Re: [R-SIG-Mac] compiling from source on ML, tcltk

2012-09-04 Thread Kasper Daniel Hansen
On Tue, Sep 4, 2012 at 11:05 AM, Simon Urbanek
 wrote:
>
> On Sep 4, 2012, at 8:47 AM, Kasper Daniel Hansen 
>  wrote:
>
>> On Mon, Sep 3, 2012 at 2:22 AM, Prof Brian Ripley  
>> wrote:
>>> On 03/09/2012 01:11, Kasper Daniel Hansen wrote:

 On Sun, Sep 2, 2012 at 3:54 PM, Prof Brian Ripley 
 wrote:
>
> Do you need R.app?  If not, I would recommend using ActiveTcl Tcl/Tk by
> e.g.
>
> ./configure ...
> --with-tcl-config=/Library/Frameworks/Tcl.framework/tclConfig.sh
> --with-tk-config=/Library/Frameworks/Tk.framework/tkConfig.sh


 I don't have (Tcl|Tk).framework in /Library/Frameworks.  I seem to
 recall they were there in earlier versions of the OS.
>>>
>>>
>>> On my system, that is where ActiveTcl installed.   Even when Apple shipped
>>> an Aqua Tcl/Tk, it was old and limited and ActiveTcl is quite way to get a
>>> pre-compiled current version.
>>>
>>>
 I did some more investigation.  Following a suggestion on Stack overflow I
 did
   sudo ln -s /opt/X11/include/X11 /usr/include
 and that suddenly gave me tcltk capabilities.  To be honest, I don't
 understand why. I use
   --X-Includes=/Usr/X11/Include --X-Libraries=/Usr/X11/Lib
>>>
>>>
>>> If you use /Usr then it may well not work.  These tools are case-sensitive
>>> 
>>
>> Thanks, the lines I have in my email I wrong - I did some manual
>> editing as I was testing some stuff.
>>
>> So to be clear, I use
>>  --x-includes=/usr/X11/include --x-libraries=/usr/X11/lib\
>> Without the symbolic link I get aqua, X11 support
>>
>> With the symbolic link I get aqua, X11, tcltk support
>>
>> Looking at config.log, it seems that all the test tcl and/or tk
>> programs complaining about missing X11/Xlib.h.  When I look at the gcc
>> command, I see stuff like
>>  /usr/bin/gcc-4.2 -arch x86_64 -std=gnu99 -E -I/usr/local/include
>> -I/usr/local/include conftest.c
>> Include directories depends on on which test we are talking about, but
>> never includes the directory I specified with --x-includes, which
>> seems wrong to me.
>>
>> This could explain why making the symlink (so we have X11/Xlib.h under
>> /usr/include) makes tcltk work.
>>
>
> That's the wrong location, certainly.
> However, could it be that your problem is a missing /usr/X11R6 symlink?

# ls -l /usr/
total 24
lrwxr-xr-x 1 root  wheel 8B Aug 28 16:57 X11 -> /opt/X11
lrwxr-xr-x 1 root  wheel 8B Aug 28 16:57 X11R6 -> /opt/X11
 : the rest of the standard directories excluded.

And /opt/X11 seems to have the standard include, lib, bin. etc, ...  Also,

# find  /opt/X11 -name "Xlib.h"
/opt/X11/include/X11/Xlib.h

This is a (pretty) fresh upgrade from Lion to Mountain Lion.  I did
have various stuff installed in Lion of course.

Kasper


>
> Cheers,
> Simon
>
>
>>
>>>
>>>
 in my configure script so I don't fully understand why the presence of
 this symbolic link fixes it.  But it is an empirical fact it does

 Kasper



>
> I've been using it for some time, and it has much nicer ttk widgets.
>
>
>
> On 02/09/2012 19:21, Kasper Daniel Hansen wrote:
>>
>>
>> I now want to get tcltk support working under ML (because I need to
>> test a few packages).
>>
>> I have installed Xquartz and I have just install the new tcltk 8.5.5
>> from CRAN (which is not signed btw).
>>
>> Running configure, I get
>>
>> checking for tclConfig.sh... no
>> checking for tclConfig.sh in library (sub)directories...
>> /usr/local/lib/tclConfig.sh
>> checking for tkConfig.sh... no
>> checking for tkConfig.sh in library (sub)directories...
>> /usr/local/lib/tkConfig.sh
>> checking tcl.h usability... yes
>> checking tcl.h presence... yes
>> checking for tcl.h... yes
>> checking tk.h usability... no
>> checking tk.h presence... no
>> checking for tk.h... no
>> checking /usr/local/include/tk8.5/generic/tk.h usability... no
>> checking /usr/local/include/tk8.5/generic/tk.h presence... no
>> checking for /usr/local/include/tk8.5/generic/tk.h... no
>> checking /usr/local/include/tk8.5/tk.h usability... no
>> checking /usr/local/include/tk8.5/tk.h presence... no
>> checking for /usr/local/include/tk8.5/tk.h... no
>> checking /usr/local/include/tcl8.5/tk.h usability... no
>> checking /usr/local/include/tcl8.5/tk.h presence... no
>> checking for /usr/local/include/tcl8.5/tk.h... no
>> checking /usr/local/include/tk.h usability... no
>> checking /usr/local/include/tk.h presence... no
>> checking for /usr/local/include/tk.h... no
>> checking for tk.h... yes
>>
>> So it seems to find tcltk.  I get a few warnings when it checks for X11
>> headers:
>>
>> checking X11/Intrinsic.h presence... no
>> configure: WARNING: X11/Intrinsic.h: accepted by the compiler,
>> rejected by the preprocessor!
>> configure: WARNING: X11/Intrinsic.h: pro

Re: [R-SIG-Mac] compiling from source on ML, tcltk

2012-09-04 Thread Simon Urbanek

On Sep 4, 2012, at 8:47 AM, Kasper Daniel Hansen  
wrote:

> On Mon, Sep 3, 2012 at 2:22 AM, Prof Brian Ripley  
> wrote:
>> On 03/09/2012 01:11, Kasper Daniel Hansen wrote:
>>> 
>>> On Sun, Sep 2, 2012 at 3:54 PM, Prof Brian Ripley 
>>> wrote:
 
 Do you need R.app?  If not, I would recommend using ActiveTcl Tcl/Tk by
 e.g.
 
 ./configure ...
 --with-tcl-config=/Library/Frameworks/Tcl.framework/tclConfig.sh
 --with-tk-config=/Library/Frameworks/Tk.framework/tkConfig.sh
>>> 
>>> 
>>> I don't have (Tcl|Tk).framework in /Library/Frameworks.  I seem to
>>> recall they were there in earlier versions of the OS.
>> 
>> 
>> On my system, that is where ActiveTcl installed.   Even when Apple shipped
>> an Aqua Tcl/Tk, it was old and limited and ActiveTcl is quite way to get a
>> pre-compiled current version.
>> 
>> 
>>> I did some more investigation.  Following a suggestion on Stack overflow I
>>> did
>>>   sudo ln -s /opt/X11/include/X11 /usr/include
>>> and that suddenly gave me tcltk capabilities.  To be honest, I don't
>>> understand why. I use
>>>   --X-Includes=/Usr/X11/Include --X-Libraries=/Usr/X11/Lib
>> 
>> 
>> If you use /Usr then it may well not work.  These tools are case-sensitive
>> 
> 
> Thanks, the lines I have in my email I wrong - I did some manual
> editing as I was testing some stuff.
> 
> So to be clear, I use
>  --x-includes=/usr/X11/include --x-libraries=/usr/X11/lib\
> Without the symbolic link I get aqua, X11 support
> 
> With the symbolic link I get aqua, X11, tcltk support
> 
> Looking at config.log, it seems that all the test tcl and/or tk
> programs complaining about missing X11/Xlib.h.  When I look at the gcc
> command, I see stuff like
>  /usr/bin/gcc-4.2 -arch x86_64 -std=gnu99 -E -I/usr/local/include
> -I/usr/local/include conftest.c
> Include directories depends on on which test we are talking about, but
> never includes the directory I specified with --x-includes, which
> seems wrong to me.
> 
> This could explain why making the symlink (so we have X11/Xlib.h under
> /usr/include) makes tcltk work.
> 

That's the wrong location, certainly.
However, could it be that your problem is a missing /usr/X11R6 symlink?

Cheers,
Simon


> 
>> 
>> 
>>> in my configure script so I don't fully understand why the presence of
>>> this symbolic link fixes it.  But it is an empirical fact it does
>>> 
>>> Kasper
>>> 
>>> 
>>> 
 
 I've been using it for some time, and it has much nicer ttk widgets.
 
 
 
 On 02/09/2012 19:21, Kasper Daniel Hansen wrote:
> 
> 
> I now want to get tcltk support working under ML (because I need to
> test a few packages).
> 
> I have installed Xquartz and I have just install the new tcltk 8.5.5
> from CRAN (which is not signed btw).
> 
> Running configure, I get
> 
> checking for tclConfig.sh... no
> checking for tclConfig.sh in library (sub)directories...
> /usr/local/lib/tclConfig.sh
> checking for tkConfig.sh... no
> checking for tkConfig.sh in library (sub)directories...
> /usr/local/lib/tkConfig.sh
> checking tcl.h usability... yes
> checking tcl.h presence... yes
> checking for tcl.h... yes
> checking tk.h usability... no
> checking tk.h presence... no
> checking for tk.h... no
> checking /usr/local/include/tk8.5/generic/tk.h usability... no
> checking /usr/local/include/tk8.5/generic/tk.h presence... no
> checking for /usr/local/include/tk8.5/generic/tk.h... no
> checking /usr/local/include/tk8.5/tk.h usability... no
> checking /usr/local/include/tk8.5/tk.h presence... no
> checking for /usr/local/include/tk8.5/tk.h... no
> checking /usr/local/include/tcl8.5/tk.h usability... no
> checking /usr/local/include/tcl8.5/tk.h presence... no
> checking for /usr/local/include/tcl8.5/tk.h... no
> checking /usr/local/include/tk.h usability... no
> checking /usr/local/include/tk.h presence... no
> checking for /usr/local/include/tk.h... no
> checking for tk.h... yes
> 
> So it seems to find tcltk.  I get a few warnings when it checks for X11
> headers:
> 
> checking X11/Intrinsic.h presence... no
> configure: WARNING: X11/Intrinsic.h: accepted by the compiler,
> rejected by the preprocessor!
> configure: WARNING: X11/Intrinsic.h: proceeding with the compiler's
> result
> 
> and
> 
> checking X11/Xmu/Atoms.h presence... no
> configure: WARNING: X11/Xmu/Atoms.h: accepted by the compiler,
> rejected by the preprocessor!
> configure: WARNING: X11/Xmu/Atoms.h: proceeding with the compiler's
> result
> 
> I assume these are unrelated.  Although configure seems to find tcltk,
> at the end of configure, I get
> 
> R is now configured for x86_64-apple-darwin12.1.0
> 
>Source directory:  ../R-2.15.x-src
>Installation directory:/usr/local/R/R-2.15.x
> 
>C c

Re: [R-SIG-Mac] compiling from source on ML, tcltk

2012-09-04 Thread Kasper Daniel Hansen
On Mon, Sep 3, 2012 at 2:22 AM, Prof Brian Ripley  wrote:
> On 03/09/2012 01:11, Kasper Daniel Hansen wrote:
>>
>> On Sun, Sep 2, 2012 at 3:54 PM, Prof Brian Ripley 
>> wrote:
>>>
>>> Do you need R.app?  If not, I would recommend using ActiveTcl Tcl/Tk by
>>> e.g.
>>>
>>> ./configure ...
>>> --with-tcl-config=/Library/Frameworks/Tcl.framework/tclConfig.sh
>>> --with-tk-config=/Library/Frameworks/Tk.framework/tkConfig.sh
>>
>>
>> I don't have (Tcl|Tk).framework in /Library/Frameworks.  I seem to
>> recall they were there in earlier versions of the OS.
>
>
> On my system, that is where ActiveTcl installed.   Even when Apple shipped
> an Aqua Tcl/Tk, it was old and limited and ActiveTcl is quite way to get a
> pre-compiled current version.
>
>
>> I did some more investigation.  Following a suggestion on Stack overflow I
>> did
>>sudo ln -s /opt/X11/include/X11 /usr/include
>> and that suddenly gave me tcltk capabilities.  To be honest, I don't
>> understand why. I use
>>--X-Includes=/Usr/X11/Include --X-Libraries=/Usr/X11/Lib
>
>
> If you use /Usr then it may well not work.  These tools are case-sensitive
> 

Thanks, the lines I have in my email I wrong - I did some manual
editing as I was testing some stuff.

So to be clear, I use
  --x-includes=/usr/X11/include --x-libraries=/usr/X11/lib\
Without the symbolic link I get aqua, X11 support

With the symbolic link I get aqua, X11, tcltk support

Looking at config.log, it seems that all the test tcl and/or tk
programs complaining about missing X11/Xlib.h.  When I look at the gcc
command, I see stuff like
  /usr/bin/gcc-4.2 -arch x86_64 -std=gnu99 -E -I/usr/local/include
-I/usr/local/include conftest.c
Include directories depends on on which test we are talking about, but
never includes the directory I specified with --x-includes, which
seems wrong to me.

This could explain why making the symlink (so we have X11/Xlib.h under
/usr/include) makes tcltk work.

Kasper

>
>
>> in my configure script so I don't fully understand why the presence of
>> this symbolic link fixes it.  But it is an empirical fact it does
>>
>> Kasper
>>
>>
>>
>>>
>>> I've been using it for some time, and it has much nicer ttk widgets.
>>>
>>>
>>>
>>> On 02/09/2012 19:21, Kasper Daniel Hansen wrote:


 I now want to get tcltk support working under ML (because I need to
 test a few packages).

 I have installed Xquartz and I have just install the new tcltk 8.5.5
 from CRAN (which is not signed btw).

 Running configure, I get

 checking for tclConfig.sh... no
 checking for tclConfig.sh in library (sub)directories...
 /usr/local/lib/tclConfig.sh
 checking for tkConfig.sh... no
 checking for tkConfig.sh in library (sub)directories...
 /usr/local/lib/tkConfig.sh
 checking tcl.h usability... yes
 checking tcl.h presence... yes
 checking for tcl.h... yes
 checking tk.h usability... no
 checking tk.h presence... no
 checking for tk.h... no
 checking /usr/local/include/tk8.5/generic/tk.h usability... no
 checking /usr/local/include/tk8.5/generic/tk.h presence... no
 checking for /usr/local/include/tk8.5/generic/tk.h... no
 checking /usr/local/include/tk8.5/tk.h usability... no
 checking /usr/local/include/tk8.5/tk.h presence... no
 checking for /usr/local/include/tk8.5/tk.h... no
 checking /usr/local/include/tcl8.5/tk.h usability... no
 checking /usr/local/include/tcl8.5/tk.h presence... no
 checking for /usr/local/include/tcl8.5/tk.h... no
 checking /usr/local/include/tk.h usability... no
 checking /usr/local/include/tk.h presence... no
 checking for /usr/local/include/tk.h... no
 checking for tk.h... yes

 So it seems to find tcltk.  I get a few warnings when it checks for X11
 headers:

 checking X11/Intrinsic.h presence... no
 configure: WARNING: X11/Intrinsic.h: accepted by the compiler,
 rejected by the preprocessor!
 configure: WARNING: X11/Intrinsic.h: proceeding with the compiler's
 result

 and

 checking X11/Xmu/Atoms.h presence... no
 configure: WARNING: X11/Xmu/Atoms.h: accepted by the compiler,
 rejected by the preprocessor!
 configure: WARNING: X11/Xmu/Atoms.h: proceeding with the compiler's
 result

 I assume these are unrelated.  Although configure seems to find tcltk,
 at the end of configure, I get

 R is now configured for x86_64-apple-darwin12.1.0

 Source directory:  ../R-2.15.x-src
 Installation directory:/usr/local/R/R-2.15.x

 C compiler:/usr/bin/gcc-4.2 -arch x86_64 -std=gnu99
 -g -O2 -std=gnu99 -march=nocona
 Fortran 77 compiler:   /usr/bin/gfortran-4.2 -arch x86_64  -g
 -O2 -march=nocona

 C++ compiler:  /usr/bin/g++-4.2 -arch x86_64  -g -O2
 -march=nocona
 Fortran 90/95 compiler:/usr/bin/gfortran-4.2 -arch x86_64

Re: [R-SIG-Mac] compiling from source on ML, tcltk

2012-09-02 Thread Prof Brian Ripley

On 03/09/2012 01:11, Kasper Daniel Hansen wrote:

On Sun, Sep 2, 2012 at 3:54 PM, Prof Brian Ripley  wrote:

Do you need R.app?  If not, I would recommend using ActiveTcl Tcl/Tk by e.g.

./configure ...
--with-tcl-config=/Library/Frameworks/Tcl.framework/tclConfig.sh
--with-tk-config=/Library/Frameworks/Tk.framework/tkConfig.sh


I don't have (Tcl|Tk).framework in /Library/Frameworks.  I seem to
recall they were there in earlier versions of the OS.


On my system, that is where ActiveTcl installed.   Even when Apple 
shipped an Aqua Tcl/Tk, it was old and limited and ActiveTcl is quite 
way to get a pre-compiled current version.



I did some more investigation.  Following a suggestion on Stack overflow I did
   sudo ln -s /opt/X11/include/X11 /usr/include
and that suddenly gave me tcltk capabilities.  To be honest, I don't
understand why. I use
   --X-Includes=/Usr/X11/Include --X-Libraries=/Usr/X11/Lib


If you use /Usr then it may well not work.  These tools are 
case-sensitive 



in my configure script so I don't fully understand why the presence of
this symbolic link fixes it.  But it is an empirical fact it does

Kasper





I've been using it for some time, and it has much nicer ttk widgets.



On 02/09/2012 19:21, Kasper Daniel Hansen wrote:


I now want to get tcltk support working under ML (because I need to
test a few packages).

I have installed Xquartz and I have just install the new tcltk 8.5.5
from CRAN (which is not signed btw).

Running configure, I get

checking for tclConfig.sh... no
checking for tclConfig.sh in library (sub)directories...
/usr/local/lib/tclConfig.sh
checking for tkConfig.sh... no
checking for tkConfig.sh in library (sub)directories...
/usr/local/lib/tkConfig.sh
checking tcl.h usability... yes
checking tcl.h presence... yes
checking for tcl.h... yes
checking tk.h usability... no
checking tk.h presence... no
checking for tk.h... no
checking /usr/local/include/tk8.5/generic/tk.h usability... no
checking /usr/local/include/tk8.5/generic/tk.h presence... no
checking for /usr/local/include/tk8.5/generic/tk.h... no
checking /usr/local/include/tk8.5/tk.h usability... no
checking /usr/local/include/tk8.5/tk.h presence... no
checking for /usr/local/include/tk8.5/tk.h... no
checking /usr/local/include/tcl8.5/tk.h usability... no
checking /usr/local/include/tcl8.5/tk.h presence... no
checking for /usr/local/include/tcl8.5/tk.h... no
checking /usr/local/include/tk.h usability... no
checking /usr/local/include/tk.h presence... no
checking for /usr/local/include/tk.h... no
checking for tk.h... yes

So it seems to find tcltk.  I get a few warnings when it checks for X11
headers:

checking X11/Intrinsic.h presence... no
configure: WARNING: X11/Intrinsic.h: accepted by the compiler,
rejected by the preprocessor!
configure: WARNING: X11/Intrinsic.h: proceeding with the compiler's result

and

checking X11/Xmu/Atoms.h presence... no
configure: WARNING: X11/Xmu/Atoms.h: accepted by the compiler,
rejected by the preprocessor!
configure: WARNING: X11/Xmu/Atoms.h: proceeding with the compiler's result

I assume these are unrelated.  Although configure seems to find tcltk,
at the end of configure, I get

R is now configured for x86_64-apple-darwin12.1.0

Source directory:  ../R-2.15.x-src
Installation directory:/usr/local/R/R-2.15.x

C compiler:/usr/bin/gcc-4.2 -arch x86_64 -std=gnu99
-g -O2 -std=gnu99 -march=nocona
Fortran 77 compiler:   /usr/bin/gfortran-4.2 -arch x86_64  -g
-O2 -march=nocona

C++ compiler:  /usr/bin/g++-4.2 -arch x86_64  -g -O2
-march=nocona
Fortran 90/95 compiler:/usr/bin/gfortran-4.2 -arch x86_64 -g -O2
-march=nocona
Obj-C compiler:/usr/bin/gcc-4.2 -arch x86_64 -g -O2
-fobjc-exceptions

Interfaces supported:  X11, aqua
External libraries:readline, BLAS(vecLib), LAPACK(in blas), ICU
Additional capabilities:   PNG, JPEG, NLS
Options enabled:   R profiling, memory profiling, Java

Recommended packages:  yes

which does not mention tcltk support under 'Interfaces supported'.

This used to work on Lion, and I am wondering what I am doing wrong.



I think you should be wondering what Apple is doing wrong   But you need
to look in config.log to find the root cause.



My configure script is

export LANG=en_US.UTF-8
../${SRCDIR}/configure SHELL='/bin/bash' \
--prefix=/usr/local/R/R-${R_VERSION} --disable-R-framework\
CC="/usr/bin/gcc-4.2 -arch x86_64 -std=gnu99" \
CFLAGS="-g -O2 -std=gnu99 -march=nocona" \
CXX="/usr/bin/g++-4.2 -arch x86_64" \
CXXFLAGS="-g -O2 -march=nocona" \
OBJC="/usr/bin/gcc-4.2 -arch x86_64" \
F77="/usr/bin/gfortran-4.2 -arch x86_64" \
FFLAGS="-g -O2 -march=nocona" \
FC="/usr/bin/gfortran-4.2 -arch x86_64" \
FCFLAGS="-g -O2 -march=nocona" \
--enable-memory-profiling\
--X-Includes=/Usr/X11/Include --X-Libraries=/Usr/X11/Lib\
--with-system-zlib\
--with-blas='-fr

Re: [R-SIG-Mac] compiling from source on ML, tcltk

2012-09-02 Thread Kasper Daniel Hansen
On Sun, Sep 2, 2012 at 3:54 PM, Prof Brian Ripley  wrote:
> Do you need R.app?  If not, I would recommend using ActiveTcl Tcl/Tk by e.g.
>
> ./configure ...
> --with-tcl-config=/Library/Frameworks/Tcl.framework/tclConfig.sh
> --with-tk-config=/Library/Frameworks/Tk.framework/tkConfig.sh

I don't have (Tcl|Tk).framework in /Library/Frameworks.  I seem to
recall they were there in earlier versions of the OS.

I did some more investigation.  Following a suggestion on Stack overflow I did
  sudo ln -s /opt/X11/include/X11 /usr/include
and that suddenly gave me tcltk capabilities.  To be honest, I don't
understand why. I use
  --X-Includes=/Usr/X11/Include --X-Libraries=/Usr/X11/Lib
in my configure script so I don't fully understand why the presence of
this symbolic link fixes it.  But it is an empirical fact it does

Kasper



>
> I've been using it for some time, and it has much nicer ttk widgets.
>
>
>
> On 02/09/2012 19:21, Kasper Daniel Hansen wrote:
>>
>> I now want to get tcltk support working under ML (because I need to
>> test a few packages).
>>
>> I have installed Xquartz and I have just install the new tcltk 8.5.5
>> from CRAN (which is not signed btw).
>>
>> Running configure, I get
>>
>> checking for tclConfig.sh... no
>> checking for tclConfig.sh in library (sub)directories...
>> /usr/local/lib/tclConfig.sh
>> checking for tkConfig.sh... no
>> checking for tkConfig.sh in library (sub)directories...
>> /usr/local/lib/tkConfig.sh
>> checking tcl.h usability... yes
>> checking tcl.h presence... yes
>> checking for tcl.h... yes
>> checking tk.h usability... no
>> checking tk.h presence... no
>> checking for tk.h... no
>> checking /usr/local/include/tk8.5/generic/tk.h usability... no
>> checking /usr/local/include/tk8.5/generic/tk.h presence... no
>> checking for /usr/local/include/tk8.5/generic/tk.h... no
>> checking /usr/local/include/tk8.5/tk.h usability... no
>> checking /usr/local/include/tk8.5/tk.h presence... no
>> checking for /usr/local/include/tk8.5/tk.h... no
>> checking /usr/local/include/tcl8.5/tk.h usability... no
>> checking /usr/local/include/tcl8.5/tk.h presence... no
>> checking for /usr/local/include/tcl8.5/tk.h... no
>> checking /usr/local/include/tk.h usability... no
>> checking /usr/local/include/tk.h presence... no
>> checking for /usr/local/include/tk.h... no
>> checking for tk.h... yes
>>
>> So it seems to find tcltk.  I get a few warnings when it checks for X11
>> headers:
>>
>> checking X11/Intrinsic.h presence... no
>> configure: WARNING: X11/Intrinsic.h: accepted by the compiler,
>> rejected by the preprocessor!
>> configure: WARNING: X11/Intrinsic.h: proceeding with the compiler's result
>>
>> and
>>
>> checking X11/Xmu/Atoms.h presence... no
>> configure: WARNING: X11/Xmu/Atoms.h: accepted by the compiler,
>> rejected by the preprocessor!
>> configure: WARNING: X11/Xmu/Atoms.h: proceeding with the compiler's result
>>
>> I assume these are unrelated.  Although configure seems to find tcltk,
>> at the end of configure, I get
>>
>> R is now configured for x86_64-apple-darwin12.1.0
>>
>>Source directory:  ../R-2.15.x-src
>>Installation directory:/usr/local/R/R-2.15.x
>>
>>C compiler:/usr/bin/gcc-4.2 -arch x86_64 -std=gnu99
>> -g -O2 -std=gnu99 -march=nocona
>>Fortran 77 compiler:   /usr/bin/gfortran-4.2 -arch x86_64  -g
>> -O2 -march=nocona
>>
>>C++ compiler:  /usr/bin/g++-4.2 -arch x86_64  -g -O2
>> -march=nocona
>>Fortran 90/95 compiler:/usr/bin/gfortran-4.2 -arch x86_64 -g -O2
>> -march=nocona
>>Obj-C compiler:/usr/bin/gcc-4.2 -arch x86_64 -g -O2
>> -fobjc-exceptions
>>
>>Interfaces supported:  X11, aqua
>>External libraries:readline, BLAS(vecLib), LAPACK(in blas), ICU
>>Additional capabilities:   PNG, JPEG, NLS
>>Options enabled:   R profiling, memory profiling, Java
>>
>>Recommended packages:  yes
>>
>> which does not mention tcltk support under 'Interfaces supported'.
>>
>> This used to work on Lion, and I am wondering what I am doing wrong.
>
>
> I think you should be wondering what Apple is doing wrong   But you need
> to look in config.log to find the root cause.
>
>>
>> My configure script is
>>
>> export LANG=en_US.UTF-8
>> ../${SRCDIR}/configure SHELL='/bin/bash' \
>>--prefix=/usr/local/R/R-${R_VERSION} --disable-R-framework\
>>CC="/usr/bin/gcc-4.2 -arch x86_64 -std=gnu99" \
>>CFLAGS="-g -O2 -std=gnu99 -march=nocona" \
>>CXX="/usr/bin/g++-4.2 -arch x86_64" \
>>CXXFLAGS="-g -O2 -march=nocona" \
>>OBJC="/usr/bin/gcc-4.2 -arch x86_64" \
>>F77="/usr/bin/gfortran-4.2 -arch x86_64" \
>>FFLAGS="-g -O2 -march=nocona" \
>>FC="/usr/bin/gfortran-4.2 -arch x86_64" \
>>FCFLAGS="-g -O2 -march=nocona" \
>>--enable-memory-profiling\
>>--X-Includes=/Usr/X11/Include --X-Libraries=/Usr/X11/Lib\
>>--with-system-zlib\
>>--with-blas='-framework vecLib' --with-lapack
>>
>> Thanks,
>> Kasper
>

Re: [R-SIG-Mac] compiling from source on ML, tcltk

2012-09-02 Thread Prof Brian Ripley

Do you need R.app?  If not, I would recommend using ActiveTcl Tcl/Tk by e.g.

./configure ... 
--with-tcl-config=/Library/Frameworks/Tcl.framework/tclConfig.sh 
--with-tk-config=/Library/Frameworks/Tk.framework/tkConfig.sh


I've been using it for some time, and it has much nicer ttk widgets.


On 02/09/2012 19:21, Kasper Daniel Hansen wrote:

I now want to get tcltk support working under ML (because I need to
test a few packages).

I have installed Xquartz and I have just install the new tcltk 8.5.5
from CRAN (which is not signed btw).

Running configure, I get

checking for tclConfig.sh... no
checking for tclConfig.sh in library (sub)directories...
/usr/local/lib/tclConfig.sh
checking for tkConfig.sh... no
checking for tkConfig.sh in library (sub)directories...
/usr/local/lib/tkConfig.sh
checking tcl.h usability... yes
checking tcl.h presence... yes
checking for tcl.h... yes
checking tk.h usability... no
checking tk.h presence... no
checking for tk.h... no
checking /usr/local/include/tk8.5/generic/tk.h usability... no
checking /usr/local/include/tk8.5/generic/tk.h presence... no
checking for /usr/local/include/tk8.5/generic/tk.h... no
checking /usr/local/include/tk8.5/tk.h usability... no
checking /usr/local/include/tk8.5/tk.h presence... no
checking for /usr/local/include/tk8.5/tk.h... no
checking /usr/local/include/tcl8.5/tk.h usability... no
checking /usr/local/include/tcl8.5/tk.h presence... no
checking for /usr/local/include/tcl8.5/tk.h... no
checking /usr/local/include/tk.h usability... no
checking /usr/local/include/tk.h presence... no
checking for /usr/local/include/tk.h... no
checking for tk.h... yes

So it seems to find tcltk.  I get a few warnings when it checks for X11 headers:

checking X11/Intrinsic.h presence... no
configure: WARNING: X11/Intrinsic.h: accepted by the compiler,
rejected by the preprocessor!
configure: WARNING: X11/Intrinsic.h: proceeding with the compiler's result

and

checking X11/Xmu/Atoms.h presence... no
configure: WARNING: X11/Xmu/Atoms.h: accepted by the compiler,
rejected by the preprocessor!
configure: WARNING: X11/Xmu/Atoms.h: proceeding with the compiler's result

I assume these are unrelated.  Although configure seems to find tcltk,
at the end of configure, I get

R is now configured for x86_64-apple-darwin12.1.0

   Source directory:  ../R-2.15.x-src
   Installation directory:/usr/local/R/R-2.15.x

   C compiler:/usr/bin/gcc-4.2 -arch x86_64 -std=gnu99
-g -O2 -std=gnu99 -march=nocona
   Fortran 77 compiler:   /usr/bin/gfortran-4.2 -arch x86_64  -g
-O2 -march=nocona

   C++ compiler:  /usr/bin/g++-4.2 -arch x86_64  -g -O2 
-march=nocona
   Fortran 90/95 compiler:/usr/bin/gfortran-4.2 -arch x86_64 -g -O2
-march=nocona
   Obj-C compiler:/usr/bin/gcc-4.2 -arch x86_64 -g -O2
-fobjc-exceptions

   Interfaces supported:  X11, aqua
   External libraries:readline, BLAS(vecLib), LAPACK(in blas), ICU
   Additional capabilities:   PNG, JPEG, NLS
   Options enabled:   R profiling, memory profiling, Java

   Recommended packages:  yes

which does not mention tcltk support under 'Interfaces supported'.

This used to work on Lion, and I am wondering what I am doing wrong.


I think you should be wondering what Apple is doing wrong   But you 
need to look in config.log to find the root cause.




My configure script is

export LANG=en_US.UTF-8
../${SRCDIR}/configure SHELL='/bin/bash' \
   --prefix=/usr/local/R/R-${R_VERSION} --disable-R-framework\
   CC="/usr/bin/gcc-4.2 -arch x86_64 -std=gnu99" \
   CFLAGS="-g -O2 -std=gnu99 -march=nocona" \
   CXX="/usr/bin/g++-4.2 -arch x86_64" \
   CXXFLAGS="-g -O2 -march=nocona" \
   OBJC="/usr/bin/gcc-4.2 -arch x86_64" \
   F77="/usr/bin/gfortran-4.2 -arch x86_64" \
   FFLAGS="-g -O2 -march=nocona" \
   FC="/usr/bin/gfortran-4.2 -arch x86_64" \
   FCFLAGS="-g -O2 -march=nocona" \
   --enable-memory-profiling\
   --x-includes=/usr/X11/include --x-libraries=/usr/X11/lib\
   --with-system-zlib\
   --with-blas='-framework vecLib' --with-lapack

Thanks,
Kasper

___
R-SIG-Mac mailing list
R-SIG-Mac@r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-mac




--
Brian D. Ripley,  rip...@stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford, Tel:  +44 1865 272861 (self)
1 South Parks Road, +44 1865 272866 (PA)
Oxford OX1 3TG, UKFax:  +44 1865 272595

___
R-SIG-Mac mailing list
R-SIG-Mac@r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-mac


[R-SIG-Mac] compiling from source on ML, tcltk

2012-09-02 Thread Kasper Daniel Hansen
I now want to get tcltk support working under ML (because I need to
test a few packages).

I have installed Xquartz and I have just install the new tcltk 8.5.5
from CRAN (which is not signed btw).

Running configure, I get

checking for tclConfig.sh... no
checking for tclConfig.sh in library (sub)directories...
/usr/local/lib/tclConfig.sh
checking for tkConfig.sh... no
checking for tkConfig.sh in library (sub)directories...
/usr/local/lib/tkConfig.sh
checking tcl.h usability... yes
checking tcl.h presence... yes
checking for tcl.h... yes
checking tk.h usability... no
checking tk.h presence... no
checking for tk.h... no
checking /usr/local/include/tk8.5/generic/tk.h usability... no
checking /usr/local/include/tk8.5/generic/tk.h presence... no
checking for /usr/local/include/tk8.5/generic/tk.h... no
checking /usr/local/include/tk8.5/tk.h usability... no
checking /usr/local/include/tk8.5/tk.h presence... no
checking for /usr/local/include/tk8.5/tk.h... no
checking /usr/local/include/tcl8.5/tk.h usability... no
checking /usr/local/include/tcl8.5/tk.h presence... no
checking for /usr/local/include/tcl8.5/tk.h... no
checking /usr/local/include/tk.h usability... no
checking /usr/local/include/tk.h presence... no
checking for /usr/local/include/tk.h... no
checking for tk.h... yes

So it seems to find tcltk.  I get a few warnings when it checks for X11 headers:

checking X11/Intrinsic.h presence... no
configure: WARNING: X11/Intrinsic.h: accepted by the compiler,
rejected by the preprocessor!
configure: WARNING: X11/Intrinsic.h: proceeding with the compiler's result

and

checking X11/Xmu/Atoms.h presence... no
configure: WARNING: X11/Xmu/Atoms.h: accepted by the compiler,
rejected by the preprocessor!
configure: WARNING: X11/Xmu/Atoms.h: proceeding with the compiler's result

I assume these are unrelated.  Although configure seems to find tcltk,
at the end of configure, I get

R is now configured for x86_64-apple-darwin12.1.0

  Source directory:  ../R-2.15.x-src
  Installation directory:/usr/local/R/R-2.15.x

  C compiler:/usr/bin/gcc-4.2 -arch x86_64 -std=gnu99
-g -O2 -std=gnu99 -march=nocona
  Fortran 77 compiler:   /usr/bin/gfortran-4.2 -arch x86_64  -g
-O2 -march=nocona

  C++ compiler:  /usr/bin/g++-4.2 -arch x86_64  -g -O2 -march=nocona
  Fortran 90/95 compiler:/usr/bin/gfortran-4.2 -arch x86_64 -g -O2
-march=nocona
  Obj-C compiler:/usr/bin/gcc-4.2 -arch x86_64 -g -O2
-fobjc-exceptions

  Interfaces supported:  X11, aqua
  External libraries:readline, BLAS(vecLib), LAPACK(in blas), ICU
  Additional capabilities:   PNG, JPEG, NLS
  Options enabled:   R profiling, memory profiling, Java

  Recommended packages:  yes

which does not mention tcltk support under 'Interfaces supported'.

This used to work on Lion, and I am wondering what I am doing wrong.

My configure script is

export LANG=en_US.UTF-8
../${SRCDIR}/configure SHELL='/bin/bash' \
  --prefix=/usr/local/R/R-${R_VERSION} --disable-R-framework\
  CC="/usr/bin/gcc-4.2 -arch x86_64 -std=gnu99" \
  CFLAGS="-g -O2 -std=gnu99 -march=nocona" \
  CXX="/usr/bin/g++-4.2 -arch x86_64" \
  CXXFLAGS="-g -O2 -march=nocona" \
  OBJC="/usr/bin/gcc-4.2 -arch x86_64" \
  F77="/usr/bin/gfortran-4.2 -arch x86_64" \
  FFLAGS="-g -O2 -march=nocona" \
  FC="/usr/bin/gfortran-4.2 -arch x86_64" \
  FCFLAGS="-g -O2 -march=nocona" \
  --enable-memory-profiling\
  --x-includes=/usr/X11/include --x-libraries=/usr/X11/lib\
  --with-system-zlib\
  --with-blas='-framework vecLib' --with-lapack

Thanks,
Kasper

___
R-SIG-Mac mailing list
R-SIG-Mac@r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-mac