svn commit: r335189 - in head/print/pdftk: . files

[Gerald Pfeifer]

On Tue, 10 Dec 2013, Kozlov Sergey wrote:
> Yes, that's correct. I think the right solution is to add a check if 
> gcj is present and give an IGNORE with suggestion to rebuild gcc.

Agreed.  If gcj is not present despite that dependency, that's a
good idea.  Would you like to propose a patch?

>>    LIB_DEPENDS=	libgcj.so:${PORTSDIR}/lang/gcc${CSUFF} \
>> 		libstdc++.so:${PORTSDIR}/lang/gcc${CSUFF} \
>> 		libgcc_s.so:${PORTSDIR}/lang/gcc${CSUFF}
>> 
>> If you insist on libgcj.so, I can accept that, but then let's drop
>> the other two least.  This is definitely covered by USE_GCC=yes.
> USE_GCC doesn't add anything to RUN_DEPENDS.

Sure it does:

  .if defined(_GCC_PORT_DEPENDS)
  BUILD_DEPENDS+= ${_GCC_PORT_DEPENDS}:${PORTSDIR}/lang/${_GCC_PORT}
  . if ${_USE_GCC} != 3.4
  RUN_DEPENDS+=   ${_GCC_PORT_DEPENDS}:${PORTSDIR}/lang/${_GCC_PORT}
  # Later GCC ports already depend on binutils; make sure whatever we
  # build leverages this as well.
  USE_BINUTILS=   yes
  . endif
  .endif

> On 08.12.2013 23:39, Gerald Pfeifer wrote:
>> Can I go ahead and just make that
>> 
>>    LIB_DEPENDS=    libgcj.so:${PORTSDIR}/lang/gcc${CSUFF}
>>    BUILD_DEPENDS=  gcj${CSUFF}:${PORTSDIR}/lang/gcc${CSUFF}
> Generally speaking, I think that's ok but we should really hear the 
> opinion of port's maintainer.

Let's see what he thinks.  And I am queuing this up on my side to 
keep an eye on it. 

Gerald