="nofollow"><FONT color=#551a8b>Cached</FONT></A><FONT color=#7777cc> - </FONT><A href="http://www.google.com/search?hl=en&lr=&q=related:www.mail-archive.com/guile-user%40gnu.org/msg00119.html+%22wendy+tarzian%22&sa=X&ei=VOQbS8HLMdH_nAer0pXUAw&ved=0CCUQHzAK" target="_top" rel="nofollow"><FONT color=#7777cc>Similar</FONT></A></SPAN></DIV></div></body></html><p>From forum: <a href="http://old.nabble.com/Gnu---Guile---User-f1687.html" embed="fixTarget[1687]" target="_top" >Gnu - Guile - User</a></p>tag:old.nabble.com,2006:post-26661710guile-aspell : spellchecking wth Guile2009-12-05T18:40:00Z2009-12-05T18:40:00ZMike GranHi-
<br><br>I wrote a bit of glue code for Guile and Gnu Aspell, a spellchecking library.
<br>I packaged it as a Guile library and saved it at
<br><a href="http://www.lonelycactus.com/aspell.html" target="_top" rel="nofollow">http://www.lonelycactus.com/aspell.html</a><br><br>I've only tried to build it on Guile 1.8.x on i686-pc-linux-gnu and
<br>i686-pc-cygwin.
<br>
<br>For the small amount of effort, it actually came out pretty useful.
<br>Here it is in action:
<br><br>-----------
<br><br>$ LTDL_LIBRARY_PATH=/usr/local/lib guile -L /usr/local/share/guile/site
<br><br>guile> (use-modules (aspell aspell))
<br><br>guile> (aspell-set-lang "en_US")
<br>#t
<br><br>guile> (aspell-correct? "recieve") ; check if it is a word
<br>#f
<br><br>guile> (aspell-word "recieve") ; get some possibilities
<br>("receive" "receiver" "Recife" "relieve" "received" "receives" "revive" "reeve"
<br>"reserve" "deceive" "reliever" "recede" "recipe" "recite" "relive")
<br><br>guile> (aspell-add-word "recieve") ; add the word to the dictionary
<br>#t
<br><br>guile> (aspell-word "flem")
<br>("Flem" "fl em" "fl-em" "Lem" "fem" "flam" "flea" "flee" "flew" "Clem" "elem" "f
<br>led" "Flem's")
<br><br>guile> (aspell-store-replacement "flem" "phlegm") ; add a preferred replacement
<br>#t
<br><br>guile> (aspell-word "flem")
<br>("Flem" "phlegm" "fl em" "fl-em" "Lem" "fem" "flam" "flea" "flee" "flew" "Clem"
<br>"elem" "fled" "Flem's")
<br><br>-----------
<br><br>Thanks,
<br><br>Mike Gran
<br><br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Guile---User-f1687.html" embed="fixTarget[1687]" target="_top" >Gnu - Guile - User</a></p>tag:old.nabble.com,2006:post-26634751Re: racing srfi-18 threads2009-12-03T14:52:44Z2009-12-03T14:52:44ZNeil JerramNeil Jerram <<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26634751&i=0" target="_top" rel="nofollow">neil@...</a>> writes:
<br><br>> [...] wondering whether and how we
<br>> will provide similar debugging facilities for compiled code as we have
<br>> in 1.8.x for interpreted code.
<br><br>Some more thoughts here, to try to build a complete new picture.
<br><br>Things that we can do quite nicely in 1.8.x are:
<br>- stack inspection
<br> - seeing the frames and what is happening, for context
<br> - mapping back to source code
<br> - querying variable values
<br> - evaluating an expression in a frame's local environment
<br>- breakpoints
<br>- tracing
<br>- single stepping.
<br><br>We should try to include profiling and code coverage in the picture.
<br><br>My thoughts for doing this in 2.0 are as follows.
<br><br>- Single stepping, breakpoints, code coverage and tracing can all be
<br> done by instrumenting code - either in Scheme, or with lower-level
<br> language ops that somehow get inserted as code is compiled.
<br><br>- The more I think about this, the more it seems clearly preferable to
<br> the 1.8.x evaluator traps model - i.e. where there are ways of
<br> _marking_ code in some way, and the evaluator or VM calls out to a
<br> hook when it sees one of these marks. It's a simpler solution, and...
<br><br>- It in particular removes the need for most of the complexity that we
<br> have in 1.8.x's (ice-9 debugging traps). The complexity is mostly
<br> about wanting to say 'do THIS when you start executing procedure FOO',
<br> but the lowlevel traps interface not allowing us to specify either
<br> 'THIS' or 'when you start executing procedure FOO' precisely. (All we
<br> can say is 'call a globally-defined function when you start executing
<br> any of the currently marked procedures'.)
<br><br>- I'm not clear how this interacts with optimisation... What
<br> happens when an optimisation reorders or eliminates code with a
<br> breakpoint? How do we present this to the user? It feels like a
<br> soluble problem though.
<br><br>- Instrumentation-based single-stepping would be more like edebug than
<br> what we have in 1.8.x - i.e. the mode of operation would probably be
<br> to instrument the whole of the body of a given function. But I think
<br> that would be fine (and consistent for our plan for future emacs
<br> domination :-)). (In contrast, 1.8.x single-stepping doesn't require
<br> prior instrumentation, and allows stepping over function boundaries.)
<br><br>- An unfortunate consequence of psyntax is backtraces being harder to
<br> read, and to correlate back to source code, because of all the
<br> alpha-renamed variables. When paused at a breakpoint, I think this
<br> also makes it harder for the user to ask what the value of a given
<br> variable is. Is there anything we can do about this - such as mapping
<br> all the variable names back to their names in the original source
<br> code?
<br><br>- As one of Andy's eval commits says, we don't have local-eval any more
<br> and so can't currently do "evaluate in a stack frame". I suppose the
<br> most important case here is querying local variable values; is there a
<br> reasonable solution for that?
<br><br>Any thoughts on that?
<br><br>> (SLIB has stuff like this too. I wonder if it would just work.)
<br><br>(Currently blocked by SLIB not loading at all in 1.9/2.0:
<br><br>scheme@(guile-user)> (use-modules (ice-9 slib))
<br>;;; note: autocompilation is enabled, set GUILE_AUTO_COMPILE=0
<br>;;; or pass the --no-autocompile argument to disable.
<br>;;; compiling /usr/share/slib/guile.init
<br>;;; WARNING: compilation of /usr/share/slib/guile.init failed:
<br>;;; key syntax-error, throw args (sc-expand "~a in ~a" ("unexpected syntax" define) #f)
<br>ERROR: In procedure sc-expand:
<br>ERROR: unexpected syntax in define
<br><br>)
<br><br>Regards,
<br> Neil
<br><br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Guile---Dev-f1686.html" embed="fixTarget[1686]" target="_top" >Gnu - Guile - Dev</a></p>tag:old.nabble.com,2006:post-26633244Re: GOOPS/compiler circular dependency2009-12-03T13:09:15Z2009-12-03T13:09:15ZNeil Jerram<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26633244&i=0" target="_top" rel="nofollow">ludo@...</a> (Ludovic Courtès) writes:
<br><br>[...]
<br>> ERROR: In procedure scm-error:
<br>> ERROR: no such language glil
<br><br>Here's a similar looking case from me.
<br><br>neil@arudy:~/SW/Guile/git$ meta/uninstalled-env guile --debug ../test-nj
<br>Backtrace:
<br>In ice-9/boot-9.scm:
<br>2368: 19 [try-module-autoload (language glil)]
<br>2371: 18 [#<program 934e760 at ice-9/boot-9.scm:2369:12 ()>]
<br>1776: 17 [save-module-excursion #<program 934e6d0 at ice-9/boot-9.scm:2373:19 ()>]
<br>2375: 16 [#<program 934e6d0 at ice-9/boot-9.scm:2373:19 ()>]
<br>In unknown file:
<br> ?: 15 [#<program 934e5c0 ()>]
<br>In ice-9/boot-9.scm:
<br>2205: 14 [process-define-module ((language glil) #:use-module (#) #:use-module ...)]
<br>1481: 13 [call-with-deferred-observers #<program 93596e0 at ice-9/boot-9.scm:2205:11 ()>]
<br>2208: 12 [#<program 93596e0 at ice-9/boot-9.scm:2205:11 ()>]
<br>3053: 11 [module-export! # #]
<br>3054: 10 [#<program 9359680 at ice-9/boot-9.scm:3052:14 (name)> glil-kw-prelude-rest]
<br>1702: 9 [module-ensure-local-variable! #<directory # 92501f8> glil-kw-prelude-rest]
<br>1474: 8 [module-modified #<directory (language glil) 92501f8>]
<br>In oop/goops/dispatch.scm:
<br> 200: 7 [#<program 9340690 at oop/goops/dispatch.scm:195:4 args> # #]
<br> 181: 6 [comp # #]
<br>In system/base/compile.scm:
<br> 239: 5 [compile (lambda (p45 c44 p43 ...) (case-lambda (# #) (args #))) #:key ...]
<br> 178: 4 [compile-passes scheme value ()]
<br>In system/base/language.scm:
<br> 87: 3 [lookup-compilation-order scheme value]
<br>In ice-9/boot-9.scm:
<br> 129: 2 [or-map #<program 935ee00 at system/base/language.scm:79:21 (pair)> (#)]
<br>In system/base/language.scm:
<br> 77: 1 [lp tree-il ((# . #<program compile-tree-il (x e opts)>))]
<br>In ice-9/boot-9.scm:
<br> 692: 0 [error "no such language" tree-il]
<br><br>ERROR: no such language tree-il
<br>neil@arudy:~/SW/Guile/git$
<br><br>This happened during investigation of the
<br>no-meta-in-compiled-define-syntax-thunk problem. I wondered if the
<br>problem would persist if I could persuade psyntax to emit scheme instead
<br>of tree-il when compiling - the hypothesis being that the bug is
<br>somewhere in one of the `(case (fluid-ref *mode*) ((c) ...' fragments of
<br>code in psyntax.scm.
<br><br>So in language/scheme/compile-tree-il.scm I changed (sc-expand x 'c ...)
<br>to (sc-expand x 'd ...), and since then have been stuck with "no such
<br>language tree-il".
<br><br> Neil
<br><br><br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Guile---Dev-f1686.html" embed="fixTarget[1686]" target="_top" >Gnu - Guile - Dev</a></p>tag:old.nabble.com,2006:post-26631394A big hole where the evaluator used to be...2009-12-03T11:08:05Z2009-12-03T11:08:05ZNeil JerramExciting times! Much kudos to Andy, who has just merged a big and
<br>interesting change, the effect of which is essentially to stop using the
<br>old C-coded evaluator.
<br><br>(Actually it's not quite true that there's only a hole left. We still
<br>need a basic C evaluator for bootstrapping, but it's much smaller and
<br>simpler than what was there before.)
<br><br>FWIW, building and make check all look good to me. There are some
<br>noticeable benchmark changes - some slower and some faster - which I'll
<br>look more into.
<br><br>Regards,
<br> Neil
<br><br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Guile---Dev-f1686.html" embed="fixTarget[1686]" target="_top" >Gnu - Guile - Dev</a></p>tag:old.nabble.com,2006:post-26631387Re: A variable that holds a string which may be the name of a variable.2009-12-03T11:00:48Z2009-12-03T11:00:48ZRichardShannOn Thu, 2009-12-03 at 18:56 +0000, Neil Jerram wrote:
<div class='shrinkable-quote'><br>> Linas Vepstas <<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26631387&i=0" target="_top" rel="nofollow">linasvepstas@...</a>> writes:
<br>>
<br>> > 2009/12/2 Neil Jerram <<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26631387&i=1" target="_top" rel="nofollow">neil@...</a>>:
<br>> > Neil,
<br>> >
<br>> > For my own edification -- is tehre any difference between
<br>> >
<br>> >> (let ((sym (with-input-from-string mything read)))
<br>> >
<br>> > and
<br>> >
<br>> > (let ((sym (string->symbol mything)))
<br>>
<br>> In the case where mything is a string consisting of symbol syntax
<br>> characters, no.
</div><br>In fact, I was somewhat interested that it should be fairly bomb proof
<br>to the contents of mything. So I tried Neil's code on various dangerous
<br>looking things...
<br>I hadn't spotted string->symbol, else I might have got there unaided; I
<br>had a hazy idea that that was what I was trying to do. Gradually the
<br>mists are clearing - thank you.
<br><br>Richard
<br><br><br><br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Guile---User-f1687.html" embed="fixTarget[1687]" target="_top" >Gnu - Guile - User</a></p>tag:old.nabble.com,2006:post-26631202Re: A variable that holds a string which may be the name of a variable.2009-12-03T10:56:18Z2009-12-03T10:56:18ZNeil JerramLinas Vepstas <<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26631202&i=0" target="_top" rel="nofollow">linasvepstas@...</a>> writes:
<br><div class='shrinkable-quote'><br>> 2009/12/2 Neil Jerram <<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26631202&i=1" target="_top" rel="nofollow">neil@...</a>>:
<br>> Neil,
<br>>
<br>> For my own edification -- is tehre any difference between
<br>>
<br>>> (let ((sym (with-input-from-string mything read)))
<br>>
<br>> and
<br>>
<br>> (let ((sym (string->symbol mything)))
</div><br>In the case where mything is a string consisting of symbol syntax
<br>characters, no.
<br><br>Maybe you're then suggesting that Richard could use string->symbol
<br>instead - in which case I agree. Using with-input-from-string and read
<br>is a bit of overkill; I forgot about string->symbol.
<br><br>Regards,
<br> Neil
<br><br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Guile---User-f1687.html" embed="fixTarget[1687]" target="_top" >Gnu - Guile - User</a></p>tag:old.nabble.com,2006:post-26631154Re: A variable that holds a string which may be the name of a variable.2009-12-03T10:52:44Z2009-12-03T10:52:44ZNeil JerramRichard Shann <<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26631154&i=0" target="_top" rel="nofollow">richard.shann@...</a>> writes:
<br><br>> Thanks to all guileans who have responded. I seem to have a workable
<br>> solution based on Neil's code. The thing I hadn't grokked was the
<br>> existence of the with-input-from-string procedure. I had been messing
<br>> with defined? and symbol? but got nowhere.
<br><br>I would say the magic is more in `read' than in
<br>`with-input-from-string'.
<br><br>`read' is responsible for the conversion from
<br>something-that-looks-like-code to actual code - which in your case is a
<br>single symbol. `with-input-from-string' just tells `read' that the
<br>something-that-looks-like-code is a string.
<br><br>Regards,
<br> Neil
<br><br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Guile---User-f1687.html" embed="fixTarget[1687]" target="_top" >Gnu - Guile - User</a></p>tag:old.nabble.com,2006:post-26630146Re: A variable that holds a string which may be the name of a variable.2009-12-03T09:23:47Z2009-12-03T09:23:47ZLinas Vepstas-32009/12/2 Neil Jerram <<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26630146&i=0" target="_top" rel="nofollow">neil@...</a>>:
<br>Neil,
<br><br>For my own edification -- is tehre any difference between
<br><br>> (let ((sym (with-input-from-string mything read)))
<br><br>and
<br><br>(let ((sym (string->symbol mything)))
<br><br>?
<br><br>--linas
<br><br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Guile---User-f1687.html" embed="fixTarget[1687]" target="_top" >Gnu - Guile - User</a></p>tag:old.nabble.com,2006:post-26629988Re: A variable that holds a string which may be the name of a variable.2009-12-03T08:11:25Z2009-12-03T08:11:25ZRichardShannThanks to all guileans who have responded. I seem to have a workable
<br>solution based on Neil's code. The thing I hadn't grokked was the
<br>existence of the with-input-from-string procedure. I had been messing
<br>with defined? and symbol? but got nowhere.
<br><br>FWIW the core bit of code I ended up with takes the name of a command,
<br><name> and looks for a string in a variable called Help-d-<name> if it
<br>exists:
<br><br>(set! help (string-append "Help-d-" name))
<br> (let ((sym (with-input-from-string help read)))
<br> (if (defined? sym)
<br> (set! help (eval sym (current-module)))
<br> (set! help "No help")
<br> ))
<br><br>This seems to cover all the cases. I haven't really got to the bottom of
<br>what that with-input-from-string procedure does - I got as far as the
<br>documentation for read which mentions an s-expression. Elsewhere I have
<br>seen references to "forms", perhaps there is something establishing some
<br>basic terminology that I have missed...
<br><br>But thanks a lot - I am on my way again...
<br><br>Richard
<br><br><br><br><br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Guile---User-f1687.html" embed="fixTarget[1687]" target="_top" >Gnu - Guile - User</a></p>tag:old.nabble.com,2006:post-26618757Re: A variable that holds a string which may be the name of a variable.2009-12-02T14:37:37Z2009-12-02T14:37:37ZLinas Vepstas-32009/12/2 Richard Shann <<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26618757&i=0" target="_top" rel="nofollow">richard.shann@...</a>>:
<div class='shrinkable-quote'><br>> I am stuck on one of those symbol/variable-name-in-a-string things
<br>> again:
<br>>
<br>> (define mything "display")
<br>> (display (eval-string mything))
<br>>
<br>> that's fine. But can I test that the string in mything is the name of a
<br>> variable before doing the eval-string and finding out the hard way? I've
<br>> been doing (symbol? mything) etc, and going witless. Do I have to do all
<br>> that catch stuff?
</div><br>If I understand you correctly, no. That's because the
<br>thing going into eval-string can be arbitrarily complex --
<br>e.g. (eval-string "(define foo (lambda (x) (+ x y))) (define (bar
<br>z y) (foo (z)) (bar 2 2)") and there's no particular way to know
<br>if evaluating that won't throw some error. -- really, you have to
<br>evaluate it to find out.
<br><br>If you want to find out if some particular variable is bound
<br>to a proceedure, you can try (proceedure? thing) see e.g.
<br><a href="http://www.gnu.org/software/guile/manual/html_node/Procedure-Properties.html" target="_top" rel="nofollow">http://www.gnu.org/software/guile/manual/html_node/Procedure-Properties.html</a><br><br>If you want to know if a variable is bound, try defined? e.g.
<br><br>guile> (defined? 'x)
<br>#f
<br>guile> (defined? (string->symbol "x"))
<br>#f
<br>guile> (define x 42)
<br>guile> (defined? 'x)
<br>#t
<br>guile> (defined? (string->symbol "x"))
<br>#t
<br><br>--linas
<br><br><br>--linas
<br><br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Guile---User-f1687.html" embed="fixTarget[1687]" target="_top" >Gnu - Guile - User</a></p>