I ma pretty sure that 1.6 and 1.8 behave the same way when one loads a
.so when in a module context. It's just that 1.8 no longer has the code
that satisfies a require by loading a .so.
So the export statements that David posted from debian look fine to me,
and the least confusing way to do thing
Hi,
"Frank J. R. Hanstick" <[EMAIL PROTECTED]> writes:
> I am trying to build guile for MacOS 10.4.11 and run into the
> following error:
>
> /bin/sh ./libtool --tag=CC --mode=compile gcc -DHAVE_CONFIG_H -I. -
> I. -I.. -I./..-g -O2 -MT readline.lo -MD -MP -MF .deps/
> readline.Tpo
2008/5/28 Frank J. R. Hanstick <[EMAIL PROTECTED]>:
> Hello,
>I am trying to build guile for MacOS 10.4.11 and run into the
> following error:
>
> readline.c:109: error: 'rl_pending_input' undeclared (first use in this
> function)
Please see the "FAQ" file, which reads as appended below.
Hello,
I am trying to build guile for MacOS 10.4.11 and run into the
following error:
/bin/sh ./libtool --tag=CC --mode=compile gcc -DHAVE_CONFIG_H -I. -
I. -I.. -I./..-g -O2 -MT readline.lo -MD -MP -MF .deps/
readline.Tpo -c -o readline.lo readline.c
mkdir .libs
gcc -DHAVE_CONFIG_
⎛⎞ David Pirotte <[EMAIL PROTECTED]>
⎝⎠ Tue, 27 May 2008 13:17:18 -0300
Here is [attached] an exemple of the actual guile-pg 0.16-4
postgres.scm file from the latest debian package.
I remember that file, but unfortunately...
The exports are done after the load-extension, but I am convi
release notes:
UNIX-domain sockets now supported for both i and o,
plus other touchups. The #:inhibit-content! is necessary
to implement HTTP HEAD if using the big dishing loop.
NB the MARKED FOR REMOVAL item, see
(info "(guile-www) big-dishing-loop")
for details.
thi
README exce
