Bill Schottstaedt | 13 Jul 19:19 2010
Picon

']' and infinite loop

In Guile 1.9.10, typing

'(1 ']x)

or lots of similar cases involving ']', to the repl
gets into an infinite loop.

Faenza, Michael E | 13 Jul 15:54 2010
Picon

Guile 1.9.10 Build Questions

Hello,

 

I’ve attempted to build Guile 1.9.10 on a 32-bit installation of Ubuntu 9.10.  I cannot seem to successfully build using a call to ‘make’.  I’ve worked through what I believe to be my full set of configuration issues.

 

Two items still trouble me:

 

1)      Configuration of Guile checks how to link with Readline.  In my case, configuration correctly determines that libreadline is available.  This happens to be a Readline v.6.1 installation.  However, I get a configure warning stating that Readline is too old and that I need v.2.1 or better.  I wonder if the checks that Guile is performing are appropriate for determining the suitability of the installed Readline library.

 

2)      ‘make’ fails due to what primarily seems to be many undefined references to GC_ functions.  ffi_ types aren’t also coming up with undefined references.  I believe I’ve set up appropriate environment variables for libffi and libgc.

 

Has anyone experienced either of these behaviors recently?  I realize I’m not including any output for you to analyze; I didn’t want you to have to pour through my scripts or output.  I’m just hoping someone sees this and realizes I’ve made a common mistake and can point that out to me.

 

Thank you,

Michael Faenza

 

 

Andy Wingo | 13 Jul 20:56 2010
Picon

Re: Guile 1.9.10 Build Questions

Hi Michael,

On Tue 13 Jul 2010 15:54, "Faenza, Michael E" <mfaenza <at> mitre.org> writes:

> I’ve attempted to build Guile 1.9.10 on a 32-bit installation of
> Ubuntu 9.10.

Please include the commands you ran (with the environment variables),
and your config.log. You can gzip the config.log if you think it is too
big.

Thanks,

Andy
--

-- 
http://wingolog.org/

Andy Wingo | 13 Jul 21:55 2010
Picon

Re: ']' and infinite loop

On Tue 13 Jul 2010 19:19, "Bill Schottstaedt" <bil <at> ccrma.Stanford.EDU> writes:

> In Guile 1.9.10, typing
>
> '(1 ']x)
>
> or lots of similar cases involving ']', to the repl
> gets into an infinite loop.

Indeed '(] causes an infinite loop. Fixed in git.

Thanks,

Andy
--

-- 
http://wingolog.org/

Andy Wingo | 17 Jul 12:59 2010
Picon

Re: Guile 1.9: bug in load

Hi Luca,

Can you file a bug for this?

  https://savannah.gnu.org/bugs/?group=guile

I didn't see an obvious solution at the start, and so I forgot about it;
but we should have an answer for it. Collecting the info in a place
other than my inbox would be a great start :)

Thanks!

Andy

On Fri 26 Mar 2010 14:02, Luca Saiu <positron <at> gnu.org> writes:

> Hello again.
>
> Andy, you had asked me to ping you if the problem was still there after
> some time. I still see it in 1.9.9.
>
> Thanks to you and the whole team for your great work,
>
> Andy Wingo wrote:
>> On Wed 18 Nov 2009 11:05, Luca Saiu <positron <at> gnu.org> writes:
>> 
>>> Andy Wingo wrote:
>>>> On Fri 13 Nov 2009 21:24, Luca Saiu <positron <at> gnu.org> writes:
>>>>
>>>>> load should call canonicalize-path relative to the directory of the file
>>>>> which loads the other one, not relative to `pwd`.
>>>> Has this changed since 1.8?
>>> Yes, it has changed.
>> 
>> OK, then we have to fix it then :)
>> 
>>> If you need a distilled test case, please tell me; it should be quite
>>> easy to build.
>> 
>> No, it's a fairly easy thing to reproduce. Please poke us if we haven't
>> fixed it within a couple weeks.
>> 
>> Happy hacking,
>> 
>> Andy
>
>
> --
> Luca Saiu
> http://www-lipn.univ-paris13.fr/~saiu
> GNU epsilon: http://www.gnu.org/software/epsilon
> Marionnet:   http://www.marionnet.org

--

-- 
http://wingolog.org/

Mike Gran | 17 Jul 13:22 2010
Picon

Re: [r6rs] #\x0 is a perfectly valid character datum representation

> >> r6rs hex  *character* escapes and Guile 1.8.x octal

> >> character escapes could  logically coexist without confusion.
> 
> > Let's do that. Shall you, or  shall I?

I pushed this change into git.

Thanks,

Mike

Mike Gran | 17 Jul 13:25 2010
Picon

[bug #29643] Current locale has higher precedence than `coding:' when determining file port encoding [1.9.10]


Update of bug #29643 (project guile):

                  Status:          Ready For Test => Fixed                  
             Open/Closed:                    Open => Closed                 

    _______________________________________________________

Follow-up Comment #2:

I pushed a branch that fixed the problem

e50d921bd87e53df33c04831ecf8b3d100af2b7b

    _______________________________________________________

Reply to this item at:

  <http://savannah.gnu.org/bugs/?29643>

_______________________________________________
  Message sent via/by Savannah
  http://savannah.gnu.org/

Andy Wingo | 17 Jul 13:38 2010
Picon

Re: [r6rs] #\x0 is a perfectly valid character datum representation

Hi,

On Sat 17 Jul 2010 13:22, Mike Gran <spk121 <at> yahoo.com> writes:

>> >> r6rs hex  *character* escapes and Guile 1.8.x octal
>
>> >> character escapes could  logically coexist without confusion.
>> 
>> > Let's do that. Shall you, or  shall I?
>
> I pushed this change into git.

Thanks a lot!

Andy
--

-- 
http://wingolog.org/

Patrick McCarty | 18 Jul 02:02 2010
Picon

[bug #30469] `procedure-source' is broken in Guile 1.9.11


URL:
  <http://savannah.gnu.org/bugs/?30469>

                 Summary: `procedure-source' is broken in Guile 1.9.11
                 Project: Guile
            Submitted by: pnorcks
            Submitted on: Sat 17 Jul 2010 05:02:52 PM PDT
                Category: None
                Severity: 3 - Normal
              Item Group: None
                  Status: None
                 Privacy: Public
             Assigned to: None
             Open/Closed: Open
         Discussion Lock: Any

    _______________________________________________________

Details:

Hello,

I am testing the latest git, and I think I've found a regression (compared to
1.8.7).  For a valid procedure definition, Guile returns `#f' for
`procedure-source':

  scheme <at> (guile-user)> (define (test a) (+ a 1))
  scheme <at> (guile-user)> (procedure-source test)
  $1 = #f
  scheme <at> (guile-user)>

The same procedure in Guile 1.8.7 returns valid info:

  guile> (define (test a) (+ a 1))
  guile> (procedure-source test)
  (lambda (a) (+ a 1))
  guile>

    _______________________________________________________

Reply to this item at:

  <http://savannah.gnu.org/bugs/?30469>

_______________________________________________
  Message sent via/by Savannah
  http://savannah.gnu.org/

Luca Saiu | 19 Jul 00:51 2010
Picon

[bug #30480] VM: load looks for files in the wrong directory


URL:
  <http://savannah.gnu.org/bugs/?30480>

                 Summary: VM: load looks for files in the wrong directory
                 Project: Guile
            Submitted by: positrone
            Submitted on: Mon 19 Jul 2010 12:51:53 AM CEST
                Category: None
                Severity: 3 - Normal
              Item Group: None
                  Status: None
                 Privacy: Public
             Assigned to: None
             Open/Closed: Open
         Discussion Lock: Any

    _______________________________________________________

Details:

The behavior of the load function in bytecode programs is wrong and different
from interpreted programs.

* The interpreter works just like 1.8 did: loaded files are looked for in
paths relative to the directory of the script calling load.  This is the
reasonable and useful behavior.
* The virtual machine looks for the file to load in paths relative to the cwd
(I don't know if at startup time or at load time).
Absolute pathnames work in either case.

Ludovic said this when we spoke about that in Göteborg:
load should call canonicalize-path relative to the directory of the file
which loads the other one, not relative to `pwd`.

Of course this is a regression compared to 1.8.

How to reproduce the malfunction, in a particularly simple case (no explicit
relative paths in the pathnames given to load):
$ cd /tmp; mkdir q; echo '(display "b\n")' > q/b.scm; echo '(load
"b.scm")(display "a\n")' > q/a.scm
$ rm -rf ~/.cache
$ guile --no-autocompile q/a.scm
b
a
$ guile --autocompile q/a.scm
;;; note: autocompilation is enabled, set GUILE_AUTO_COMPILE=0
;;;       or pass the --no-autocompile argument to disable.
;;; compiling q/a.scm
;;; compiled /home/luca/.cache/guile/ccache/2.0-0.Q-LE-8/tmp/q/a.scm.go
ERROR: In procedure open-file:
ERROR: No such file or directory: "b.scm"

    _______________________________________________________

Reply to this item at:

  <http://savannah.gnu.org/bugs/?30480>

_______________________________________________
  Message sent via/by Savannah
  http://savannah.gnu.org/


Gmane