From maj at fortinbras.us  Fri May  1 14:11:49 2009
From: maj at fortinbras.us (Mark A. Jensen)
Date: Fri, 1 May 2009 14:11:49 -0400
Subject: [BioLib-dev] installation on cygwin
Message-ID: <40844D11A9E84D05BBCE66BE6C76223B@NewLife>

Hey guys,
Am trying to build on cygwin (I know, I know...). Have got cmake. I do

$ cmake -DBUILD_PERL:BOOLEAN=TRUE .

in the install directory (biolib-0.0.3), and receive

-- The C compiler identification is GNU
CMake Error: Could not find cmake module 
file:/cygdrive/c/....../biolib-0.0.3/CMakeFiles/CMakeCCompiler.cmake
-- The CXX compiler identification is GNU
CMake Error: Could not find cmake module 
file:/cygdrive/c/...../biolib-0.0.3/CMakeFiles/CMakeCXXCompiler.cmake
...

Have set CYGWIN=1, CMAKE_INCLUDE_PATH=/usr/share/cmake-2.6.2/include, 
CMAKE_MODULE_PATH=/usr/share/cmake-2.6.2/Modules, still no dice. In
/usr/share/cmake-2.6.2/Modules, there exists CMakeCCompiler.cmake.in and 
CMakeCXXCompiler.cmake.in, but not
plain .cmake files.

Any tips appreciated-
cheers Mark


From pjotr.public14 at thebird.nl  Fri May  1 16:46:12 2009
From: pjotr.public14 at thebird.nl (Pjotr Prins)
Date: Fri, 1 May 2009 22:46:12 +0200
Subject: [BioLib-dev] installation on cygwin
In-Reply-To: <40844D11A9E84D05BBCE66BE6C76223B@NewLife>
References: <40844D11A9E84D05BBCE66BE6C76223B@NewLife>
Message-ID: <20090501204612.GA6590@thebird.nl>

Live dangerously ;-). The problem is that the cmake files have only
been tried on Linux and OSX. Well, I guess I'll have to try myself. I have
Windows somewhere, it is on my list now. 

Pj.

On Fri, May 01, 2009 at 02:11:49PM -0400, Mark A. Jensen wrote:
> Hey guys,
> Am trying to build on cygwin (I know, I know...). Have got cmake. I do
>
> $ cmake -DBUILD_PERL:BOOLEAN=TRUE .
>
> in the install directory (biolib-0.0.3), and receive
>
> -- The C compiler identification is GNU
> CMake Error: Could not find cmake module  
> file:/cygdrive/c/....../biolib-0.0.3/CMakeFiles/CMakeCCompiler.cmake
> -- The CXX compiler identification is GNU
> CMake Error: Could not find cmake module  
> file:/cygdrive/c/...../biolib-0.0.3/CMakeFiles/CMakeCXXCompiler.cmake
> ...
>
> Have set CYGWIN=1, CMAKE_INCLUDE_PATH=/usr/share/cmake-2.6.2/include,  
> CMAKE_MODULE_PATH=/usr/share/cmake-2.6.2/Modules, still no dice. In
> /usr/share/cmake-2.6.2/Modules, there exists CMakeCCompiler.cmake.in and  
> CMakeCXXCompiler.cmake.in, but not
> plain .cmake files.
>
> Any tips appreciated-
> cheers Mark
>
> _______________________________________________
> BioLib-dev mailing list
> BioLib-dev at lists.open-bio.org
> http://lists.open-bio.org/mailman/listinfo/biolib-dev
>

From maj at fortinbras.us  Fri May  1 16:57:58 2009
From: maj at fortinbras.us (Mark A. Jensen)
Date: Fri, 1 May 2009 16:57:58 -0400
Subject: [BioLib-dev] installation on cygwin
In-Reply-To: <20090501204612.GA6590@thebird.nl>
References: <40844D11A9E84D05BBCE66BE6C76223B@NewLife>
	<20090501204612.GA6590@thebird.nl>
Message-ID: <B62873826C8C4AB982248FDA1748A21F@NewLife>

I always seem to be stumbling into these "volunteer opportunities..."
I'll see if I can make it go, then let you know-- cheers MAJ
----- Original Message ----- 
From: "Pjotr Prins" <pjotr.public14 at thebird.nl>
To: "Mark A. Jensen" <maj at fortinbras.us>
Cc: "BioLib Project" <biolib-dev at lists.open-bio.org>
Sent: Friday, May 01, 2009 4:46 PM
Subject: Re: [BioLib-dev] installation on cygwin


> Live dangerously ;-). The problem is that the cmake files have only
> been tried on Linux and OSX. Well, I guess I'll have to try myself. I have
> Windows somewhere, it is on my list now. 
> 
> Pj.
> 
> On Fri, May 01, 2009 at 02:11:49PM -0400, Mark A. Jensen wrote:
>> Hey guys,
>> Am trying to build on cygwin (I know, I know...). Have got cmake. I do
>>
>> $ cmake -DBUILD_PERL:BOOLEAN=TRUE .
>>
>> in the install directory (biolib-0.0.3), and receive
>>
>> -- The C compiler identification is GNU
>> CMake Error: Could not find cmake module  
>> file:/cygdrive/c/....../biolib-0.0.3/CMakeFiles/CMakeCCompiler.cmake
>> -- The CXX compiler identification is GNU
>> CMake Error: Could not find cmake module  
>> file:/cygdrive/c/...../biolib-0.0.3/CMakeFiles/CMakeCXXCompiler.cmake
>> ...
>>
>> Have set CYGWIN=1, CMAKE_INCLUDE_PATH=/usr/share/cmake-2.6.2/include,  
>> CMAKE_MODULE_PATH=/usr/share/cmake-2.6.2/Modules, still no dice. In
>> /usr/share/cmake-2.6.2/Modules, there exists CMakeCCompiler.cmake.in and  
>> CMakeCXXCompiler.cmake.in, but not
>> plain .cmake files.
>>
>> Any tips appreciated-
>> cheers Mark
>>
>> _______________________________________________
>> BioLib-dev mailing list
>> BioLib-dev at lists.open-bio.org
>> http://lists.open-bio.org/mailman/listinfo/biolib-dev
>>
> 
>

From pjotr.public14 at thebird.nl  Sat May  2 06:38:30 2009
From: pjotr.public14 at thebird.nl (Pjotr Prins)
Date: Sat, 2 May 2009 12:38:30 +0200
Subject: [BioLib-dev] installation on cygwin
In-Reply-To: <40844D11A9E84D05BBCE66BE6C76223B@NewLife>
References: <40844D11A9E84D05BBCE66BE6C76223B@NewLife>
Message-ID: <20090502103830.GA10664@thebird.nl>

If have cygwin running and a version of example and staden that
builds for Perl. Only problem is the test fails to find the dynamic
library. I'll fix that soon. Checkout the cygwin branch:

  cd biolib
  git checkout -b cygwin     # create new branch
  git remote add github git://github.com/pjotrp/biolib.git
  git pull github cygwin     # pull my branch
  ./configure --with-example
  make 
  make test
  ./configure --with-staden
  make
  make test

The latest version of biolib has moved the CMake modules into a
separate repository - that is shared with ASciLib.

The problem you were seeing actually had to do with missing
submodules for git. This should be pulled in with configure:

  git submodule init
  git submodule update

Prerequisites for Cygwin are:

The BioLib build system supports building on Cygwin
(http://www.cygwin.com/). First install Cygwin on your machine
selecting the following packages:

* cmake   (development)
* gcc-g++ (development)
* make    (development)
* swig    (development)
* your preferred language: perl, python and/or ruby

and, optionally:

* git (to fetch the latest repositories)

for ASciLib:

* autoconf (development)
* automake (development)

NOTE: The AffyIO module does not yet work on Cygwin - there is a
dependency on the R libraries. I'll sort that soon.

See also ./doc/WINDOWS.txt for the latest.

On Fri, May 01, 2009 at 02:11:49PM -0400, Mark A. Jensen wrote:
> Hey guys,
> Am trying to build on cygwin (I know, I know...). Have got cmake. I do
>
> $ cmake -DBUILD_PERL:BOOLEAN=TRUE .
>
> in the install directory (biolib-0.0.3), and receive
>
> -- The C compiler identification is GNU
> CMake Error: Could not find cmake module  
> file:/cygdrive/c/....../biolib-0.0.3/CMakeFiles/CMakeCCompiler.cmake
> -- The CXX compiler identification is GNU
> CMake Error: Could not find cmake module  
> file:/cygdrive/c/...../biolib-0.0.3/CMakeFiles/CMakeCXXCompiler.cmake
> ...
>
> Have set CYGWIN=1, CMAKE_INCLUDE_PATH=/usr/share/cmake-2.6.2/include,  
> CMAKE_MODULE_PATH=/usr/share/cmake-2.6.2/Modules, still no dice. In
> /usr/share/cmake-2.6.2/Modules, there exists CMakeCCompiler.cmake.in and  
> CMakeCXXCompiler.cmake.in, but not
> plain .cmake files.
>
> Any tips appreciated-
> cheers Mark
>
> _______________________________________________
> BioLib-dev mailing list
> BioLib-dev at lists.open-bio.org
> http://lists.open-bio.org/mailman/listinfo/biolib-dev
>

From pjotr.public14 at thebird.nl  Sat May  2 14:51:39 2009
From: pjotr.public14 at thebird.nl (Pjotr Prins)
Date: Sat, 2 May 2009 20:51:39 +0200
Subject: [BioLib-dev] installation on cygwin
In-Reply-To: <20090502103830.GA10664@thebird.nl>
References: <40844D11A9E84D05BBCE66BE6C76223B@NewLife>
	<20090502103830.GA10664@thebird.nl>
Message-ID: <20090502185139.GA15817@thebird.nl>

I have merged the BioLib cygwin branch into master - preparing for a
new release. The major change is the use of modified cmake modules
which are shared with ASciLib. 

The Cygwin tests are still broken. Dynaload.pm complains it can't find
shared libraries. I have to tell cmake how to find the Cygwin shared
libs - either due to the different naming convention (.dll) or because
of some missing linker switches exporting the correct bindings.

It is just a matter of poking around.

Pj.

On Sat, May 02, 2009 at 12:38:30PM +0200, Pjotr Prins wrote:
> If have cygwin running and a version of example and staden that
> builds for Perl. Only problem is the test fails to find the dynamic
> library. I'll fix that soon. Checkout the cygwin branch:
> 
>   cd biolib
>   git checkout -b cygwin     # create new branch
>   git remote add github git://github.com/pjotrp/biolib.git
>   git pull github cygwin     # pull my branch
>   ./configure --with-example
>   make 
>   make test
>   ./configure --with-staden
>   make
>   make test
> 
> The latest version of biolib has moved the CMake modules into a
> separate repository - that is shared with ASciLib.
> 
> The problem you were seeing actually had to do with missing
> submodules for git. This should be pulled in with configure:
> 
>   git submodule init
>   git submodule update
> 
> Prerequisites for Cygwin are:
> 
> The BioLib build system supports building on Cygwin
> (http://www.cygwin.com/). First install Cygwin on your machine
> selecting the following packages:
> 
> * cmake   (development)
> * gcc-g++ (development)
> * make    (development)
> * swig    (development)
> * your preferred language: perl, python and/or ruby
> 
> and, optionally:
> 
> * git (to fetch the latest repositories)
> 
> for ASciLib:
> 
> * autoconf (development)
> * automake (development)
> 
> NOTE: The AffyIO module does not yet work on Cygwin - there is a
> dependency on the R libraries. I'll sort that soon.
> 
> See also ./doc/WINDOWS.txt for the latest.
> 
> On Fri, May 01, 2009 at 02:11:49PM -0400, Mark A. Jensen wrote:
> > Hey guys,
> > Am trying to build on cygwin (I know, I know...). Have got cmake. I do
> >
> > $ cmake -DBUILD_PERL:BOOLEAN=TRUE .
> >
> > in the install directory (biolib-0.0.3), and receive
> >
> > -- The C compiler identification is GNU
> > CMake Error: Could not find cmake module  
> > file:/cygdrive/c/....../biolib-0.0.3/CMakeFiles/CMakeCCompiler.cmake
> > -- The CXX compiler identification is GNU
> > CMake Error: Could not find cmake module  
> > file:/cygdrive/c/...../biolib-0.0.3/CMakeFiles/CMakeCXXCompiler.cmake
> > ...
> >
> > Have set CYGWIN=1, CMAKE_INCLUDE_PATH=/usr/share/cmake-2.6.2/include,  
> > CMAKE_MODULE_PATH=/usr/share/cmake-2.6.2/Modules, still no dice. In
> > /usr/share/cmake-2.6.2/Modules, there exists CMakeCCompiler.cmake.in and  
> > CMakeCXXCompiler.cmake.in, but not
> > plain .cmake files.
> >
> > Any tips appreciated-
> > cheers Mark
> >
> > _______________________________________________
> > BioLib-dev mailing list
> > BioLib-dev at lists.open-bio.org
> > http://lists.open-bio.org/mailman/listinfo/biolib-dev
> >
> _______________________________________________
> BioLib-dev mailing list
> BioLib-dev at lists.open-bio.org
> http://lists.open-bio.org/mailman/listinfo/biolib-dev

From maj at fortinbras.us  Sat May  2 18:32:22 2009
From: maj at fortinbras.us (Mark A. Jensen)
Date: Sat, 2 May 2009 18:32:22 -0400
Subject: [BioLib-dev] installation on cygwin
In-Reply-To: <20090502185139.GA15817@thebird.nl>
References: <40844D11A9E84D05BBCE66BE6C76223B@NewLife><20090502103830.GA10664@thebird.nl>
	<20090502185139.GA15817@thebird.nl>
Message-ID: <97AB226A4E2C48D18943FD85CDC00236@NewLife>

Hey Pjotr--
I found this which got me quite a long ways --

this is key:
>From KDE-Windows:
>The thing is, cmake separates between "ARCHIVE", "LIBRARY" and "RUNTIME"
>components of targets.
>On Windows, the dll-file is considered a "RUNTIME" component, since it
> should go into the same directory as the executable, and the .lib file is
> considered an ARCHIVE component. On UNIX the .so file is a LIBRARY target,
> a .a file is (obviously) an ARCHIVE target.
>So if you say
>install(TARGETS mylib LIBRARY DESTINATION lib)
>under Windows then you didn't specifiy the destination directories for the
> dll and the lib file, that's why cmake complains.
>So you should do:
>install(TARGETS mylib LIBRARY DESTINATION lib${LIB_SUFFIX}
>                     ARCHIVE DESTINATION lib${LIB_SUFFIX}
>                     RUNTIME DESTINATION bin )
>
>Then it will do the right thing both on Windows and UNIX. If it's inside a
> KDE project, you can also use
>install(TARGETS mylib ${INSTALL_TARGETS_DEFAULT_ARGS} )
>which is a variable set in FindKDE4Internal.cmake, which basically contains
>just these settings.
>
>The simple way would be
>install(TARGETS mylib DESTINATION lib)
>but then Windows users may complain, because the dll won't be in the same
>directory as executables, so they'll have top adjust the PATH to make the
>executables run.

Not quite sure of the source of the original source of this tip (it was quoted 
where I found it), but doing

install ( TARGETS mylib
            LIBRARY DESTINATION lib
            RUNTIME DESTINATION bin )

in all the relevant CMakeLists.txt puts .dlls in the right place--
cheers- MAJ
----- Original Message ----- 
From: "Pjotr Prins" <pjotr.public14 at thebird.nl>
To: "BioLib Project" <biolib-dev at lists.open-bio.org>
Sent: Saturday, May 02, 2009 2:51 PM
Subject: Re: [BioLib-dev] installation on cygwin


>I have merged the BioLib cygwin branch into master - preparing for a
> new release. The major change is the use of modified cmake modules
> which are shared with ASciLib.
>
> The Cygwin tests are still broken. Dynaload.pm complains it can't find
> shared libraries. I have to tell cmake how to find the Cygwin shared
> libs - either due to the different naming convention (.dll) or because
> of some missing linker switches exporting the correct bindings.
>
> It is just a matter of poking around.
>
> Pj.
>
> On Sat, May 02, 2009 at 12:38:30PM +0200, Pjotr Prins wrote:
>> If have cygwin running and a version of example and staden that
>> builds for Perl. Only problem is the test fails to find the dynamic
>> library. I'll fix that soon. Checkout the cygwin branch:
>>
>>   cd biolib
>>   git checkout -b cygwin     # create new branch
>>   git remote add github git://github.com/pjotrp/biolib.git
>>   git pull github cygwin     # pull my branch
>>   ./configure --with-example
>>   make
>>   make test
>>   ./configure --with-staden
>>   make
>>   make test
>>
>> The latest version of biolib has moved the CMake modules into a
>> separate repository - that is shared with ASciLib.
>>
>> The problem you were seeing actually had to do with missing
>> submodules for git. This should be pulled in with configure:
>>
>>   git submodule init
>>   git submodule update
>>
>> Prerequisites for Cygwin are:
>>
>> The BioLib build system supports building on Cygwin
>> (http://www.cygwin.com/). First install Cygwin on your machine
>> selecting the following packages:
>>
>> * cmake   (development)
>> * gcc-g++ (development)
>> * make    (development)
>> * swig    (development)
>> * your preferred language: perl, python and/or ruby
>>
>> and, optionally:
>>
>> * git (to fetch the latest repositories)
>>
>> for ASciLib:
>>
>> * autoconf (development)
>> * automake (development)
>>
>> NOTE: The AffyIO module does not yet work on Cygwin - there is a
>> dependency on the R libraries. I'll sort that soon.
>>
>> See also ./doc/WINDOWS.txt for the latest.
>>
>> On Fri, May 01, 2009 at 02:11:49PM -0400, Mark A. Jensen wrote:
>> > Hey guys,
>> > Am trying to build on cygwin (I know, I know...). Have got cmake. I do
>> >
>> > $ cmake -DBUILD_PERL:BOOLEAN=TRUE .
>> >
>> > in the install directory (biolib-0.0.3), and receive
>> >
>> > -- The C compiler identification is GNU
>> > CMake Error: Could not find cmake module
>> > file:/cygdrive/c/....../biolib-0.0.3/CMakeFiles/CMakeCCompiler.cmake
>> > -- The CXX compiler identification is GNU
>> > CMake Error: Could not find cmake module
>> > file:/cygdrive/c/...../biolib-0.0.3/CMakeFiles/CMakeCXXCompiler.cmake
>> > ...
>> >
>> > Have set CYGWIN=1, CMAKE_INCLUDE_PATH=/usr/share/cmake-2.6.2/include,
>> > CMAKE_MODULE_PATH=/usr/share/cmake-2.6.2/Modules, still no dice. In
>> > /usr/share/cmake-2.6.2/Modules, there exists CMakeCCompiler.cmake.in and
>> > CMakeCXXCompiler.cmake.in, but not
>> > plain .cmake files.
>> >
>> > Any tips appreciated-
>> > cheers Mark
>> >
>> > _______________________________________________
>> > BioLib-dev mailing list
>> > BioLib-dev at lists.open-bio.org
>> > http://lists.open-bio.org/mailman/listinfo/biolib-dev
>> >
>> _______________________________________________
>> BioLib-dev mailing list
>> BioLib-dev at lists.open-bio.org
>> http://lists.open-bio.org/mailman/listinfo/biolib-dev
> _______________________________________________
> BioLib-dev mailing list
> BioLib-dev at lists.open-bio.org
> http://lists.open-bio.org/mailman/listinfo/biolib-dev
>
> 


From pjotr.public14 at thebird.nl  Mon May  4 08:18:18 2009
From: pjotr.public14 at thebird.nl (Pjotr Prins)
Date: Mon, 4 May 2009 14:18:18 +0200
Subject: [BioLib-dev] installation on cygwin
In-Reply-To: <40844D11A9E84D05BBCE66BE6C76223B@NewLife>
References: <40844D11A9E84D05BBCE66BE6C76223B@NewLife>
Message-ID: <20090504121818.GA2606@thebird.nl>

On Fri, May 01, 2009 at 02:11:49PM -0400, Mark A. Jensen wrote:
> Hey guys,
> Am trying to build on cygwin (I know, I know...). Have got cmake. I do

I am adding automated testing for BioLib. Here the current results
for Cygwin. You can see Ruby and Python are fine. Perl segfaults
after loading the DLL's.

Pj.

BioLib
0.0.4rc1
CYGWIN_NT-5.1 1.5.25(0.156/4/2) 2008-06-12 19:34
cmake version 2.6-patch 2

SWIG Version 1.3.36

Compiled with g++ [i686-pc-cygwin]

commit 2b86377a107be1443ef579cb54943658e0b91cdc
Author: pjotrp <pjotr.public01 at thebird.nl>
Date:   Mon May 4 13:42:40 2009 +0200

    Fix Perl test to find DLL
biolib git at github.com:pjotrp/biolib.git

This is perl, v5.10.0 built for cygwin-thread-multi-64int
(with 6 registered patches, see perl -V for more detail)

Running tests...
Start processing tests
Test project /home/xxx/autotest/biolib
  1/  2 Testing TestPerlExample               ***Exception: SegFault
  2/  2 Testing TestPerlStadenIOLib           ***Exception: SegFault

0% tests passed, 2 tests failed out of 2
	  1 - TestPerlExample (SEGFAULT)
	  2 - TestPerlStadenIOLib (SEGFAULT)
ruby 1.8.7 (2008-08-11 patchlevel 72) [i386-cygwin]
Running tests...
Start processing tests
Test project /home/xxx/autotest/biolib
  1/  2 Testing TestRubyExample                  Passed
  2/  2 Testing TestRubyStadenIOLib              Passed

100% tests passed, 0 tests failed out of 2
Running tests...
Start processing tests
Test project /home/xxx/autotest/biolib
  1/  2 Testing TestPythonExample                Passed
  2/  2 Testing TestPythonStadenIOLib            Passed

100% tests passed, 0 tests failed out of 2

From maj at fortinbras.us  Tue May  5 08:17:29 2009
From: maj at fortinbras.us (Mark A. Jensen)
Date: Tue, 5 May 2009 08:17:29 -0400
Subject: [BioLib-dev] installation on cygwin
In-Reply-To: <20090504121818.GA2606@thebird.nl>
References: <40844D11A9E84D05BBCE66BE6C76223B@NewLife>
	<20090504121818.GA2606@thebird.nl>
Message-ID: <70BB7AD164D44BAB8CEAB4FF24CB6E6F@NewLife>

The latest configured and made fine; I also get the segfaults. Will 
investigate--cheers
----- Original Message ----- 
From: "Pjotr Prins" <pjotr.public14 at thebird.nl>
To: "Mark A. Jensen" <maj at fortinbras.us>
Cc: "BioLib Project" <biolib-dev at lists.open-bio.org>
Sent: Monday, May 04, 2009 8:18 AM
Subject: Re: [BioLib-dev] installation on cygwin


> On Fri, May 01, 2009 at 02:11:49PM -0400, Mark A. Jensen wrote:
>> Hey guys,
>> Am trying to build on cygwin (I know, I know...). Have got cmake. I do
>
> I am adding automated testing for BioLib. Here the current results
> for Cygwin. You can see Ruby and Python are fine. Perl segfaults
> after loading the DLL's.
>
> Pj.
>
> BioLib
> 0.0.4rc1
> CYGWIN_NT-5.1 1.5.25(0.156/4/2) 2008-06-12 19:34
> cmake version 2.6-patch 2
>
> SWIG Version 1.3.36
>
> Compiled with g++ [i686-pc-cygwin]
>
> commit 2b86377a107be1443ef579cb54943658e0b91cdc
> Author: pjotrp <pjotr.public01 at thebird.nl>
> Date:   Mon May 4 13:42:40 2009 +0200
>
>    Fix Perl test to find DLL
> biolib git at github.com:pjotrp/biolib.git
>
> This is perl, v5.10.0 built for cygwin-thread-multi-64int
> (with 6 registered patches, see perl -V for more detail)
>
> Running tests...
> Start processing tests
> Test project /home/xxx/autotest/biolib
>  1/  2 Testing TestPerlExample               ***Exception: SegFault
>  2/  2 Testing TestPerlStadenIOLib           ***Exception: SegFault
>
> 0% tests passed, 2 tests failed out of 2
>   1 - TestPerlExample (SEGFAULT)
>   2 - TestPerlStadenIOLib (SEGFAULT)
> ruby 1.8.7 (2008-08-11 patchlevel 72) [i386-cygwin]
> Running tests...
> Start processing tests
> Test project /home/xxx/autotest/biolib
>  1/  2 Testing TestRubyExample                  Passed
>  2/  2 Testing TestRubyStadenIOLib              Passed
>
> 100% tests passed, 0 tests failed out of 2
> Running tests...
> Start processing tests
> Test project /home/xxx/autotest/biolib
>  1/  2 Testing TestPythonExample                Passed
>  2/  2 Testing TestPythonStadenIOLib            Passed
>
> 100% tests passed, 0 tests failed out of 2
>
> 


From pjotr.public14 at thebird.nl  Tue May  5 17:16:19 2009
From: pjotr.public14 at thebird.nl (Pjotr Prins)
Date: Tue, 5 May 2009 23:16:19 +0200
Subject: [BioLib-dev] 0.0.4 release
Message-ID: <20090505211619.GA15941@thebird.nl>

BioLib is getting adolescent. With automated integration testing and a
build system I am happy with. 

Cygwin is supported - though the Perl tests are still broken (Python
and Ruby are fine). It shouldn't be too hard to also support native
Windows Perl etc.

On Linux everything should work. With OS/X Perl Staden works.

Download 0.0.4 from http://bio3.biobeat.org/download/nix-biology/ or 
preferably:

  git clone git://github.com/pjotrp/biolib.git

---------------------------------------------------------------------
BIOLIB 0.04  a3de2f3619d05b613dc15280773c341bf3a33267     May 5, 2009
---------------------------------------------------------------------

- Full Perl, Python and Ruby support
- Cygwin support (Microsoft Windows) - except for Affyio as it
  requires Rlib. Ruby and Python are supported. Perl segfaults.
- Moved cmake modules into own git repository (for sharing with
  ASciLib) - see http://github.com/pjotrp/cmake-support/
- Automated testing for different setups introduced
- Generate XML docs from SWIG
- Added R/QTL module


From xshuai at umail.iu.edu  Thu May 14 00:22:10 2009
From: xshuai at umail.iu.edu (Xin Shuai)
Date: Thu, 14 May 2009 00:22:10 -0400
Subject: [BioLib-dev] CMakeLists.txt
Message-ID: <3a7743460905132122q4013614fh10c7db434cb36d25@mail.gmail.com>

HI! It seems that writing CMakeLists.txt is a big job in BioLib. And I have
create a CMakeLists.txt file for each directory. Do I have to specially
study related command of writing a CMakeLIsts.txt file?

-- 
Xin Shuai (David)
PhD of Complex System in School of Informatics
Indiana University Bloomington
812-606-8019

From pjotr.public14 at thebird.nl  Thu May 14 02:03:56 2009
From: pjotr.public14 at thebird.nl (Pjotr Prins)
Date: Thu, 14 May 2009 08:03:56 +0200
Subject: [BioLib-dev] CMakeLists.txt
In-Reply-To: <3a7743460905132122q4013614fh10c7db434cb36d25@mail.gmail.com>
References: <3a7743460905132122q4013614fh10c7db434cb36d25@mail.gmail.com>
Message-ID: <20090514060356.GC4164@thebird.nl>

CMake is the back bone of biolib sorting out all dependencies and
locations. Fortunately CMake is very straightforward, it is really a
little programming language (macro language) that generates the make
files. Rather like the often used autoconf tools (i.e. 'configure').
Unlike autoconf it is easy to understand and rather nice at status
messages. I have learned to love it, despite the fact it is another
little language to learn.

The most important thing to remember is that when you change a CMake
file you have to remove the cache files. CMake caches all settings -
probably necessary for big build systems like KDE.

Initially just copy the CMakeLists.txt file from other places and
adapt it to your tree. If there is 'deeper' functionality required I can
always help get it to work.

Read the CMake online documentation (take an hour, and the FAQ). It'll
pay off.

Pj.

On Thu, May 14, 2009 at 12:22:10AM -0400, Xin Shuai wrote:
> HI! It seems that writing CMakeLists.txt is a big job in BioLib. And I have
> create a CMakeLists.txt file for each directory. Do I have to specially
> study related command of writing a CMakeLIsts.txt file?
> 
> -- 
> Xin Shuai (David)
> PhD of Complex System in School of Informatics
> Indiana University Bloomington
> 812-606-8019
> _______________________________________________
> BioLib-dev mailing list
> BioLib-dev at lists.open-bio.org
> http://lists.open-bio.org/mailman/listinfo/biolib-dev
> 

From cjfields at illinois.edu  Thu May 14 17:14:49 2009
From: cjfields at illinois.edu (Chris Fields)
Date: Thu, 14 May 2009 16:14:49 -0500
Subject: [BioLib-dev] [ANNOUNCEMENT] Google Summer of Code student Xin Shuai
Message-ID: <7A388EB0-E9B2-4579-80EA-83AC95817EF9@illinois.edu>

All,

I am proud to introduce Xin 'David' Shuai, my student for the Google  
Summer of Code 2009, to the Open Bioinformatics community.  David's  
project centers on developing SWIG-based bindings to libsequence (a  
population genetics library) for the BioLib project:

http://biolib.open-bio.org/wiki/Main_Page

Besides myself, David will be co-mentored by Mark Jensen and Pjotr  
Prins.

As the BioLib project centers on creating common, maintainable SWIG- 
based bindings to popular bioinformatics libraries for the various  
Bio* toolkits, we will likely need input from the various Open Bio  
communities at various stages in the project.  At this time, David's  
initial plans are to develop and test libsequence bindings for Perl  
and Python.

David's proposal and project plan are available here:

http://biolib.open-bio.org/wiki/User:David

Congratulations David, and welcome to the Open-Bio community!

Sincerely,

Christopher Fields
University of Illinois Urbana-Champaign
Institute for Genomic Biology
Urbana, IL 61801

From hlapp at gmx.net  Thu May 14 18:16:30 2009
From: hlapp at gmx.net (Hilmar Lapp)
Date: Thu, 14 May 2009 18:16:30 -0400
Subject: [BioLib-dev] [Bioperl-l] [ANNOUNCEMENT] Google Summer of Code
	student Xin Shuai
In-Reply-To: <7A388EB0-E9B2-4579-80EA-83AC95817EF9@illinois.edu>
References: <7A388EB0-E9B2-4579-80EA-83AC95817EF9@illinois.edu>
Message-ID: <4482F182-B0B0-4C71-B8DE-B5B7A4EC4D81@gmx.net>

Welcome David, good luck with your project, and I hope (actually, am  
quite certain) that you'll enjoy your summer with us.

	-hilmar

On May 14, 2009, at 5:14 PM, Chris Fields wrote:

> All,
>
> I am proud to introduce Xin 'David' Shuai, my student for the Google  
> Summer of Code 2009, to the Open Bioinformatics community.  David's  
> project centers on developing SWIG-based bindings to libsequence (a  
> population genetics library) for the BioLib project:
>
> http://biolib.open-bio.org/wiki/Main_Page
>
> Besides myself, David will be co-mentored by Mark Jensen and Pjotr  
> Prins.
>
> As the BioLib project centers on creating common, maintainable SWIG- 
> based bindings to popular bioinformatics libraries for the various  
> Bio* toolkits, we will likely need input from the various Open Bio  
> communities at various stages in the project.  At this time, David's  
> initial plans are to develop and test libsequence bindings for Perl  
> and Python.
>
> David's proposal and project plan are available here:
>
> http://biolib.open-bio.org/wiki/User:David
>
> Congratulations David, and welcome to the Open-Bio community!
>
> Sincerely,
>
> Christopher Fields
> University of Illinois Urbana-Champaign
> Institute for Genomic Biology
> Urbana, IL 61801
> _______________________________________________
> Bioperl-l mailing list
> Bioperl-l at lists.open-bio.org
> http://lists.open-bio.org/mailman/listinfo/bioperl-l

-- 
===========================================================
: Hilmar Lapp  -:-  Durham, NC  -:-  hlapp at gmx dot net :
===========================================================




From xshuai at umail.iu.edu  Thu May 14 21:52:40 2009
From: xshuai at umail.iu.edu (Xin Shuai)
Date: Thu, 14 May 2009 21:52:40 -0400
Subject: [BioLib-dev] [Bioperl-l] [ANNOUNCEMENT] Google Summer of Code
	student Xin Shuai
In-Reply-To: <4482F182-B0B0-4C71-B8DE-B5B7A4EC4D81@gmx.net>
References: <7A388EB0-E9B2-4579-80EA-83AC95817EF9@illinois.edu>
	<4482F182-B0B0-4C71-B8DE-B5B7A4EC4D81@gmx.net>
Message-ID: <3a7743460905141852x69e32266oc440e7557d59d84f@mail.gmail.com>

Thank you for everyone's help during my application. I will do my best to
accomplish it.Since I'm a totally newcomer in Bio* language program, I will
have a lot to ask during the following months and hope to get your help!

On Thu, May 14, 2009 at 6:16 PM, Hilmar Lapp <hlapp at gmx.net> wrote:

> Welcome David, good luck with your project, and I hope (actually, am quite
> certain) that you'll enjoy your summer with us.
>
>        -hilmar
>
>
> On May 14, 2009, at 5:14 PM, Chris Fields wrote:
>
>  All,
>>
>> I am proud to introduce Xin 'David' Shuai, my student for the Google
>> Summer of Code 2009, to the Open Bioinformatics community.  David's project
>> centers on developing SWIG-based bindings to libsequence (a population
>> genetics library) for the BioLib project:
>>
>> http://biolib.open-bio.org/wiki/Main_Page
>>
>> Besides myself, David will be co-mentored by Mark Jensen and Pjotr Prins.
>>
>> As the BioLib project centers on creating common, maintainable SWIG-based
>> bindings to popular bioinformatics libraries for the various Bio* toolkits,
>> we will likely need input from the various Open Bio communities at various
>> stages in the project.  At this time, David's initial plans are to develop
>> and test libsequence bindings for Perl and Python.
>>
>> David's proposal and project plan are available here:
>>
>> http://biolib.open-bio.org/wiki/User:David
>>
>> Congratulations David, and welcome to the Open-Bio community!
>>
>> Sincerely,
>>
>> Christopher Fields
>> University of Illinois Urbana-Champaign
>> Institute for Genomic Biology
>> Urbana, IL 61801
>> _______________________________________________
>> Bioperl-l mailing list
>> Bioperl-l at lists.open-bio.org
>> http://lists.open-bio.org/mailman/listinfo/bioperl-l
>>
>
> --
> ===========================================================
> : Hilmar Lapp  -:-  Durham, NC  -:-  hlapp at gmx dot net :
> ===========================================================
>
>
>
>


-- 
Xin Shuai (David)
PhD of Complex System in School of Informatics
Indiana University Bloomington
812-606-8019

From xshuai at umail.iu.edu  Thu May 14 23:57:30 2009
From: xshuai at umail.iu.edu (Xin Shuai)
Date: Thu, 14 May 2009 23:57:30 -0400
Subject: [BioLib-dev] Where can I see "swig"
Message-ID: <3a7743460905142057j6d87070w83d5098e20a656ce@mail.gmail.com>

I tried to build the example module with python bindings and the result
seems fine.However, I feel a little puzzled that where does "swig" take
effect in the whole process?
In other words, in which file can I find the command that is related to
using "swig"?

-- 
Xin Shuai (David)
PhD of Complex System in School of Informatics
Indiana University Bloomington
812-606-8019

From maj at fortinbras.us  Fri May 15 00:11:15 2009
From: maj at fortinbras.us (Mark A. Jensen)
Date: Fri, 15 May 2009 00:11:15 -0400
Subject: [BioLib-dev] Where can I see "swig"
In-Reply-To: <3a7743460905142057j6d87070w83d5098e20a656ce@mail.gmail.com>
References: <3a7743460905142057j6d87070w83d5098e20a656ce@mail.gmail.com>
Message-ID: <D9D0A7C2077645269E3ED7E4694918AA@NewLife>

Hey David,
look in [install dir]/src/mappings/swig; the interface files (which swig uses to 
create the bindings) are there for all the mappings. Then you can go down to 
python/example/CMakeLists.txt to see cmake routines that are running swig. The 
lines that do the work are (Pj. will correct me if I'm wrong)...

 INCLUDE(${SWIG_USE_FILE})

 SET(SWIG_FLAGS "")
 SET_SOURCE_FILES_PROPERTIES(${INTERFACE} PROPERTIES SWIG_FLAGS "-includeall")

 SWIG_ADD_MODULE(${MY_MODULE} python ${INTERFACE} ${SOURCES} )
 SWIG_LINK_LIBRARIES(${MY_MODULE} ${PYTHON_LIBRARIES} )

Mark

----- Original Message ----- 
From: "Xin Shuai" <xshuai at umail.iu.edu>
To: "BioLib Project" <biolib-dev at lists.open-bio.org>
Sent: Thursday, May 14, 2009 11:57 PM
Subject: [BioLib-dev] Where can I see "swig"


>I tried to build the example module with python bindings and the result
> seems fine.However, I feel a little puzzled that where does "swig" take
> effect in the whole process?
> In other words, in which file can I find the command that is related to
> using "swig"?
>
> -- 
> Xin Shuai (David)
> PhD of Complex System in School of Informatics
> Indiana University Bloomington
> 812-606-8019
> _______________________________________________
> BioLib-dev mailing list
> BioLib-dev at lists.open-bio.org
> http://lists.open-bio.org/mailman/listinfo/biolib-dev
>
> 


From xshuai at umail.iu.edu  Fri May 15 01:29:04 2009
From: xshuai at umail.iu.edu (Xin Shuai)
Date: Fri, 15 May 2009 01:29:04 -0400
Subject: [BioLib-dev] Where can I see "swig"
In-Reply-To: <D9D0A7C2077645269E3ED7E4694918AA@NewLife>
References: <3a7743460905142057j6d87070w83d5098e20a656ce@mail.gmail.com>
	<D9D0A7C2077645269E3ED7E4694918AA@NewLife>
Message-ID: <3a7743460905142229t402757c6w7410ea784a2f4d4d@mail.gmail.com>

Hi, Mark! I'm afraid I cannot see as you post it in
python/example/CMakeLists.txt. The following is what I can see in that file:#
Builds the Python biolib_example library after resolving paths.

cmake_minimum_required(VERSION 2.6)

ENABLE_TESTING()

SET (M_NAME example)
SET (M_VERSION 1.0)
SET (INTERFACE python_${M_NAME}.i)

IF(NOT BUILD_LIBS)
  SET (MAP_ROOT ../../../../..)
  SET (CMAKE_MODULE_PATH ${MAP_ROOT}/tools/cmake-support/modules)
ENDIF(NOT BUILD_LIBS)

SET(CMAKE_LIBRARY_OUTPUT_DIRECTORY biolib)

FIND_PACKAGE(MapPython REQUIRED)

POST_BUILD_PYTHON_BINDINGS()
TEST_PYTHON_BINDINGS()
INSTALL_PYTHON_BINDINGS()

I still cannot see any "swig" related command.

On Fri, May 15, 2009 at 12:11 AM, Mark A. Jensen <maj at fortinbras.us> wrote:

> Hey David,
> look in [install dir]/src/mappings/swig; the interface files (which swig
> uses to create the bindings) are there for all the mappings. Then you can go
> down to python/example/CMakeLists.txt to see cmake routines that are running
> swig. The lines that do the work are (Pj. will correct me if I'm wrong)...
>
> INCLUDE(${SWIG_USE_FILE})
>
> SET(SWIG_FLAGS "")
> SET_SOURCE_FILES_PROPERTIES(${INTERFACE} PROPERTIES SWIG_FLAGS
> "-includeall")
>
> SWIG_ADD_MODULE(${MY_MODULE} python ${INTERFACE} ${SOURCES} )
> SWIG_LINK_LIBRARIES(${MY_MODULE} ${PYTHON_LIBRARIES} )
>
> Mark
>
> ----- Original Message ----- From: "Xin Shuai" <xshuai at umail.iu.edu>
> To: "BioLib Project" <biolib-dev at lists.open-bio.org>
> Sent: Thursday, May 14, 2009 11:57 PM
> Subject: [BioLib-dev] Where can I see "swig"
>
>
>  I tried to build the example module with python bindings and the result
>> seems fine.However, I feel a little puzzled that where does "swig" take
>> effect in the whole process?
>> In other words, in which file can I find the command that is related to
>> using "swig"?
>>
>> --
>> Xin Shuai (David)
>> PhD of Complex System in School of Informatics
>> Indiana University Bloomington
>> 812-606-8019
>> _______________________________________________
>> BioLib-dev mailing list
>> BioLib-dev at lists.open-bio.org
>> http://lists.open-bio.org/mailman/listinfo/biolib-dev
>>
>>
>>
>


-- 
Xin Shuai (David)
PhD of Complex System in School of Informatics
Indiana University Bloomington
812-606-8019

From pjotr.public14 at thebird.nl  Fri May 15 01:52:28 2009
From: pjotr.public14 at thebird.nl (Pjotr Prins)
Date: Fri, 15 May 2009 07:52:28 +0200
Subject: [BioLib-dev] Where can I see "swig"
In-Reply-To: <3a7743460905142229t402757c6w7410ea784a2f4d4d@mail.gmail.com>
References: <3a7743460905142057j6d87070w83d5098e20a656ce@mail.gmail.com>
	<D9D0A7C2077645269E3ED7E4694918AA@NewLife>
	<3a7743460905142229t402757c6w7410ea784a2f4d4d@mail.gmail.com>
Message-ID: <20090515055228.GA15185@thebird.nl>

On Fri, May 15, 2009 at 01:29:04AM -0400, Xin Shuai wrote:
> POST_BUILD_PYTHON_BINDINGS()
> TEST_PYTHON_BINDINGS()
> INSTALL_PYTHON_BINDINGS()
> 
> I still cannot see any "swig" related command.

Above are macros defined in ./tools/cmake-support/modules/. They,
indeed, do the hard work of managing SWIG for Python. The command:

  FIND_PACKAGE(MapPython REQUIRED)

loads the relevant cmake files.

Pj.


From pjotr.public14 at thebird.nl  Fri May 15 02:15:01 2009
From: pjotr.public14 at thebird.nl (Pjotr Prins)
Date: Fri, 15 May 2009 08:15:01 +0200
Subject: [BioLib-dev] BioLib Talk at BOSC 2009
Message-ID: <20090515061501.GA16563@thebird.nl>

I will give a talk at BOSC 2009 on the BioLib project, the abstract
was accepted. I hope to see many of you :-). I also requested a Birds
of a Feather session on incorporating old and new C libraries into
BioLib. If that fails we will do something after hours.

Pjotr.

From pjotr.public14 at thebird.nl  Fri May 15 02:18:27 2009
From: pjotr.public14 at thebird.nl (Pjotr Prins)
Date: Fri, 15 May 2009 08:18:27 +0200
Subject: [BioLib-dev] [ANNOUNCEMENT] Google Summer of Code student Adam Smith
In-Reply-To: <20090515061051.GA15908@thebird.nl>
References: <20090515061051.GA15908@thebird.nl>
Message-ID: <20090515061827.GA16958@thebird.nl>

Hi Everyone,

I am glad to announce the participation of Adam Smith in BioLib for
the Google Summer of Code 2009. Adam will work on embedding Bio++ into
BioLib, which news has been well received by both projects (Julien is
the project leader for Bio++).  Bio++ is a set of C++ libraries for
Bioinformatics, including sequence analysis, phylogenetics, molecular
evolution and population genetics. Adam's project is described here:

https://www.nescent.org/wg/phyloinformatics/index.php?title=Phyloinformatics_Summer_of_Code_2009#Mapping_the_Bio.2B.2B_Phylogenetics_toolkit_to_BioPerl.2C_R.2FBioConductor_and_BioJAVA_using_BioLib

Adam has defended his Ph.D. this week, and from now on he can work
fully on this project over the summer. Adam's home page is here:

  http://pages.cs.wisc.edu/~aasmith/

and will announce on this list when he has updated his project page.

He will also give a talk at ISMB, so we'll get a chance to meet in
person.

Pjotr.

From xshuai at umail.iu.edu  Fri May 15 11:53:28 2009
From: xshuai at umail.iu.edu (Xin Shuai)
Date: Fri, 15 May 2009 11:53:28 -0400
Subject: [BioLib-dev] Where can I see "swig"
In-Reply-To: <20090515055228.GA15185@thebird.nl>
References: <3a7743460905142057j6d87070w83d5098e20a656ce@mail.gmail.com>
	<D9D0A7C2077645269E3ED7E4694918AA@NewLife>
	<3a7743460905142229t402757c6w7410ea784a2f4d4d@mail.gmail.com>
	<20090515055228.GA15185@thebird.nl>
Message-ID: <3a7743460905150853v690fd8e8g8a81ac0553d85d8@mail.gmail.com>

It seems like those .cmake files in ./tools/cmake-support/modules are all
created automatically by CMake and I don't have to care about it. However, I
should specify in CMakeLists.txt that how to create those macros, is that
right?
By the way, at this moment I can only build modules in BioLib one by one
using
 ./configure command  --with-(module name)
Since when I want to configure the whole modules I was always told that
R_LIBRARY is not found. Is it possible for me to avoid such problem but
still configure the whole modules?

On Fri, May 15, 2009 at 1:52 AM, Pjotr Prins <pjotr.public14 at thebird.nl>wrote:

> On Fri, May 15, 2009 at 01:29:04AM -0400, Xin Shuai wrote:
> > POST_BUILD_PYTHON_BINDINGS()
> > TEST_PYTHON_BINDINGS()
> > INSTALL_PYTHON_BINDINGS()
> >
> > I still cannot see any "swig" related command.
>
> Above are macros defined in ./tools/cmake-support/modules/. They,
> indeed, do the hard work of managing SWIG for Python. The command:
>
>  FIND_PACKAGE(MapPython REQUIRED)
>
> loads the relevant cmake files.
>
> Pj.
>
>


-- 
Xin Shuai (David)
PhD of Complex System in School of Informatics
Indiana University Bloomington
812-606-8019

From pjotr.public14 at thebird.nl  Fri May 15 15:28:39 2009
From: pjotr.public14 at thebird.nl (Pjotr Prins)
Date: Fri, 15 May 2009 21:28:39 +0200
Subject: [BioLib-dev] Where can I see "swig"
In-Reply-To: <3a7743460905150853v690fd8e8g8a81ac0553d85d8@mail.gmail.com>
References: <3a7743460905142057j6d87070w83d5098e20a656ce@mail.gmail.com>
	<D9D0A7C2077645269E3ED7E4694918AA@NewLife>
	<3a7743460905142229t402757c6w7410ea784a2f4d4d@mail.gmail.com>
	<20090515055228.GA15185@thebird.nl>
	<3a7743460905150853v690fd8e8g8a81ac0553d85d8@mail.gmail.com>
Message-ID: <20090515192839.GD25030@thebird.nl>

BioLib has a modular setup. Which is what it should be. Only the
Affyio library requires R, and on some platforms I have not fixed
finding them. It will come soon.

Pj.

On Fri, May 15, 2009 at 11:53:28AM -0400, Xin Shuai wrote:
> It seems like those .cmake files in ./tools/cmake-support/modules are all
> created automatically by CMake and I don't have to care about it. However, I
> should specify in CMakeLists.txt that how to create those macros, is that
> right?
> By the way, at this moment I can only build modules in BioLib one by one
> using
>  ./configure command  --with-(module name)
> Since when I want to configure the whole modules I was always told that
> R_LIBRARY is not found. Is it possible for me to avoid such problem but
> still configure the whole modules?
> 
> On Fri, May 15, 2009 at 1:52 AM, Pjotr Prins <pjotr.public14 at thebird.nl>wrote:
> 
> > On Fri, May 15, 2009 at 01:29:04AM -0400, Xin Shuai wrote:
> > > POST_BUILD_PYTHON_BINDINGS()
> > > TEST_PYTHON_BINDINGS()
> > > INSTALL_PYTHON_BINDINGS()
> > >
> > > I still cannot see any "swig" related command.
> >
> > Above are macros defined in ./tools/cmake-support/modules/. They,
> > indeed, do the hard work of managing SWIG for Python. The command:
> >
> >  FIND_PACKAGE(MapPython REQUIRED)
> >
> > loads the relevant cmake files.
> >
> > Pj.
> >
> >
> 
> 
> -- 
> Xin Shuai (David)
> PhD of Complex System in School of Informatics
> Indiana University Bloomington
> 812-606-8019

From xshuai at umail.iu.edu  Fri May 15 16:09:53 2009
From: xshuai at umail.iu.edu (Xin Shuai)
Date: Fri, 15 May 2009 16:09:53 -0400
Subject: [BioLib-dev] Where can I see "swig"
In-Reply-To: <20090515192839.GD25030@thebird.nl>
References: <3a7743460905142057j6d87070w83d5098e20a656ce@mail.gmail.com>
	<D9D0A7C2077645269E3ED7E4694918AA@NewLife>
	<3a7743460905142229t402757c6w7410ea784a2f4d4d@mail.gmail.com>
	<20090515055228.GA15185@thebird.nl>
	<3a7743460905150853v690fd8e8g8a81ac0553d85d8@mail.gmail.com>
	<20090515192839.GD25030@thebird.nl>
Message-ID: <3a7743460905151309r2f93d095xd8082c8bb2548c3c@mail.gmail.com>

I noticed that such form of file " _staden_io_lib.so"will be created
together with corresponding .py and .pyc files
this .so file is a type of binary file
What's the function of this file?

On Fri, May 15, 2009 at 3:28 PM, Pjotr Prins <pjotr.public14 at thebird.nl>wrote:

> BioLib has a modular setup. Which is what it should be. Only the
> Affyio library requires R, and on some platforms I have not fixed
> finding them. It will come soon.
>
> Pj.
>
> On Fri, May 15, 2009 at 11:53:28AM -0400, Xin Shuai wrote:
> > It seems like those .cmake files in ./tools/cmake-support/modules are all
> > created automatically by CMake and I don't have to care about it.
> However, I
> > should specify in CMakeLists.txt that how to create those macros, is that
> > right?
> > By the way, at this moment I can only build modules in BioLib one by one
> > using
> >  ./configure command  --with-(module name)
> > Since when I want to configure the whole modules I was always told that
> > R_LIBRARY is not found. Is it possible for me to avoid such problem but
> > still configure the whole modules?
> >
> > On Fri, May 15, 2009 at 1:52 AM, Pjotr Prins <pjotr.public14 at thebird.nl
> >wrote:
> >
> > > On Fri, May 15, 2009 at 01:29:04AM -0400, Xin Shuai wrote:
> > > > POST_BUILD_PYTHON_BINDINGS()
> > > > TEST_PYTHON_BINDINGS()
> > > > INSTALL_PYTHON_BINDINGS()
> > > >
> > > > I still cannot see any "swig" related command.
> > >
> > > Above are macros defined in ./tools/cmake-support/modules/. They,
> > > indeed, do the hard work of managing SWIG for Python. The command:
> > >
> > >  FIND_PACKAGE(MapPython REQUIRED)
> > >
> > > loads the relevant cmake files.
> > >
> > > Pj.
> > >
> > >
> >
> >
> > --
> > Xin Shuai (David)
> > PhD of Complex System in School of Informatics
> > Indiana University Bloomington
> > 812-606-8019
>



-- 
Xin Shuai (David)
PhD of Complex System in School of Informatics
Indiana University Bloomington
812-606-8019

From pjotr.public14 at thebird.nl  Fri May 15 19:09:27 2009
From: pjotr.public14 at thebird.nl (Pjotr Prins)
Date: Sat, 16 May 2009 01:09:27 +0200
Subject: [BioLib-dev] Where can I see "swig"
In-Reply-To: <3a7743460905151309r2f93d095xd8082c8bb2548c3c@mail.gmail.com>
References: <3a7743460905142057j6d87070w83d5098e20a656ce@mail.gmail.com>
	<D9D0A7C2077645269E3ED7E4694918AA@NewLife>
	<3a7743460905142229t402757c6w7410ea784a2f4d4d@mail.gmail.com>
	<20090515055228.GA15185@thebird.nl>
	<3a7743460905150853v690fd8e8g8a81ac0553d85d8@mail.gmail.com>
	<20090515192839.GD25030@thebird.nl>
	<3a7743460905151309r2f93d095xd8082c8bb2548c3c@mail.gmail.com>
Message-ID: <20090515230927.GA25644@thebird.nl>

On Fri, May 15, 2009 at 04:09:53PM -0400, Xin Shuai wrote:
> I noticed that such form of file " _staden_io_lib.so"will be created
> together with corresponding .py and .pyc files
> this .so file is a type of binary file
> What's the function of this file?

It contains Python specific bindings - the C counterpart of the
Python file(s). Some languages have both, Ruby has only a C lib.

The bindings handle the interface between the high level language and
the low level (generic) C libraries - like the functionality in
biolib-staden_io.so (which links against the bindings of all
languages and contains the actual functionality we aim to bind).

Hope that helps. You can try and read the C code.

Pj.

From xshuai at umail.iu.edu  Sat May 16 12:15:27 2009
From: xshuai at umail.iu.edu (Xin Shuai)
Date: Sat, 16 May 2009 12:15:27 -0400
Subject: [BioLib-dev] Submit my revision
Message-ID: <3a7743460905160915i3018728cgdd31881511a8c651@mail.gmail.com>

HI, Pjotr:  I forked your biolib repo and cloned my fork using "Your clone
URL"
  Then I made some a few changes and what to submit my revision.
  I went into the directory of changed "biolib" in my computer and then type
the following commands:


xin-shuais-macbook-2:biolib diavy$ git push origin masterEverything
up-to-date

However, when I checked my fork in github.com again and didn't see any
change at all. Why?


-- 
Xin Shuai (David)
PhD of Complex System in School of Informatics
Indiana University Bloomington
812-606-8019

From pjotr.public14 at thebird.nl  Sat May 16 14:18:55 2009
From: pjotr.public14 at thebird.nl (Pjotr Prins)
Date: Sat, 16 May 2009 20:18:55 +0200
Subject: [BioLib-dev] Submit my revision
In-Reply-To: <3a7743460905160915i3018728cgdd31881511a8c651@mail.gmail.com>
References: <3a7743460905160915i3018728cgdd31881511a8c651@mail.gmail.com>
Message-ID: <20090516181855.GA344@thebird.nl>

On Sat, May 16, 2009 at 12:15:27PM -0400, Xin Shuai wrote:
> HI, Pjotr:  I forked your biolib repo and cloned my fork using "Your clone
> URL"
>   Then I made some a few changes and what to submit my revision.
>   I went into the directory of changed "biolib" in my computer and then type
> the following commands:
> 
> 
> xin-shuais-macbook-2:biolib diavy$ git push origin masterEverything
> up-to-date
> 
> However, when I checked my fork in github.com again and didn't see any
> change at all. Why?

Did you commit your changes? 

man git-commit

And see the github howto.

Pj.

From aasmith at cs.wisc.edu  Thu May 21 03:05:50 2009
From: aasmith at cs.wisc.edu (Adam A Smith)
Date: Thu, 21 May 2009 02:05:50 -0500 (CDT)
Subject: [BioLib-dev] [ANNOUNCEMENT] Google Summer of Code student Adam
 Smith
In-Reply-To: <20090515061827.GA16958@thebird.nl>
References: <20090515061051.GA15908@thebird.nl>
	<20090515061827.GA16958@thebird.nl>
Message-ID: <Pine.LNX.4.64.0905210203360.13454@king06.cs.wisc.edu>

> and will announce on this list when he has updated his project page.

project page is now updated.

http://www.nescent.org/wg_phyloinformatics/PhyloSoC:Mapping_the_Bio%2B%2B_Phylogenetics_toolkit_to_R/BioConductor_and_BioJAVA_using_BioLib

sorry everyone for the delay in getting up & running.  had some important, 
unexpected family obligations that came up right as i was trying to 
defend.  but i'm here now.

Adam


>
> He will also give a talk at ISMB, so we'll get a chance to meet in
> person.
>
> Pjotr.
> _______________________________________________
> BioLib-dev mailing list
> BioLib-dev at lists.open-bio.org
> http://lists.open-bio.org/mailman/listinfo/biolib-dev
>

From pjotr.public14 at thebird.nl  Fri May 22 13:01:13 2009
From: pjotr.public14 at thebird.nl (Pjotr Prins)
Date: Fri, 22 May 2009 19:01:13 +0200
Subject: [BioLib-dev] Adding Bio++ to BioLib
Message-ID: <20090522170113.GA341@thebird.nl>

For Adam I have created a Bio++ git repository on github at:

  http://github.com/pjotrp/biopp/tree/master

And a special biopp branch of the Biolib repository which includes
Bio++ as a git submodule (in ./config/biopp), with one example build.

I'll do something similar for David.

Pj.

From maj at fortinbras.us  Fri May  1 18:11:49 2009
From: maj at fortinbras.us (Mark A. Jensen)
Date: Fri, 1 May 2009 14:11:49 -0400
Subject: [BioLib-dev] installation on cygwin
Message-ID: <40844D11A9E84D05BBCE66BE6C76223B@NewLife>

Hey guys,
Am trying to build on cygwin (I know, I know...). Have got cmake. I do

$ cmake -DBUILD_PERL:BOOLEAN=TRUE .

in the install directory (biolib-0.0.3), and receive

-- The C compiler identification is GNU
CMake Error: Could not find cmake module 
file:/cygdrive/c/....../biolib-0.0.3/CMakeFiles/CMakeCCompiler.cmake
-- The CXX compiler identification is GNU
CMake Error: Could not find cmake module 
file:/cygdrive/c/...../biolib-0.0.3/CMakeFiles/CMakeCXXCompiler.cmake
...

Have set CYGWIN=1, CMAKE_INCLUDE_PATH=/usr/share/cmake-2.6.2/include, 
CMAKE_MODULE_PATH=/usr/share/cmake-2.6.2/Modules, still no dice. In
/usr/share/cmake-2.6.2/Modules, there exists CMakeCCompiler.cmake.in and 
CMakeCXXCompiler.cmake.in, but not
plain .cmake files.

Any tips appreciated-
cheers Mark



From pjotr.public14 at thebird.nl  Fri May  1 20:46:12 2009
From: pjotr.public14 at thebird.nl (Pjotr Prins)
Date: Fri, 1 May 2009 22:46:12 +0200
Subject: [BioLib-dev] installation on cygwin
In-Reply-To: <40844D11A9E84D05BBCE66BE6C76223B@NewLife>
References: <40844D11A9E84D05BBCE66BE6C76223B@NewLife>
Message-ID: <20090501204612.GA6590@thebird.nl>

Live dangerously ;-). The problem is that the cmake files have only
been tried on Linux and OSX. Well, I guess I'll have to try myself. I have
Windows somewhere, it is on my list now. 

Pj.

On Fri, May 01, 2009 at 02:11:49PM -0400, Mark A. Jensen wrote:
> Hey guys,
> Am trying to build on cygwin (I know, I know...). Have got cmake. I do
>
> $ cmake -DBUILD_PERL:BOOLEAN=TRUE .
>
> in the install directory (biolib-0.0.3), and receive
>
> -- The C compiler identification is GNU
> CMake Error: Could not find cmake module  
> file:/cygdrive/c/....../biolib-0.0.3/CMakeFiles/CMakeCCompiler.cmake
> -- The CXX compiler identification is GNU
> CMake Error: Could not find cmake module  
> file:/cygdrive/c/...../biolib-0.0.3/CMakeFiles/CMakeCXXCompiler.cmake
> ...
>
> Have set CYGWIN=1, CMAKE_INCLUDE_PATH=/usr/share/cmake-2.6.2/include,  
> CMAKE_MODULE_PATH=/usr/share/cmake-2.6.2/Modules, still no dice. In
> /usr/share/cmake-2.6.2/Modules, there exists CMakeCCompiler.cmake.in and  
> CMakeCXXCompiler.cmake.in, but not
> plain .cmake files.
>
> Any tips appreciated-
> cheers Mark
>
> _______________________________________________
> BioLib-dev mailing list
> BioLib-dev at lists.open-bio.org
> http://lists.open-bio.org/mailman/listinfo/biolib-dev
>


From maj at fortinbras.us  Fri May  1 20:57:58 2009
From: maj at fortinbras.us (Mark A. Jensen)
Date: Fri, 1 May 2009 16:57:58 -0400
Subject: [BioLib-dev] installation on cygwin
In-Reply-To: <20090501204612.GA6590@thebird.nl>
References: <40844D11A9E84D05BBCE66BE6C76223B@NewLife>
	<20090501204612.GA6590@thebird.nl>
Message-ID: <B62873826C8C4AB982248FDA1748A21F@NewLife>

I always seem to be stumbling into these "volunteer opportunities..."
I'll see if I can make it go, then let you know-- cheers MAJ
----- Original Message ----- 
From: "Pjotr Prins" <pjotr.public14 at thebird.nl>
To: "Mark A. Jensen" <maj at fortinbras.us>
Cc: "BioLib Project" <biolib-dev at lists.open-bio.org>
Sent: Friday, May 01, 2009 4:46 PM
Subject: Re: [BioLib-dev] installation on cygwin


> Live dangerously ;-). The problem is that the cmake files have only
> been tried on Linux and OSX. Well, I guess I'll have to try myself. I have
> Windows somewhere, it is on my list now. 
> 
> Pj.
> 
> On Fri, May 01, 2009 at 02:11:49PM -0400, Mark A. Jensen wrote:
>> Hey guys,
>> Am trying to build on cygwin (I know, I know...). Have got cmake. I do
>>
>> $ cmake -DBUILD_PERL:BOOLEAN=TRUE .
>>
>> in the install directory (biolib-0.0.3), and receive
>>
>> -- The C compiler identification is GNU
>> CMake Error: Could not find cmake module  
>> file:/cygdrive/c/....../biolib-0.0.3/CMakeFiles/CMakeCCompiler.cmake
>> -- The CXX compiler identification is GNU
>> CMake Error: Could not find cmake module  
>> file:/cygdrive/c/...../biolib-0.0.3/CMakeFiles/CMakeCXXCompiler.cmake
>> ...
>>
>> Have set CYGWIN=1, CMAKE_INCLUDE_PATH=/usr/share/cmake-2.6.2/include,  
>> CMAKE_MODULE_PATH=/usr/share/cmake-2.6.2/Modules, still no dice. In
>> /usr/share/cmake-2.6.2/Modules, there exists CMakeCCompiler.cmake.in and  
>> CMakeCXXCompiler.cmake.in, but not
>> plain .cmake files.
>>
>> Any tips appreciated-
>> cheers Mark
>>
>> _______________________________________________
>> BioLib-dev mailing list
>> BioLib-dev at lists.open-bio.org
>> http://lists.open-bio.org/mailman/listinfo/biolib-dev
>>
> 
>


From pjotr.public14 at thebird.nl  Sat May  2 10:38:30 2009
From: pjotr.public14 at thebird.nl (Pjotr Prins)
Date: Sat, 2 May 2009 12:38:30 +0200
Subject: [BioLib-dev] installation on cygwin
In-Reply-To: <40844D11A9E84D05BBCE66BE6C76223B@NewLife>
References: <40844D11A9E84D05BBCE66BE6C76223B@NewLife>
Message-ID: <20090502103830.GA10664@thebird.nl>

If have cygwin running and a version of example and staden that
builds for Perl. Only problem is the test fails to find the dynamic
library. I'll fix that soon. Checkout the cygwin branch:

  cd biolib
  git checkout -b cygwin     # create new branch
  git remote add github git://github.com/pjotrp/biolib.git
  git pull github cygwin     # pull my branch
  ./configure --with-example
  make 
  make test
  ./configure --with-staden
  make
  make test

The latest version of biolib has moved the CMake modules into a
separate repository - that is shared with ASciLib.

The problem you were seeing actually had to do with missing
submodules for git. This should be pulled in with configure:

  git submodule init
  git submodule update

Prerequisites for Cygwin are:

The BioLib build system supports building on Cygwin
(http://www.cygwin.com/). First install Cygwin on your machine
selecting the following packages:

* cmake   (development)
* gcc-g++ (development)
* make    (development)
* swig    (development)
* your preferred language: perl, python and/or ruby

and, optionally:

* git (to fetch the latest repositories)

for ASciLib:

* autoconf (development)
* automake (development)

NOTE: The AffyIO module does not yet work on Cygwin - there is a
dependency on the R libraries. I'll sort that soon.

See also ./doc/WINDOWS.txt for the latest.

On Fri, May 01, 2009 at 02:11:49PM -0400, Mark A. Jensen wrote:
> Hey guys,
> Am trying to build on cygwin (I know, I know...). Have got cmake. I do
>
> $ cmake -DBUILD_PERL:BOOLEAN=TRUE .
>
> in the install directory (biolib-0.0.3), and receive
>
> -- The C compiler identification is GNU
> CMake Error: Could not find cmake module  
> file:/cygdrive/c/....../biolib-0.0.3/CMakeFiles/CMakeCCompiler.cmake
> -- The CXX compiler identification is GNU
> CMake Error: Could not find cmake module  
> file:/cygdrive/c/...../biolib-0.0.3/CMakeFiles/CMakeCXXCompiler.cmake
> ...
>
> Have set CYGWIN=1, CMAKE_INCLUDE_PATH=/usr/share/cmake-2.6.2/include,  
> CMAKE_MODULE_PATH=/usr/share/cmake-2.6.2/Modules, still no dice. In
> /usr/share/cmake-2.6.2/Modules, there exists CMakeCCompiler.cmake.in and  
> CMakeCXXCompiler.cmake.in, but not
> plain .cmake files.
>
> Any tips appreciated-
> cheers Mark
>
> _______________________________________________
> BioLib-dev mailing list
> BioLib-dev at lists.open-bio.org
> http://lists.open-bio.org/mailman/listinfo/biolib-dev
>


From pjotr.public14 at thebird.nl  Sat May  2 18:51:39 2009
From: pjotr.public14 at thebird.nl (Pjotr Prins)
Date: Sat, 2 May 2009 20:51:39 +0200
Subject: [BioLib-dev] installation on cygwin
In-Reply-To: <20090502103830.GA10664@thebird.nl>
References: <40844D11A9E84D05BBCE66BE6C76223B@NewLife>
	<20090502103830.GA10664@thebird.nl>
Message-ID: <20090502185139.GA15817@thebird.nl>

I have merged the BioLib cygwin branch into master - preparing for a
new release. The major change is the use of modified cmake modules
which are shared with ASciLib. 

The Cygwin tests are still broken. Dynaload.pm complains it can't find
shared libraries. I have to tell cmake how to find the Cygwin shared
libs - either due to the different naming convention (.dll) or because
of some missing linker switches exporting the correct bindings.

It is just a matter of poking around.

Pj.

On Sat, May 02, 2009 at 12:38:30PM +0200, Pjotr Prins wrote:
> If have cygwin running and a version of example and staden that
> builds for Perl. Only problem is the test fails to find the dynamic
> library. I'll fix that soon. Checkout the cygwin branch:
> 
>   cd biolib
>   git checkout -b cygwin     # create new branch
>   git remote add github git://github.com/pjotrp/biolib.git
>   git pull github cygwin     # pull my branch
>   ./configure --with-example
>   make 
>   make test
>   ./configure --with-staden
>   make
>   make test
> 
> The latest version of biolib has moved the CMake modules into a
> separate repository - that is shared with ASciLib.
> 
> The problem you were seeing actually had to do with missing
> submodules for git. This should be pulled in with configure:
> 
>   git submodule init
>   git submodule update
> 
> Prerequisites for Cygwin are:
> 
> The BioLib build system supports building on Cygwin
> (http://www.cygwin.com/). First install Cygwin on your machine
> selecting the following packages:
> 
> * cmake   (development)
> * gcc-g++ (development)
> * make    (development)
> * swig    (development)
> * your preferred language: perl, python and/or ruby
> 
> and, optionally:
> 
> * git (to fetch the latest repositories)
> 
> for ASciLib:
> 
> * autoconf (development)
> * automake (development)
> 
> NOTE: The AffyIO module does not yet work on Cygwin - there is a
> dependency on the R libraries. I'll sort that soon.
> 
> See also ./doc/WINDOWS.txt for the latest.
> 
> On Fri, May 01, 2009 at 02:11:49PM -0400, Mark A. Jensen wrote:
> > Hey guys,
> > Am trying to build on cygwin (I know, I know...). Have got cmake. I do
> >
> > $ cmake -DBUILD_PERL:BOOLEAN=TRUE .
> >
> > in the install directory (biolib-0.0.3), and receive
> >
> > -- The C compiler identification is GNU
> > CMake Error: Could not find cmake module  
> > file:/cygdrive/c/....../biolib-0.0.3/CMakeFiles/CMakeCCompiler.cmake
> > -- The CXX compiler identification is GNU
> > CMake Error: Could not find cmake module  
> > file:/cygdrive/c/...../biolib-0.0.3/CMakeFiles/CMakeCXXCompiler.cmake
> > ...
> >
> > Have set CYGWIN=1, CMAKE_INCLUDE_PATH=/usr/share/cmake-2.6.2/include,  
> > CMAKE_MODULE_PATH=/usr/share/cmake-2.6.2/Modules, still no dice. In
> > /usr/share/cmake-2.6.2/Modules, there exists CMakeCCompiler.cmake.in and  
> > CMakeCXXCompiler.cmake.in, but not
> > plain .cmake files.
> >
> > Any tips appreciated-
> > cheers Mark
> >
> > _______________________________________________
> > BioLib-dev mailing list
> > BioLib-dev at lists.open-bio.org
> > http://lists.open-bio.org/mailman/listinfo/biolib-dev
> >
> _______________________________________________
> BioLib-dev mailing list
> BioLib-dev at lists.open-bio.org
> http://lists.open-bio.org/mailman/listinfo/biolib-dev


From maj at fortinbras.us  Sat May  2 22:32:22 2009
From: maj at fortinbras.us (Mark A. Jensen)
Date: Sat, 2 May 2009 18:32:22 -0400
Subject: [BioLib-dev] installation on cygwin
In-Reply-To: <20090502185139.GA15817@thebird.nl>
References: <40844D11A9E84D05BBCE66BE6C76223B@NewLife><20090502103830.GA10664@thebird.nl>
	<20090502185139.GA15817@thebird.nl>
Message-ID: <97AB226A4E2C48D18943FD85CDC00236@NewLife>

Hey Pjotr--
I found this which got me quite a long ways --

this is key:
>From KDE-Windows:
>The thing is, cmake separates between "ARCHIVE", "LIBRARY" and "RUNTIME"
>components of targets.
>On Windows, the dll-file is considered a "RUNTIME" component, since it
> should go into the same directory as the executable, and the .lib file is
> considered an ARCHIVE component. On UNIX the .so file is a LIBRARY target,
> a .a file is (obviously) an ARCHIVE target.
>So if you say
>install(TARGETS mylib LIBRARY DESTINATION lib)
>under Windows then you didn't specifiy the destination directories for the
> dll and the lib file, that's why cmake complains.
>So you should do:
>install(TARGETS mylib LIBRARY DESTINATION lib${LIB_SUFFIX}
>                     ARCHIVE DESTINATION lib${LIB_SUFFIX}
>                     RUNTIME DESTINATION bin )
>
>Then it will do the right thing both on Windows and UNIX. If it's inside a
> KDE project, you can also use
>install(TARGETS mylib ${INSTALL_TARGETS_DEFAULT_ARGS} )
>which is a variable set in FindKDE4Internal.cmake, which basically contains
>just these settings.
>
>The simple way would be
>install(TARGETS mylib DESTINATION lib)
>but then Windows users may complain, because the dll won't be in the same
>directory as executables, so they'll have top adjust the PATH to make the
>executables run.

Not quite sure of the source of the original source of this tip (it was quoted 
where I found it), but doing

install ( TARGETS mylib
            LIBRARY DESTINATION lib
            RUNTIME DESTINATION bin )

in all the relevant CMakeLists.txt puts .dlls in the right place--
cheers- MAJ
----- Original Message ----- 
From: "Pjotr Prins" <pjotr.public14 at thebird.nl>
To: "BioLib Project" <biolib-dev at lists.open-bio.org>
Sent: Saturday, May 02, 2009 2:51 PM
Subject: Re: [BioLib-dev] installation on cygwin


>I have merged the BioLib cygwin branch into master - preparing for a
> new release. The major change is the use of modified cmake modules
> which are shared with ASciLib.
>
> The Cygwin tests are still broken. Dynaload.pm complains it can't find
> shared libraries. I have to tell cmake how to find the Cygwin shared
> libs - either due to the different naming convention (.dll) or because
> of some missing linker switches exporting the correct bindings.
>
> It is just a matter of poking around.
>
> Pj.
>
> On Sat, May 02, 2009 at 12:38:30PM +0200, Pjotr Prins wrote:
>> If have cygwin running and a version of example and staden that
>> builds for Perl. Only problem is the test fails to find the dynamic
>> library. I'll fix that soon. Checkout the cygwin branch:
>>
>>   cd biolib
>>   git checkout -b cygwin     # create new branch
>>   git remote add github git://github.com/pjotrp/biolib.git
>>   git pull github cygwin     # pull my branch
>>   ./configure --with-example
>>   make
>>   make test
>>   ./configure --with-staden
>>   make
>>   make test
>>
>> The latest version of biolib has moved the CMake modules into a
>> separate repository - that is shared with ASciLib.
>>
>> The problem you were seeing actually had to do with missing
>> submodules for git. This should be pulled in with configure:
>>
>>   git submodule init
>>   git submodule update
>>
>> Prerequisites for Cygwin are:
>>
>> The BioLib build system supports building on Cygwin
>> (http://www.cygwin.com/). First install Cygwin on your machine
>> selecting the following packages:
>>
>> * cmake   (development)
>> * gcc-g++ (development)
>> * make    (development)
>> * swig    (development)
>> * your preferred language: perl, python and/or ruby
>>
>> and, optionally:
>>
>> * git (to fetch the latest repositories)
>>
>> for ASciLib:
>>
>> * autoconf (development)
>> * automake (development)
>>
>> NOTE: The AffyIO module does not yet work on Cygwin - there is a
>> dependency on the R libraries. I'll sort that soon.
>>
>> See also ./doc/WINDOWS.txt for the latest.
>>
>> On Fri, May 01, 2009 at 02:11:49PM -0400, Mark A. Jensen wrote:
>> > Hey guys,
>> > Am trying to build on cygwin (I know, I know...). Have got cmake. I do
>> >
>> > $ cmake -DBUILD_PERL:BOOLEAN=TRUE .
>> >
>> > in the install directory (biolib-0.0.3), and receive
>> >
>> > -- The C compiler identification is GNU
>> > CMake Error: Could not find cmake module
>> > file:/cygdrive/c/....../biolib-0.0.3/CMakeFiles/CMakeCCompiler.cmake
>> > -- The CXX compiler identification is GNU
>> > CMake Error: Could not find cmake module
>> > file:/cygdrive/c/...../biolib-0.0.3/CMakeFiles/CMakeCXXCompiler.cmake
>> > ...
>> >
>> > Have set CYGWIN=1, CMAKE_INCLUDE_PATH=/usr/share/cmake-2.6.2/include,
>> > CMAKE_MODULE_PATH=/usr/share/cmake-2.6.2/Modules, still no dice. In
>> > /usr/share/cmake-2.6.2/Modules, there exists CMakeCCompiler.cmake.in and
>> > CMakeCXXCompiler.cmake.in, but not
>> > plain .cmake files.
>> >
>> > Any tips appreciated-
>> > cheers Mark
>> >
>> > _______________________________________________
>> > BioLib-dev mailing list
>> > BioLib-dev at lists.open-bio.org
>> > http://lists.open-bio.org/mailman/listinfo/biolib-dev
>> >
>> _______________________________________________
>> BioLib-dev mailing list
>> BioLib-dev at lists.open-bio.org
>> http://lists.open-bio.org/mailman/listinfo/biolib-dev
> _______________________________________________
> BioLib-dev mailing list
> BioLib-dev at lists.open-bio.org
> http://lists.open-bio.org/mailman/listinfo/biolib-dev
>
> 



From pjotr.public14 at thebird.nl  Mon May  4 12:18:18 2009
From: pjotr.public14 at thebird.nl (Pjotr Prins)
Date: Mon, 4 May 2009 14:18:18 +0200
Subject: [BioLib-dev] installation on cygwin
In-Reply-To: <40844D11A9E84D05BBCE66BE6C76223B@NewLife>
References: <40844D11A9E84D05BBCE66BE6C76223B@NewLife>
Message-ID: <20090504121818.GA2606@thebird.nl>

On Fri, May 01, 2009 at 02:11:49PM -0400, Mark A. Jensen wrote:
> Hey guys,
> Am trying to build on cygwin (I know, I know...). Have got cmake. I do

I am adding automated testing for BioLib. Here the current results
for Cygwin. You can see Ruby and Python are fine. Perl segfaults
after loading the DLL's.

Pj.

BioLib
0.0.4rc1
CYGWIN_NT-5.1 1.5.25(0.156/4/2) 2008-06-12 19:34
cmake version 2.6-patch 2

SWIG Version 1.3.36

Compiled with g++ [i686-pc-cygwin]

commit 2b86377a107be1443ef579cb54943658e0b91cdc
Author: pjotrp <pjotr.public01 at thebird.nl>
Date:   Mon May 4 13:42:40 2009 +0200

    Fix Perl test to find DLL
biolib git at github.com:pjotrp/biolib.git

This is perl, v5.10.0 built for cygwin-thread-multi-64int
(with 6 registered patches, see perl -V for more detail)

Running tests...
Start processing tests
Test project /home/xxx/autotest/biolib
  1/  2 Testing TestPerlExample               ***Exception: SegFault
  2/  2 Testing TestPerlStadenIOLib           ***Exception: SegFault

0% tests passed, 2 tests failed out of 2
	  1 - TestPerlExample (SEGFAULT)
	  2 - TestPerlStadenIOLib (SEGFAULT)
ruby 1.8.7 (2008-08-11 patchlevel 72) [i386-cygwin]
Running tests...
Start processing tests
Test project /home/xxx/autotest/biolib
  1/  2 Testing TestRubyExample                  Passed
  2/  2 Testing TestRubyStadenIOLib              Passed

100% tests passed, 0 tests failed out of 2
Running tests...
Start processing tests
Test project /home/xxx/autotest/biolib
  1/  2 Testing TestPythonExample                Passed
  2/  2 Testing TestPythonStadenIOLib            Passed

100% tests passed, 0 tests failed out of 2


From maj at fortinbras.us  Tue May  5 12:17:29 2009
From: maj at fortinbras.us (Mark A. Jensen)
Date: Tue, 5 May 2009 08:17:29 -0400
Subject: [BioLib-dev] installation on cygwin
In-Reply-To: <20090504121818.GA2606@thebird.nl>
References: <40844D11A9E84D05BBCE66BE6C76223B@NewLife>
	<20090504121818.GA2606@thebird.nl>
Message-ID: <70BB7AD164D44BAB8CEAB4FF24CB6E6F@NewLife>

The latest configured and made fine; I also get the segfaults. Will 
investigate--cheers
----- Original Message ----- 
From: "Pjotr Prins" <pjotr.public14 at thebird.nl>
To: "Mark A. Jensen" <maj at fortinbras.us>
Cc: "BioLib Project" <biolib-dev at lists.open-bio.org>
Sent: Monday, May 04, 2009 8:18 AM
Subject: Re: [BioLib-dev] installation on cygwin


> On Fri, May 01, 2009 at 02:11:49PM -0400, Mark A. Jensen wrote:
>> Hey guys,
>> Am trying to build on cygwin (I know, I know...). Have got cmake. I do
>
> I am adding automated testing for BioLib. Here the current results
> for Cygwin. You can see Ruby and Python are fine. Perl segfaults
> after loading the DLL's.
>
> Pj.
>
> BioLib
> 0.0.4rc1
> CYGWIN_NT-5.1 1.5.25(0.156/4/2) 2008-06-12 19:34
> cmake version 2.6-patch 2
>
> SWIG Version 1.3.36
>
> Compiled with g++ [i686-pc-cygwin]
>
> commit 2b86377a107be1443ef579cb54943658e0b91cdc
> Author: pjotrp <pjotr.public01 at thebird.nl>
> Date:   Mon May 4 13:42:40 2009 +0200
>
>    Fix Perl test to find DLL
> biolib git at github.com:pjotrp/biolib.git
>
> This is perl, v5.10.0 built for cygwin-thread-multi-64int
> (with 6 registered patches, see perl -V for more detail)
>
> Running tests...
> Start processing tests
> Test project /home/xxx/autotest/biolib
>  1/  2 Testing TestPerlExample               ***Exception: SegFault
>  2/  2 Testing TestPerlStadenIOLib           ***Exception: SegFault
>
> 0% tests passed, 2 tests failed out of 2
>   1 - TestPerlExample (SEGFAULT)
>   2 - TestPerlStadenIOLib (SEGFAULT)
> ruby 1.8.7 (2008-08-11 patchlevel 72) [i386-cygwin]
> Running tests...
> Start processing tests
> Test project /home/xxx/autotest/biolib
>  1/  2 Testing TestRubyExample                  Passed
>  2/  2 Testing TestRubyStadenIOLib              Passed
>
> 100% tests passed, 0 tests failed out of 2
> Running tests...
> Start processing tests
> Test project /home/xxx/autotest/biolib
>  1/  2 Testing TestPythonExample                Passed
>  2/  2 Testing TestPythonStadenIOLib            Passed
>
> 100% tests passed, 0 tests failed out of 2
>
> 



From pjotr.public14 at thebird.nl  Tue May  5 21:16:19 2009
From: pjotr.public14 at thebird.nl (Pjotr Prins)
Date: Tue, 5 May 2009 23:16:19 +0200
Subject: [BioLib-dev] 0.0.4 release
Message-ID: <20090505211619.GA15941@thebird.nl>

BioLib is getting adolescent. With automated integration testing and a
build system I am happy with. 

Cygwin is supported - though the Perl tests are still broken (Python
and Ruby are fine). It shouldn't be too hard to also support native
Windows Perl etc.

On Linux everything should work. With OS/X Perl Staden works.

Download 0.0.4 from http://bio3.biobeat.org/download/nix-biology/ or 
preferably:

  git clone git://github.com/pjotrp/biolib.git

---------------------------------------------------------------------
BIOLIB 0.04  a3de2f3619d05b613dc15280773c341bf3a33267     May 5, 2009
---------------------------------------------------------------------

- Full Perl, Python and Ruby support
- Cygwin support (Microsoft Windows) - except for Affyio as it
  requires Rlib. Ruby and Python are supported. Perl segfaults.
- Moved cmake modules into own git repository (for sharing with
  ASciLib) - see http://github.com/pjotrp/cmake-support/
- Automated testing for different setups introduced
- Generate XML docs from SWIG
- Added R/QTL module



From xshuai at umail.iu.edu  Thu May 14 04:22:10 2009
From: xshuai at umail.iu.edu (Xin Shuai)
Date: Thu, 14 May 2009 00:22:10 -0400
Subject: [BioLib-dev] CMakeLists.txt
Message-ID: <3a7743460905132122q4013614fh10c7db434cb36d25@mail.gmail.com>

HI! It seems that writing CMakeLists.txt is a big job in BioLib. And I have
create a CMakeLists.txt file for each directory. Do I have to specially
study related command of writing a CMakeLIsts.txt file?

-- 
Xin Shuai (David)
PhD of Complex System in School of Informatics
Indiana University Bloomington
812-606-8019


From pjotr.public14 at thebird.nl  Thu May 14 06:03:56 2009
From: pjotr.public14 at thebird.nl (Pjotr Prins)
Date: Thu, 14 May 2009 08:03:56 +0200
Subject: [BioLib-dev] CMakeLists.txt
In-Reply-To: <3a7743460905132122q4013614fh10c7db434cb36d25@mail.gmail.com>
References: <3a7743460905132122q4013614fh10c7db434cb36d25@mail.gmail.com>
Message-ID: <20090514060356.GC4164@thebird.nl>

CMake is the back bone of biolib sorting out all dependencies and
locations. Fortunately CMake is very straightforward, it is really a
little programming language (macro language) that generates the make
files. Rather like the often used autoconf tools (i.e. 'configure').
Unlike autoconf it is easy to understand and rather nice at status
messages. I have learned to love it, despite the fact it is another
little language to learn.

The most important thing to remember is that when you change a CMake
file you have to remove the cache files. CMake caches all settings -
probably necessary for big build systems like KDE.

Initially just copy the CMakeLists.txt file from other places and
adapt it to your tree. If there is 'deeper' functionality required I can
always help get it to work.

Read the CMake online documentation (take an hour, and the FAQ). It'll
pay off.

Pj.

On Thu, May 14, 2009 at 12:22:10AM -0400, Xin Shuai wrote:
> HI! It seems that writing CMakeLists.txt is a big job in BioLib. And I have
> create a CMakeLists.txt file for each directory. Do I have to specially
> study related command of writing a CMakeLIsts.txt file?
> 
> -- 
> Xin Shuai (David)
> PhD of Complex System in School of Informatics
> Indiana University Bloomington
> 812-606-8019
> _______________________________________________
> BioLib-dev mailing list
> BioLib-dev at lists.open-bio.org
> http://lists.open-bio.org/mailman/listinfo/biolib-dev
> 


From cjfields at illinois.edu  Thu May 14 21:14:49 2009
From: cjfields at illinois.edu (Chris Fields)
Date: Thu, 14 May 2009 16:14:49 -0500
Subject: [BioLib-dev] [ANNOUNCEMENT] Google Summer of Code student Xin Shuai
Message-ID: <7A388EB0-E9B2-4579-80EA-83AC95817EF9@illinois.edu>

All,

I am proud to introduce Xin 'David' Shuai, my student for the Google  
Summer of Code 2009, to the Open Bioinformatics community.  David's  
project centers on developing SWIG-based bindings to libsequence (a  
population genetics library) for the BioLib project:

http://biolib.open-bio.org/wiki/Main_Page

Besides myself, David will be co-mentored by Mark Jensen and Pjotr  
Prins.

As the BioLib project centers on creating common, maintainable SWIG- 
based bindings to popular bioinformatics libraries for the various  
Bio* toolkits, we will likely need input from the various Open Bio  
communities at various stages in the project.  At this time, David's  
initial plans are to develop and test libsequence bindings for Perl  
and Python.

David's proposal and project plan are available here:

http://biolib.open-bio.org/wiki/User:David

Congratulations David, and welcome to the Open-Bio community!

Sincerely,

Christopher Fields
University of Illinois Urbana-Champaign
Institute for Genomic Biology
Urbana, IL 61801


From hlapp at gmx.net  Thu May 14 22:16:30 2009
From: hlapp at gmx.net (Hilmar Lapp)
Date: Thu, 14 May 2009 18:16:30 -0400
Subject: [BioLib-dev] [Bioperl-l] [ANNOUNCEMENT] Google Summer of Code
	student Xin Shuai
In-Reply-To: <7A388EB0-E9B2-4579-80EA-83AC95817EF9@illinois.edu>
References: <7A388EB0-E9B2-4579-80EA-83AC95817EF9@illinois.edu>
Message-ID: <4482F182-B0B0-4C71-B8DE-B5B7A4EC4D81@gmx.net>

Welcome David, good luck with your project, and I hope (actually, am  
quite certain) that you'll enjoy your summer with us.

	-hilmar

On May 14, 2009, at 5:14 PM, Chris Fields wrote:

> All,
>
> I am proud to introduce Xin 'David' Shuai, my student for the Google  
> Summer of Code 2009, to the Open Bioinformatics community.  David's  
> project centers on developing SWIG-based bindings to libsequence (a  
> population genetics library) for the BioLib project:
>
> http://biolib.open-bio.org/wiki/Main_Page
>
> Besides myself, David will be co-mentored by Mark Jensen and Pjotr  
> Prins.
>
> As the BioLib project centers on creating common, maintainable SWIG- 
> based bindings to popular bioinformatics libraries for the various  
> Bio* toolkits, we will likely need input from the various Open Bio  
> communities at various stages in the project.  At this time, David's  
> initial plans are to develop and test libsequence bindings for Perl  
> and Python.
>
> David's proposal and project plan are available here:
>
> http://biolib.open-bio.org/wiki/User:David
>
> Congratulations David, and welcome to the Open-Bio community!
>
> Sincerely,
>
> Christopher Fields
> University of Illinois Urbana-Champaign
> Institute for Genomic Biology
> Urbana, IL 61801
> _______________________________________________
> Bioperl-l mailing list
> Bioperl-l at lists.open-bio.org
> http://lists.open-bio.org/mailman/listinfo/bioperl-l

-- 
===========================================================
: Hilmar Lapp  -:-  Durham, NC  -:-  hlapp at gmx dot net :
===========================================================





From xshuai at umail.iu.edu  Fri May 15 01:52:40 2009
From: xshuai at umail.iu.edu (Xin Shuai)
Date: Thu, 14 May 2009 21:52:40 -0400
Subject: [BioLib-dev] [Bioperl-l] [ANNOUNCEMENT] Google Summer of Code
	student Xin Shuai
In-Reply-To: <4482F182-B0B0-4C71-B8DE-B5B7A4EC4D81@gmx.net>
References: <7A388EB0-E9B2-4579-80EA-83AC95817EF9@illinois.edu>
	<4482F182-B0B0-4C71-B8DE-B5B7A4EC4D81@gmx.net>
Message-ID: <3a7743460905141852x69e32266oc440e7557d59d84f@mail.gmail.com>

Thank you for everyone's help during my application. I will do my best to
accomplish it.Since I'm a totally newcomer in Bio* language program, I will
have a lot to ask during the following months and hope to get your help!

On Thu, May 14, 2009 at 6:16 PM, Hilmar Lapp <hlapp at gmx.net> wrote:

> Welcome David, good luck with your project, and I hope (actually, am quite
> certain) that you'll enjoy your summer with us.
>
>        -hilmar
>
>
> On May 14, 2009, at 5:14 PM, Chris Fields wrote:
>
>  All,
>>
>> I am proud to introduce Xin 'David' Shuai, my student for the Google
>> Summer of Code 2009, to the Open Bioinformatics community.  David's project
>> centers on developing SWIG-based bindings to libsequence (a population
>> genetics library) for the BioLib project:
>>
>> http://biolib.open-bio.org/wiki/Main_Page
>>
>> Besides myself, David will be co-mentored by Mark Jensen and Pjotr Prins.
>>
>> As the BioLib project centers on creating common, maintainable SWIG-based
>> bindings to popular bioinformatics libraries for the various Bio* toolkits,
>> we will likely need input from the various Open Bio communities at various
>> stages in the project.  At this time, David's initial plans are to develop
>> and test libsequence bindings for Perl and Python.
>>
>> David's proposal and project plan are available here:
>>
>> http://biolib.open-bio.org/wiki/User:David
>>
>> Congratulations David, and welcome to the Open-Bio community!
>>
>> Sincerely,
>>
>> Christopher Fields
>> University of Illinois Urbana-Champaign
>> Institute for Genomic Biology
>> Urbana, IL 61801
>> _______________________________________________
>> Bioperl-l mailing list
>> Bioperl-l at lists.open-bio.org
>> http://lists.open-bio.org/mailman/listinfo/bioperl-l
>>
>
> --
> ===========================================================
> : Hilmar Lapp  -:-  Durham, NC  -:-  hlapp at gmx dot net :
> ===========================================================
>
>
>
>


-- 
Xin Shuai (David)
PhD of Complex System in School of Informatics
Indiana University Bloomington
812-606-8019


From xshuai at umail.iu.edu  Fri May 15 03:57:30 2009
From: xshuai at umail.iu.edu (Xin Shuai)
Date: Thu, 14 May 2009 23:57:30 -0400
Subject: [BioLib-dev] Where can I see "swig"
Message-ID: <3a7743460905142057j6d87070w83d5098e20a656ce@mail.gmail.com>

I tried to build the example module with python bindings and the result
seems fine.However, I feel a little puzzled that where does "swig" take
effect in the whole process?
In other words, in which file can I find the command that is related to
using "swig"?

-- 
Xin Shuai (David)
PhD of Complex System in School of Informatics
Indiana University Bloomington
812-606-8019


From maj at fortinbras.us  Fri May 15 04:11:15 2009
From: maj at fortinbras.us (Mark A. Jensen)
Date: Fri, 15 May 2009 00:11:15 -0400
Subject: [BioLib-dev] Where can I see "swig"
In-Reply-To: <3a7743460905142057j6d87070w83d5098e20a656ce@mail.gmail.com>
References: <3a7743460905142057j6d87070w83d5098e20a656ce@mail.gmail.com>
Message-ID: <D9D0A7C2077645269E3ED7E4694918AA@NewLife>

Hey David,
look in [install dir]/src/mappings/swig; the interface files (which swig uses to 
create the bindings) are there for all the mappings. Then you can go down to 
python/example/CMakeLists.txt to see cmake routines that are running swig. The 
lines that do the work are (Pj. will correct me if I'm wrong)...

 INCLUDE(${SWIG_USE_FILE})

 SET(SWIG_FLAGS "")
 SET_SOURCE_FILES_PROPERTIES(${INTERFACE} PROPERTIES SWIG_FLAGS "-includeall")

 SWIG_ADD_MODULE(${MY_MODULE} python ${INTERFACE} ${SOURCES} )
 SWIG_LINK_LIBRARIES(${MY_MODULE} ${PYTHON_LIBRARIES} )

Mark

----- Original Message ----- 
From: "Xin Shuai" <xshuai at umail.iu.edu>
To: "BioLib Project" <biolib-dev at lists.open-bio.org>
Sent: Thursday, May 14, 2009 11:57 PM
Subject: [BioLib-dev] Where can I see "swig"


>I tried to build the example module with python bindings and the result
> seems fine.However, I feel a little puzzled that where does "swig" take
> effect in the whole process?
> In other words, in which file can I find the command that is related to
> using "swig"?
>
> -- 
> Xin Shuai (David)
> PhD of Complex System in School of Informatics
> Indiana University Bloomington
> 812-606-8019
> _______________________________________________
> BioLib-dev mailing list
> BioLib-dev at lists.open-bio.org
> http://lists.open-bio.org/mailman/listinfo/biolib-dev
>
> 



From xshuai at umail.iu.edu  Fri May 15 05:29:04 2009
From: xshuai at umail.iu.edu (Xin Shuai)
Date: Fri, 15 May 2009 01:29:04 -0400
Subject: [BioLib-dev] Where can I see "swig"
In-Reply-To: <D9D0A7C2077645269E3ED7E4694918AA@NewLife>
References: <3a7743460905142057j6d87070w83d5098e20a656ce@mail.gmail.com>
	<D9D0A7C2077645269E3ED7E4694918AA@NewLife>
Message-ID: <3a7743460905142229t402757c6w7410ea784a2f4d4d@mail.gmail.com>

Hi, Mark! I'm afraid I cannot see as you post it in
python/example/CMakeLists.txt. The following is what I can see in that file:#
Builds the Python biolib_example library after resolving paths.

cmake_minimum_required(VERSION 2.6)

ENABLE_TESTING()

SET (M_NAME example)
SET (M_VERSION 1.0)
SET (INTERFACE python_${M_NAME}.i)

IF(NOT BUILD_LIBS)
  SET (MAP_ROOT ../../../../..)
  SET (CMAKE_MODULE_PATH ${MAP_ROOT}/tools/cmake-support/modules)
ENDIF(NOT BUILD_LIBS)

SET(CMAKE_LIBRARY_OUTPUT_DIRECTORY biolib)

FIND_PACKAGE(MapPython REQUIRED)

POST_BUILD_PYTHON_BINDINGS()
TEST_PYTHON_BINDINGS()
INSTALL_PYTHON_BINDINGS()

I still cannot see any "swig" related command.

On Fri, May 15, 2009 at 12:11 AM, Mark A. Jensen <maj at fortinbras.us> wrote:

> Hey David,
> look in [install dir]/src/mappings/swig; the interface files (which swig
> uses to create the bindings) are there for all the mappings. Then you can go
> down to python/example/CMakeLists.txt to see cmake routines that are running
> swig. The lines that do the work are (Pj. will correct me if I'm wrong)...
>
> INCLUDE(${SWIG_USE_FILE})
>
> SET(SWIG_FLAGS "")
> SET_SOURCE_FILES_PROPERTIES(${INTERFACE} PROPERTIES SWIG_FLAGS
> "-includeall")
>
> SWIG_ADD_MODULE(${MY_MODULE} python ${INTERFACE} ${SOURCES} )
> SWIG_LINK_LIBRARIES(${MY_MODULE} ${PYTHON_LIBRARIES} )
>
> Mark
>
> ----- Original Message ----- From: "Xin Shuai" <xshuai at umail.iu.edu>
> To: "BioLib Project" <biolib-dev at lists.open-bio.org>
> Sent: Thursday, May 14, 2009 11:57 PM
> Subject: [BioLib-dev] Where can I see "swig"
>
>
>  I tried to build the example module with python bindings and the result
>> seems fine.However, I feel a little puzzled that where does "swig" take
>> effect in the whole process?
>> In other words, in which file can I find the command that is related to
>> using "swig"?
>>
>> --
>> Xin Shuai (David)
>> PhD of Complex System in School of Informatics
>> Indiana University Bloomington
>> 812-606-8019
>> _______________________________________________
>> BioLib-dev mailing list
>> BioLib-dev at lists.open-bio.org
>> http://lists.open-bio.org/mailman/listinfo/biolib-dev
>>
>>
>>
>


-- 
Xin Shuai (David)
PhD of Complex System in School of Informatics
Indiana University Bloomington
812-606-8019


From pjotr.public14 at thebird.nl  Fri May 15 05:52:28 2009
From: pjotr.public14 at thebird.nl (Pjotr Prins)
Date: Fri, 15 May 2009 07:52:28 +0200
Subject: [BioLib-dev] Where can I see "swig"
In-Reply-To: <3a7743460905142229t402757c6w7410ea784a2f4d4d@mail.gmail.com>
References: <3a7743460905142057j6d87070w83d5098e20a656ce@mail.gmail.com>
	<D9D0A7C2077645269E3ED7E4694918AA@NewLife>
	<3a7743460905142229t402757c6w7410ea784a2f4d4d@mail.gmail.com>
Message-ID: <20090515055228.GA15185@thebird.nl>

On Fri, May 15, 2009 at 01:29:04AM -0400, Xin Shuai wrote:
> POST_BUILD_PYTHON_BINDINGS()
> TEST_PYTHON_BINDINGS()
> INSTALL_PYTHON_BINDINGS()
> 
> I still cannot see any "swig" related command.

Above are macros defined in ./tools/cmake-support/modules/. They,
indeed, do the hard work of managing SWIG for Python. The command:

  FIND_PACKAGE(MapPython REQUIRED)

loads the relevant cmake files.

Pj.



From pjotr.public14 at thebird.nl  Fri May 15 06:15:01 2009
From: pjotr.public14 at thebird.nl (Pjotr Prins)
Date: Fri, 15 May 2009 08:15:01 +0200
Subject: [BioLib-dev] BioLib Talk at BOSC 2009
Message-ID: <20090515061501.GA16563@thebird.nl>

I will give a talk at BOSC 2009 on the BioLib project, the abstract
was accepted. I hope to see many of you :-). I also requested a Birds
of a Feather session on incorporating old and new C libraries into
BioLib. If that fails we will do something after hours.

Pjotr.


From pjotr.public14 at thebird.nl  Fri May 15 06:18:27 2009
From: pjotr.public14 at thebird.nl (Pjotr Prins)
Date: Fri, 15 May 2009 08:18:27 +0200
Subject: [BioLib-dev] [ANNOUNCEMENT] Google Summer of Code student Adam Smith
In-Reply-To: <20090515061051.GA15908@thebird.nl>
References: <20090515061051.GA15908@thebird.nl>
Message-ID: <20090515061827.GA16958@thebird.nl>

Hi Everyone,

I am glad to announce the participation of Adam Smith in BioLib for
the Google Summer of Code 2009. Adam will work on embedding Bio++ into
BioLib, which news has been well received by both projects (Julien is
the project leader for Bio++).  Bio++ is a set of C++ libraries for
Bioinformatics, including sequence analysis, phylogenetics, molecular
evolution and population genetics. Adam's project is described here:

https://www.nescent.org/wg/phyloinformatics/index.php?title=Phyloinformatics_Summer_of_Code_2009#Mapping_the_Bio.2B.2B_Phylogenetics_toolkit_to_BioPerl.2C_R.2FBioConductor_and_BioJAVA_using_BioLib

Adam has defended his Ph.D. this week, and from now on he can work
fully on this project over the summer. Adam's home page is here:

  http://pages.cs.wisc.edu/~aasmith/

and will announce on this list when he has updated his project page.

He will also give a talk at ISMB, so we'll get a chance to meet in
person.

Pjotr.


From xshuai at umail.iu.edu  Fri May 15 15:53:28 2009
From: xshuai at umail.iu.edu (Xin Shuai)
Date: Fri, 15 May 2009 11:53:28 -0400
Subject: [BioLib-dev] Where can I see "swig"
In-Reply-To: <20090515055228.GA15185@thebird.nl>
References: <3a7743460905142057j6d87070w83d5098e20a656ce@mail.gmail.com>
	<D9D0A7C2077645269E3ED7E4694918AA@NewLife>
	<3a7743460905142229t402757c6w7410ea784a2f4d4d@mail.gmail.com>
	<20090515055228.GA15185@thebird.nl>
Message-ID: <3a7743460905150853v690fd8e8g8a81ac0553d85d8@mail.gmail.com>

It seems like those .cmake files in ./tools/cmake-support/modules are all
created automatically by CMake and I don't have to care about it. However, I
should specify in CMakeLists.txt that how to create those macros, is that
right?
By the way, at this moment I can only build modules in BioLib one by one
using
 ./configure command  --with-(module name)
Since when I want to configure the whole modules I was always told that
R_LIBRARY is not found. Is it possible for me to avoid such problem but
still configure the whole modules?

On Fri, May 15, 2009 at 1:52 AM, Pjotr Prins <pjotr.public14 at thebird.nl>wrote:

> On Fri, May 15, 2009 at 01:29:04AM -0400, Xin Shuai wrote:
> > POST_BUILD_PYTHON_BINDINGS()
> > TEST_PYTHON_BINDINGS()
> > INSTALL_PYTHON_BINDINGS()
> >
> > I still cannot see any "swig" related command.
>
> Above are macros defined in ./tools/cmake-support/modules/. They,
> indeed, do the hard work of managing SWIG for Python. The command:
>
>  FIND_PACKAGE(MapPython REQUIRED)
>
> loads the relevant cmake files.
>
> Pj.
>
>


-- 
Xin Shuai (David)
PhD of Complex System in School of Informatics
Indiana University Bloomington
812-606-8019


From pjotr.public14 at thebird.nl  Fri May 15 19:28:39 2009
From: pjotr.public14 at thebird.nl (Pjotr Prins)
Date: Fri, 15 May 2009 21:28:39 +0200
Subject: [BioLib-dev] Where can I see "swig"
In-Reply-To: <3a7743460905150853v690fd8e8g8a81ac0553d85d8@mail.gmail.com>
References: <3a7743460905142057j6d87070w83d5098e20a656ce@mail.gmail.com>
	<D9D0A7C2077645269E3ED7E4694918AA@NewLife>
	<3a7743460905142229t402757c6w7410ea784a2f4d4d@mail.gmail.com>
	<20090515055228.GA15185@thebird.nl>
	<3a7743460905150853v690fd8e8g8a81ac0553d85d8@mail.gmail.com>
Message-ID: <20090515192839.GD25030@thebird.nl>

BioLib has a modular setup. Which is what it should be. Only the
Affyio library requires R, and on some platforms I have not fixed
finding them. It will come soon.

Pj.

On Fri, May 15, 2009 at 11:53:28AM -0400, Xin Shuai wrote:
> It seems like those .cmake files in ./tools/cmake-support/modules are all
> created automatically by CMake and I don't have to care about it. However, I
> should specify in CMakeLists.txt that how to create those macros, is that
> right?
> By the way, at this moment I can only build modules in BioLib one by one
> using
>  ./configure command  --with-(module name)
> Since when I want to configure the whole modules I was always told that
> R_LIBRARY is not found. Is it possible for me to avoid such problem but
> still configure the whole modules?
> 
> On Fri, May 15, 2009 at 1:52 AM, Pjotr Prins <pjotr.public14 at thebird.nl>wrote:
> 
> > On Fri, May 15, 2009 at 01:29:04AM -0400, Xin Shuai wrote:
> > > POST_BUILD_PYTHON_BINDINGS()
> > > TEST_PYTHON_BINDINGS()
> > > INSTALL_PYTHON_BINDINGS()
> > >
> > > I still cannot see any "swig" related command.
> >
> > Above are macros defined in ./tools/cmake-support/modules/. They,
> > indeed, do the hard work of managing SWIG for Python. The command:
> >
> >  FIND_PACKAGE(MapPython REQUIRED)
> >
> > loads the relevant cmake files.
> >
> > Pj.
> >
> >
> 
> 
> -- 
> Xin Shuai (David)
> PhD of Complex System in School of Informatics
> Indiana University Bloomington
> 812-606-8019


From xshuai at umail.iu.edu  Fri May 15 20:09:53 2009
From: xshuai at umail.iu.edu (Xin Shuai)
Date: Fri, 15 May 2009 16:09:53 -0400
Subject: [BioLib-dev] Where can I see "swig"
In-Reply-To: <20090515192839.GD25030@thebird.nl>
References: <3a7743460905142057j6d87070w83d5098e20a656ce@mail.gmail.com>
	<D9D0A7C2077645269E3ED7E4694918AA@NewLife>
	<3a7743460905142229t402757c6w7410ea784a2f4d4d@mail.gmail.com>
	<20090515055228.GA15185@thebird.nl>
	<3a7743460905150853v690fd8e8g8a81ac0553d85d8@mail.gmail.com>
	<20090515192839.GD25030@thebird.nl>
Message-ID: <3a7743460905151309r2f93d095xd8082c8bb2548c3c@mail.gmail.com>

I noticed that such form of file " _staden_io_lib.so"will be created
together with corresponding .py and .pyc files
this .so file is a type of binary file
What's the function of this file?

On Fri, May 15, 2009 at 3:28 PM, Pjotr Prins <pjotr.public14 at thebird.nl>wrote:

> BioLib has a modular setup. Which is what it should be. Only the
> Affyio library requires R, and on some platforms I have not fixed
> finding them. It will come soon.
>
> Pj.
>
> On Fri, May 15, 2009 at 11:53:28AM -0400, Xin Shuai wrote:
> > It seems like those .cmake files in ./tools/cmake-support/modules are all
> > created automatically by CMake and I don't have to care about it.
> However, I
> > should specify in CMakeLists.txt that how to create those macros, is that
> > right?
> > By the way, at this moment I can only build modules in BioLib one by one
> > using
> >  ./configure command  --with-(module name)
> > Since when I want to configure the whole modules I was always told that
> > R_LIBRARY is not found. Is it possible for me to avoid such problem but
> > still configure the whole modules?
> >
> > On Fri, May 15, 2009 at 1:52 AM, Pjotr Prins <pjotr.public14 at thebird.nl
> >wrote:
> >
> > > On Fri, May 15, 2009 at 01:29:04AM -0400, Xin Shuai wrote:
> > > > POST_BUILD_PYTHON_BINDINGS()
> > > > TEST_PYTHON_BINDINGS()
> > > > INSTALL_PYTHON_BINDINGS()
> > > >
> > > > I still cannot see any "swig" related command.
> > >
> > > Above are macros defined in ./tools/cmake-support/modules/. They,
> > > indeed, do the hard work of managing SWIG for Python. The command:
> > >
> > >  FIND_PACKAGE(MapPython REQUIRED)
> > >
> > > loads the relevant cmake files.
> > >
> > > Pj.
> > >
> > >
> >
> >
> > --
> > Xin Shuai (David)
> > PhD of Complex System in School of Informatics
> > Indiana University Bloomington
> > 812-606-8019
>



-- 
Xin Shuai (David)
PhD of Complex System in School of Informatics
Indiana University Bloomington
812-606-8019


From pjotr.public14 at thebird.nl  Fri May 15 23:09:27 2009
From: pjotr.public14 at thebird.nl (Pjotr Prins)
Date: Sat, 16 May 2009 01:09:27 +0200
Subject: [BioLib-dev] Where can I see "swig"
In-Reply-To: <3a7743460905151309r2f93d095xd8082c8bb2548c3c@mail.gmail.com>
References: <3a7743460905142057j6d87070w83d5098e20a656ce@mail.gmail.com>
	<D9D0A7C2077645269E3ED7E4694918AA@NewLife>
	<3a7743460905142229t402757c6w7410ea784a2f4d4d@mail.gmail.com>
	<20090515055228.GA15185@thebird.nl>
	<3a7743460905150853v690fd8e8g8a81ac0553d85d8@mail.gmail.com>
	<20090515192839.GD25030@thebird.nl>
	<3a7743460905151309r2f93d095xd8082c8bb2548c3c@mail.gmail.com>
Message-ID: <20090515230927.GA25644@thebird.nl>

On Fri, May 15, 2009 at 04:09:53PM -0400, Xin Shuai wrote:
> I noticed that such form of file " _staden_io_lib.so"will be created
> together with corresponding .py and .pyc files
> this .so file is a type of binary file
> What's the function of this file?

It contains Python specific bindings - the C counterpart of the
Python file(s). Some languages have both, Ruby has only a C lib.

The bindings handle the interface between the high level language and
the low level (generic) C libraries - like the functionality in
biolib-staden_io.so (which links against the bindings of all
languages and contains the actual functionality we aim to bind).

Hope that helps. You can try and read the C code.

Pj.


From xshuai at umail.iu.edu  Sat May 16 16:15:27 2009
From: xshuai at umail.iu.edu (Xin Shuai)
Date: Sat, 16 May 2009 12:15:27 -0400
Subject: [BioLib-dev] Submit my revision
Message-ID: <3a7743460905160915i3018728cgdd31881511a8c651@mail.gmail.com>

HI, Pjotr:  I forked your biolib repo and cloned my fork using "Your clone
URL"
  Then I made some a few changes and what to submit my revision.
  I went into the directory of changed "biolib" in my computer and then type
the following commands:


xin-shuais-macbook-2:biolib diavy$ git push origin masterEverything
up-to-date

However, when I checked my fork in github.com again and didn't see any
change at all. Why?


-- 
Xin Shuai (David)
PhD of Complex System in School of Informatics
Indiana University Bloomington
812-606-8019


From pjotr.public14 at thebird.nl  Sat May 16 18:18:55 2009
From: pjotr.public14 at thebird.nl (Pjotr Prins)
Date: Sat, 16 May 2009 20:18:55 +0200
Subject: [BioLib-dev] Submit my revision
In-Reply-To: <3a7743460905160915i3018728cgdd31881511a8c651@mail.gmail.com>
References: <3a7743460905160915i3018728cgdd31881511a8c651@mail.gmail.com>
Message-ID: <20090516181855.GA344@thebird.nl>

On Sat, May 16, 2009 at 12:15:27PM -0400, Xin Shuai wrote:
> HI, Pjotr:  I forked your biolib repo and cloned my fork using "Your clone
> URL"
>   Then I made some a few changes and what to submit my revision.
>   I went into the directory of changed "biolib" in my computer and then type
> the following commands:
> 
> 
> xin-shuais-macbook-2:biolib diavy$ git push origin masterEverything
> up-to-date
> 
> However, when I checked my fork in github.com again and didn't see any
> change at all. Why?

Did you commit your changes? 

man git-commit

And see the github howto.

Pj.


From aasmith at cs.wisc.edu  Thu May 21 07:05:50 2009
From: aasmith at cs.wisc.edu (Adam A Smith)
Date: Thu, 21 May 2009 02:05:50 -0500 (CDT)
Subject: [BioLib-dev] [ANNOUNCEMENT] Google Summer of Code student Adam
 Smith
In-Reply-To: <20090515061827.GA16958@thebird.nl>
References: <20090515061051.GA15908@thebird.nl>
	<20090515061827.GA16958@thebird.nl>
Message-ID: <Pine.LNX.4.64.0905210203360.13454@king06.cs.wisc.edu>

> and will announce on this list when he has updated his project page.

project page is now updated.

http://www.nescent.org/wg_phyloinformatics/PhyloSoC:Mapping_the_Bio%2B%2B_Phylogenetics_toolkit_to_R/BioConductor_and_BioJAVA_using_BioLib

sorry everyone for the delay in getting up & running.  had some important, 
unexpected family obligations that came up right as i was trying to 
defend.  but i'm here now.

Adam


>
> He will also give a talk at ISMB, so we'll get a chance to meet in
> person.
>
> Pjotr.
> _______________________________________________
> BioLib-dev mailing list
> BioLib-dev at lists.open-bio.org
> http://lists.open-bio.org/mailman/listinfo/biolib-dev
>


From pjotr.public14 at thebird.nl  Fri May 22 17:01:13 2009
From: pjotr.public14 at thebird.nl (Pjotr Prins)
Date: Fri, 22 May 2009 19:01:13 +0200
Subject: [BioLib-dev] Adding Bio++ to BioLib
Message-ID: <20090522170113.GA341@thebird.nl>

For Adam I have created a Bio++ git repository on github at:

  http://github.com/pjotrp/biopp/tree/master

And a special biopp branch of the Biolib repository which includes
Bio++ as a git submodule (in ./config/biopp), with one example build.

I'll do something similar for David.

Pj.