raptor: should there be a note in UPDATING?

[Robert Backhaus-2]

I just had a problem building liblrdf. The cause was raptor-1 was
still installed. Checking UPDATING shows nothing that mentions raptor
after the KDE update last November.
Does there need to a note in UPDATING to uninstall raptor?
_______________________________________________
[hidden email] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-ports
To unsubscribe, send any mail to "[hidden email]"

[Raphael Kubo da Costa-2]

Robert Backhaus <[hidden email]> writes:

> I just had a problem building liblrdf. The cause was raptor-1 was
> still installed. Checking UPDATING shows nothing that mentions raptor
> after the KDE update last November.
> Does there need to a note in UPDATING to uninstall raptor?

It seems more appropriate to fix the port: according to the build log
posted in <[hidden email]> (the other liblrdf
thread), -I/usr/local/include is being passed before
-I/usr/local/include/raptor2, and it is likely that the source code
#includes <raptor.h>, which is installed by both raptor and raptor2
(albeit into different directories).

Someone (TM) should check whether the first include is really necessary,
or if the code can't include <raptor2/raptor.h>, or if
-I/usr/local/include/raptor2 can't be passed before
-I/usr/local/include.

_______________________________________________
[hidden email] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-ports
To unsubscribe, send any mail to "[hidden email]"

[Raphael Kubo da Costa-2]

Raphael Kubo da Costa <[hidden email]> writes:

> Someone (TM) should check whether the first include is really necessary,
> or if the code can't include <raptor2/raptor.h>, or if
> -I/usr/local/include/raptor2 can't be passed before
> -I/usr/local/include.

Everything built fine without the {CPP,LD}FLAGS additions here. I'm
tempted to commit that change, but wonder if it will end up breaking
some weird setup...

_______________________________________________
[hidden email] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-ports
To unsubscribe, send any mail to "[hidden email]"

[Raphael Kubo da Costa-2]

Raphael Kubo da Costa <[hidden email]> writes:

> Raphael Kubo da Costa <[hidden email]> writes:
>
>> Someone (TM) should check whether the first include is really necessary,
>> or if the code can't include <raptor2/raptor.h>, or if
>> -I/usr/local/include/raptor2 can't be passed before
>> -I/usr/local/include.
>
> Everything built fine without the {CPP,LD}FLAGS additions here. I'm
> tempted to commit that change, but wonder if it will end up breaking
> some weird setup...

Coming to think of it again, removing those lines changing {CPP,LD}FLAGS
would have an impact if one is using openssl from ports, as the check
for MD5() in libcrypto.so would be done for the library in base, not
${LOCALBASE}.

_______________________________________________
[hidden email] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-ports
To unsubscribe, send any mail to "[hidden email]"

[Raphael Kubo da Costa-2]

Raphael Kubo da Costa <[hidden email]> writes:

> Raphael Kubo da Costa <[hidden email]> writes:
>
>> Raphael Kubo da Costa <[hidden email]> writes:
>>
>>> Someone (TM) should check whether the first include is really necessary,
>>> or if the code can't include <raptor2/raptor.h>, or if
>>> -I/usr/local/include/raptor2 can't be passed before
>>> -I/usr/local/include.
>>
>> Everything built fine without the {CPP,LD}FLAGS additions here. I'm
>> tempted to commit that change, but wonder if it will end up breaking
>> some weird setup...
>
> Coming to think of it again, removing those lines changing {CPP,LD}FLAGS
> would have an impact if one is using openssl from ports, as the check
> for MD5() in libcrypto.so would be done for the library in base, not
> ${LOCALBASE}.

I think I've finally fixed the issue in r302968.

_______________________________________________
[hidden email] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-ports
To unsubscribe, send any mail to "[hidden email]"