Re: pipe buffering
Neil Jerram ossau.uklinux.net> writes:
> Well, as the manual says:
>
> -- Macro: expect clause ...
> [...] The procedures
> are called in turn after each character is read from the port,
> [...]
>
> The test is successful if the procedure returns a non-false value.
>
> Your lambda returns *unspecified*, which counts as a non-false value.
> So the whole (expect ...) invocation completes after reading just one
> character.
I must've missed that "each character" bit.
Maybe it would be better to post my original program that would just
hang. I was only using expect to print debugging messages; my
original program used expect-strings.
--Aidan
#! /usr/bin/guile -s
!#
(use-modules (ice-9 expect))
(use-modules (ice-9 popen))
(define ie-io (open-pipe* OPEN_BOTH
"/usr/bin/telnet"
"ienabler.canterbury.ac.nz" "259"))
(let ((expect-port ie-io))
(expect-strings
("^User:" (begin (display "[snip]\n" ie-io) (display "gave user")))
("^passwd:" (begin (display "[snip]\n" ie-io) (display "gave passwd")
Unbound variable: quote
Hello,
I'm trying to load a file at the REPL that loads some modules from a
subdirectory then defines a function. At reading my use-modules
statements, I get the following error:
guile> %load-path
("/home/joel/Public/src/age_dep/sim/trunk" "/usr/share/guile/site"
"/usr/share/guile/1.8" "/usr/share/guile")
guile> (load "age_dep.scm")
Backtrace:
In unknown file:
...
?: 20 (letrec ((load-file #)) (dynamic-wind (lambda () #) (lambda () #)
...) ...)
?: 21* [dynamic-wind # # #]
?: 22* [#]
?: 23* (let* ((file #)) (cond (# => #) (# => #)))
?: 24 [#
"/home/joel/Public/src/age_dep/sim/trunk/age-dep-scm/age-dep-interface.scm"]
?: 25 [with-fluid* # #f #]
?: 26* [#]
?: 27* [load-file # ...]
?: 28* [save-module-excursion #]
?: 29 (let (# #) (dynamic-wind # thunk #))
?: 30 [dynamic-wind # # #]
?: 31* [#]
?: 32* [primitive-load
"/home/joel/Public/src/age_dep/sim/trunk/age-dep-scm/age-dep-interface.scm"]
In /home/joel/Public/src/age_dep/sim/trunk/age-dep-scm/age-dep-interface.scm:
36: 33* (use-modules (ice-9 popen))
36: 34 (eval-case (# # *unspecified*) (else #))
36: 35 (begin (process-use-modules (list (list #))) *unspecified*)
In unknown file:
?: 36* [process-use-modules ...
?: 37* [list ...
?: 38* [list ...
?: 39* (quote (ice-9 popen))
: In expression (quote (ice-9 popen)):
: Unbound variable: quote
ABORT: (unbound-variable)
However, at the REPL I can load the popen module just fine (with a
warning):
guile> (use-modules (ice-9 popen))
guile> (use-modules (age-dep-scm age-dep-interface))
WARNING: (guile-user): `quote' imported from both (ice-9 null) and (age-dep-scm
age-dep-interface)
Why would `quote' be redefined? What's going on here and how can I fix it?
Thanks,
Joel
--
Joel J. Adamson
Servedio Lab
University of North Carolina at Chapel Hill
FSF Member #8164
http://www.unc.edu/~adamsonj
pgpnplCF1AGmq.pgp
Description: PGP signature
Re: pipe buffering
On Wed, Sep 22, 2010 at 09:17:08PM +0100, Neil Jerram wrote: > Well, as the manual says: > > -- Macro: expect clause ... > [...] The procedures > are called in turn after each character is read from the port, > [...] > > The test is successful if the procedure returns a non-false value. > > Your lambda returns *unspecified*, which counts as a non-false value. > So the whole (expect ...) invocation completes after reading just one > character. I must've missed that "each character" bit. Maybe it would be better to post my original program that would just hang. I was only using expect to print debugging messages; my original program used expect-strings. --Aidan signature.asc Description: Digital signature
