[Ecls-list] Latest changes, 1/10/2009

Gabriel Dos Reis gdr at integrable-solutions.net
Mon Oct 5 17:25:31 UTC 2009


On Mon, Oct 5, 2009 at 11:39 AM, Juan Jose Garcia-Ripoll
<juanjose.garciaripoll at googlemail.com> wrote:
> On Thu, Oct 1, 2009 at 5:39 PM, Gabriel Dos Reis
> <gdr at integrable-solutions.net> wrote:

>> Hi Juanjo,
>>
>>   I looked at the automated tests and it seems that noweb is not
>> installed on the
>> machine so OpenAxiom could not be built.  Is that a known issue?
>
> It seems there is no port of noweb/Fink for OS X 10.6 yet. That means
> OpenAxiom will probably fail to build unless I set it up manually.
>
> I wanted to ask you also about a different problem. The last reports
> of open-axiom fail because they look for X11 development libraries. Is
> this new?

Looking for X11 libraries is not new -- it has always been there.
However, if no X11 headers or libraries is found, the build should
still progress -- the only think that would happen is that the parts
that depend on X11 will not be built.  You'll still have a fully
functional interpreter and compiler.

> Or have I misconfigured something?

If you still have the config.log around, I would like to have a look at it
because this is alarming and would be a serious OpenAxiom regression.

>  BTW, I just installed the
> required libraries (xorg-dev on Ubuntu) so if this was the only
> problem everything should be ok by tomorrow
>
> ../../libtool --mode=link gcc -static -o
> ../../x86_64-unknown-linux/lib/htadd addfile.o htadd.o lex.o
> -L../../src/lib -lspad -L../../src/lib -lopen-axiom-core -lutil   -lm
> -lm -lXpm  -lX11  -lm
> libtool: link: gcc -o ../../x86_64-unknown-linux/lib/htadd addfile.o
> htadd.o lex.o  -L/home/jjgarcia/tmp/intel64.butthreads/lisp/oa-build/src/lib
> -lspad -lopen-axiom-core -lutil -lXpm -lX11 -lm
> /usr/bin/ld: cannot find -lXpm

Ah, that is a very thorny issue.   There seems to be no complete reliable
way to test for Xpm.


For the automated ECL testing, I would suggest to configure with

      --without-x

until I figured a far more robust test.


Thanks for the feddback.

-- Gaby




More information about the ecl-devel mailing list