openbabel-scripting — This list is deprecated. Please use openbabel-discuss instead.

[openbabel-scripting] Re: Python module loading problems...

[Geoffrey H. <ge...@ge...>]

Good news!

I just got a very helpful posting from the SWIG mailing list:

On Dec 8, 2005, at 4:11 PM, Josh Cherry wrote:
> A while ago I was doing something where I
> had to mess around with Python's dll parameters, something like
>
> import sys
> import dl
> sys.setdlopenflags(sys.getdlopenflags() | dl.RTLD_GLOBAL)

This goes before the "import openbabel" bits in a Python script. I've  
verified that it is a workaround on Linux.

I'd still like to make sure this won't be needed in the future (e.g.,  
it's already in openbabel.py). At the worst, I think it can probably  
get added to the top of openbabel.py before anything else happens.

Please let me know if this works for you or if there are still issues.

Cheers,
-Geoff

Re: [openbabel-scripting] A Pythonic interface to OpenBabel

[Noel O'B. <no...@ca...>]

On Tue, 2005-12-06 at 10:11 -0800, S Joshua Swamidass wrote:
> Noel, how did you fix the compile/linking problems?
I didn't fix those problems, but those problems only occur when you try
file conversion. So I have just written code to use the other parts, so
far.

> Josh
> 
> On 06/12/05, Noel O'Boyle <no...@ca...> wrote:
> > On Tue, 2005-12-06 at 09:16 -0500, Geoffrey Hutchison wrote:
> > > On Dec 6, 2005, at 5:02 AM, Noel O'Boyle wrote:
> > >
> > > > It should be possible to access 90% of the OpenBabel library following
> > > > these guidelines, and the other 10% can be accessed by 'advanced
> > > > users'
> > > > through the original OpenBabel objects, atom.OBAtom in the example
> > > > above.
> > >
> > > Sounds great! This certainly offers better syntax than the original
> > > SWIG wrapper. Is it relatively easy to update the code as the library
> > > changes?
> >
> > Well, I won't argue "better" or "worse", but "easier for me" is
> > certainly true. I'll let you know how easy it is to update once I've
> > written it all - at the moment, piece of cake :-) (But I get your point)
> >
> > > > I have done some work towards this goal (attached) although file
> > > > format
> > > > conversion will have to wait until OpenBabel is ready.
> > >
> > > > If there is some interest in this, I will put it up on SourceForge.
> > > > Comments, suggestions
> > > > are welcome, of course.
> > >
> > > Can this be included alongside the current SWIG wrapper in scripts/
> > > python? I'd ultimately like to have a separate download "package" for
> > > each scripting module on SourceForge, but want to get slightly more
> > > bulletproof code first.
> >
> > Well, that'd be great.
> >
> > > Thanks!
> > > -Geoff
> > >
> >
> >
> >
> > -------------------------------------------------------
> > This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
> > for problems?  Stop!  Download the new AJAX search engine that makes
> > searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
> > http://ads.osdn.com/?ad_id=7637&alloc_id=16865&op=click
> > _______________________________________________
> > OpenBabel-scripting mailing list
> > Ope...@li...
> > https://lists.sourceforge.net/lists/listinfo/openbabel-scripting
> >

Re: [openbabel-scripting] A Pythonic interface to OpenBabel

[S J. S. <swa...@gm...>]

Noel, how did you fix the compile/linking problems?

Josh

On 06/12/05, Noel O'Boyle <no...@ca...> wrote:
> On Tue, 2005-12-06 at 09:16 -0500, Geoffrey Hutchison wrote:
> > On Dec 6, 2005, at 5:02 AM, Noel O'Boyle wrote:
> >
> > > It should be possible to access 90% of the OpenBabel library followin=
g
> > > these guidelines, and the other 10% can be accessed by 'advanced
> > > users'
> > > through the original OpenBabel objects, atom.OBAtom in the example
> > > above.
> >
> > Sounds great! This certainly offers better syntax than the original
> > SWIG wrapper. Is it relatively easy to update the code as the library
> > changes?
>
> Well, I won't argue "better" or "worse", but "easier for me" is
> certainly true. I'll let you know how easy it is to update once I've
> written it all - at the moment, piece of cake :-) (But I get your point)
>
> > > I have done some work towards this goal (attached) although file
> > > format
> > > conversion will have to wait until OpenBabel is ready.
> >
> > > If there is some interest in this, I will put it up on SourceForge.
> > > Comments, suggestions
> > > are welcome, of course.
> >
> > Can this be included alongside the current SWIG wrapper in scripts/
> > python? I'd ultimately like to have a separate download "package" for
> > each scripting module on SourceForge, but want to get slightly more
> > bulletproof code first.
>
> Well, that'd be great.
>
> > Thanks!
> > -Geoff
> >
>
>
>
> -------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc. Do you grep through log fi=
les
> for problems?  Stop!  Download the new AJAX search engine that makes
> searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
> http://ads.osdn.com/?ad_id=3D7637&alloc_id=3D16865&op=3Dclick
> _______________________________________________
> OpenBabel-scripting mailing list
> Ope...@li...
> https://lists.sourceforge.net/lists/listinfo/openbabel-scripting
>

Re: [openbabel-scripting] A Pythonic interface to OpenBabel

[Noel O'B. <no...@ca...>]

On Tue, 2005-12-06 at 09:16 -0500, Geoffrey Hutchison wrote:
> On Dec 6, 2005, at 5:02 AM, Noel O'Boyle wrote:
> 
> > It should be possible to access 90% of the OpenBabel library following
> > these guidelines, and the other 10% can be accessed by 'advanced  
> > users'
> > through the original OpenBabel objects, atom.OBAtom in the example
> > above.
> 
> Sounds great! This certainly offers better syntax than the original  
> SWIG wrapper. Is it relatively easy to update the code as the library  
> changes?

Well, I won't argue "better" or "worse", but "easier for me" is
certainly true. I'll let you know how easy it is to update once I've
written it all - at the moment, piece of cake :-) (But I get your point)

> > I have done some work towards this goal (attached) although file  
> > format
> > conversion will have to wait until OpenBabel is ready.
> 
> > If there is some interest in this, I will put it up on SourceForge.  
> > Comments, suggestions
> > are welcome, of course.
> 
> Can this be included alongside the current SWIG wrapper in scripts/ 
> python? I'd ultimately like to have a separate download "package" for  
> each scripting module on SourceForge, but want to get slightly more  
> bulletproof code first.

Well, that'd be great.

> Thanks!
> -Geoff
> 

Re: [openbabel-scripting] A Pythonic interface to OpenBabel

[Geoffrey H. <ge...@ge...>]

On Dec 6, 2005, at 5:02 AM, Noel O'Boyle wrote:

> It should be possible to access 90% of the OpenBabel library following
> these guidelines, and the other 10% can be accessed by 'advanced  
> users'
> through the original OpenBabel objects, atom.OBAtom in the example
> above.

Sounds great! This certainly offers better syntax than the original  
SWIG wrapper. Is it relatively easy to update the code as the library  
changes?

> I have done some work towards this goal (attached) although file  
> format
> conversion will have to wait until OpenBabel is ready.

> If there is some interest in this, I will put it up on SourceForge.  
> Comments, suggestions
> are welcome, of course.

Can this be included alongside the current SWIG wrapper in scripts/ 
python? I'd ultimately like to have a separate download "package" for  
each scripting module on SourceForge, but want to get slightly more  
bulletproof code first.

Thanks!
-Geoff

[openbabel-scripting] A Pythonic interface to OpenBabel

[Noel O'B. <no...@ca...>]

Dear all,

I have been thinking about developing a more Pythonic interface to
OpenBabel, PyOpenBabel, along the same idea as PyDaylight, for the
Daylight library (Andrew Dalke). It is a wrapper around the SWIG
interface to OpenBabel.

The motivation is:
(1) to provide easy access to attributes and methods (no need for
GetVar())
(2) to keep things simple (only one good way to do things)
(3) things should work 'as expected' in Python (iterators, etc.)

For example, the following sort of commands should work:
-----------------------
for atom in molecule.atoms: (or even 'for atom in molecule' !)
	print atom.coords
	print atom.element

text = molecule.write("SDF")
molecule.write("SMILES",filename="mymolecule.smi")

print vars(atom).keys() # Show the attributes of 'atom', like dir(atom)

['index', 'hyb', 'valence', 'coordidx', 'formalcharge', 'idx',
'implicitvalence', 'type', 'partialcharge', 'heterovalence', 'isotope',
'cidx', 'coords', 'vector', 'atomicnum', 'atomicmass', 'OBAtom', 'spin',
'exactmass', 'heavyvalence', 'data']
-----------------------

It should be possible to access 90% of the OpenBabel library following
these guidelines, and the other 10% can be accessed by 'advanced users'
through the original OpenBabel objects, atom.OBAtom in the example
above.

I have done some work towards this goal (attached) although file format
conversion will have to wait until OpenBabel is ready. If there is some
interest in this, I will put it up on SourceForge. Comments, suggestions
are welcome, of course.

Regards,
Noel

Re: [openbabel-scripting] Re: "did not load properly"

[Geoffrey H. <ge...@ge...>]

On Dec 2, 2005, at 3:21 PM, Thomas Huebner wrote:

> The error message probably comes from LoadFormatFiles in  
> obconversion.cpp (if that helps to debug;-)
> ...
>            cerr << *itr << " did not load properly" << endl;

Yes, thanks.

> which would scream if dlopen in dlhandler_unix.cpp fails.
> If I let dlerror() print, I get the message:
> /usr/local/lib/openbabel/chem3dformat.so: undefined symbol:  
> _ZN9OpenBabel8OBFormat22TargetClassDescriptionEv

Well, that was very helpful -- it's a "mangled" C++ debugging symbol.  
If you use c++filt to decode, you get:
OpenBabel::OBFormat::TargetClassDescription()

Hmm, I'll have to see what's going on there. Maybe the Linux loader  
isn't resolving that symbol in the libopenbabel library?

Thanks,
-Geoff

[openbabel-scripting] Re: "did not load properly"

[Thomas H. <thu...@t2...>]

Hello,

I'm having the same "*.so did not load properly" problem with python, ope=
nbabel and linux (suse 9.1).

Geoffrey Hutchison wrote:=20

>> Do you think we coudl statically link them? Where is the code that =20
>> loads
>> them dynamically?
>
>You don't want to statically link them. In fact, there are problems =20
>with the static builds of Open Babel because the linking is tricky.
>
>The C++ code that loads the file formats is in src/=20
>dlhandler_unix.cpp. It has no error messages like the ones you =20
>report, which is why I don't know what to debug. :-(

The error message probably comes from LoadFormatFiles in obconversion.cpp=
 (if that helps to debug;-)
      =20
for(itr=3Dfiles.begin();itr!=3Dfiles.end();itr++)
    {
        if(DLHandler::openLib(*itr))
            count++;
        else
            cerr << *itr << " did not load properly" << endl;
     }

which would scream if dlopen in dlhandler_unix.cpp fails.=20

If I let dlerror() print, I get the message:
/usr/local/lib/openbabel/chem3dformat.so: undefined symbol: _ZN9OpenBabel=
8OBFormat22TargetClassDescriptionEv
Which doesn't mean anything to me, but maybe there is someone out there s=
marter?

Having a working python binding, would be really lovely, txs.

Bye
Thomas=20

--=20
Dr. Thomas H=FCbner
T2-Consult Wein & H=FCbner,  M=FCnchen

Phone: +49 (89) 21 58 16 80
http://www.t2-consult.de

Re: [openbabel-scripting] Re: "did not load properly"

[Noel O'B. <no...@ca...>]

Done. Same problem as in Python, it seems.


$ perl Makefile.PL; make; make test

Checking if your kit is complete...
Warning: the following files are missing in your kit:
        openbabel.pm
Please inform the author.
Writing Makefile for Chemistry::OpenBabel
cp OpenBabel.pm blib/lib/Chemistry/OpenBabel.pm
g++ -c  -I../../src -D_REENTRANT -D_GNU_SOURCE -DTHREADS_HAVE_PIDS
-DDEBIAN -fno-strict-aliasing -I/usr/local/include -D_LARGEFILE_SOURCE
-D_FILE_OFFSET_BITS=64 -O2   -DVERSION=\"1.0.0\" -DXS_VERSION=\"1.0.0\"
-fPIC "-I/usr/lib/perl/5.8/CORE"   openbabel_perl.cpp
Running Mkbootstrap for Chemistry::OpenBabel ()
chmod 644 OpenBabel.bs
rm -f blib/arch/auto/Chemistry/OpenBabel/OpenBabel.so
LD_RUN_PATH="" g++  -shared -L/usr/local/lib
openbabel_perl.o ../../src/.libs/libopenbabel.a -lz  -o
blib/arch/auto/Chemistry/OpenBabel/OpenBabel.so
chmod 755 blib/arch/auto/Chemistry/OpenBabel/OpenBabel.so
cp OpenBabel.bs blib/arch/auto/Chemistry/OpenBabel/OpenBabel.bs
chmod 644 blib/arch/auto/Chemistry/OpenBabel/OpenBabel.bs
PERL_DL_NONLAZY=1 /usr/bin/perl "-MExtUtils::Command::MM" "-e"
"test_harness(0, 'blib/lib', 'blib/arch')" t/*.t
t/example..../usr/local/lib/openbabel/APIInterface.so did not load
properly
/usr/local/lib/openbabel/CSRformat.so did not load properly
/usr/local/lib/openbabel/PQSformat.so did not load properly
/usr/local/lib/openbabel/alchemyformat.so did not load properly
/usr/local/lib/openbabel/amberformat.so did not load properly
/usr/local/lib/openbabel/balstformat.so did not load properly
/usr/local/lib/openbabel/bgfformat.so did not load properly
t/example....ok 5/11/usr/local/lib/openbabel/boxformat.so did not load
properly
/usr/local/lib/openbabel/cacaoformat.so did not load properly
/usr/local/lib/openbabel/cacheformat.so did not load properly
/usr/local/lib/openbabel/carformat.so did not load properly
/usr/local/lib/openbabel/cccformat.so did not load properly
/usr/local/lib/openbabel/chem3dformat.so did not load properly
/usr/local/lib/openbabel/chemdrawformat.so did not load properly
/usr/local/lib/openbabel/chemtoolformat.so did not load properly
/usr/local/lib/openbabel/copyformat.so did not load properly
/usr/local/lib/openbabel/crkformat.so did not load properly
/usr/local/lib/openbabel/cssrformat.so did not load properly
/usr/local/lib/openbabel/dmolformat.so did not load properly
/usr/local/lib/openbabel/fastsearchformat.so did not load properly
/usr/local/lib/openbabel/featformat.so did not load properly
/usr/local/lib/openbabel/fhformat.so did not load properly
/usr/local/lib/openbabel/fingerprintformat.so did not load properly
/usr/local/lib/openbabel/freefracformat.so did not load properly
/usr/local/lib/openbabel/gamessformat.so did not load properly
/usr/local/lib/openbabel/gaussformat.so did not load properly
/usr/local/lib/openbabel/ghemicalformat.so did not load properly
/usr/local/lib/openbabel/gromos96format.so did not load properly
/usr/local/lib/openbabel/hinformat.so did not load properly
/usr/local/lib/openbabel/inchiformat.so did not load properly
/usr/local/lib/openbabel/jaguarformat.so did not load properly
/usr/local/lib/openbabel/mdlformat.so did not load properly
/usr/local/lib/openbabel/mmodformat.so did not load properly
/usr/local/lib/openbabel/mpdformat.so did not load properly
/usr/local/lib/openbabel/mol2format.so did not load properly
/usr/local/lib/openbabel/mopacformat.so did not load properly
/usr/local/lib/openbabel/mpqcformat.so did not load properly
/usr/local/lib/openbabel/nwchemformat.so did not load properly
/usr/local/lib/openbabel/pcmodelformat.so did not load properly
/usr/local/lib/openbabel/pdbformat.so did not load properly
/usr/local/lib/openbabel/povrayformat.so did not load properly
/usr/local/lib/openbabel/qchemformat.so did not load properly
/usr/local/lib/openbabel/reportformat.so did not load properly
/usr/local/lib/openbabel/rxnformat.so did not load properly
/usr/local/lib/openbabel/shelxformat.so did not load properly
/usr/local/lib/openbabel/smilesformat.so did not load properly
/usr/local/lib/openbabel/tinkerformat.so did not load properly
/usr/local/lib/openbabel/turbomoleformat.so did not load properly
/usr/local/lib/openbabel/unichemformat.so did not load properly
/usr/local/lib/openbabel/viewmolformat.so did not load properly
/usr/local/lib/openbabel/xedformat.so did not load properly
/usr/local/lib/openbabel/xyzformat.so did not load properly
/usr/local/lib/openbabel/yasaraformat.so did not load properly
/usr/local/lib/openbabel/zindoformat.so did not load properly
t/example....FAILED tests 8-9
        Failed 2/11 tests, 81.82% okay
Failed Test Stat Wstat Total Fail  Failed  List of Failed
-------------------------------------------------------------------------------
t/example.t               11    2  18.18%  8-9
Failed 1/1 test scripts, 0.00% okay. 2/11 subtests failed, 81.82% okay.
make: *** [test_dynamic] Error 255



On Thu, 2005-12-01 at 10:58 -0500, Geoffrey Hutchison wrote:
> On Dec 1, 2005, at 10:01 AM, Noel O'Boyle wrote:
> 
> > Testing Perl...phew, doesn't work for Perl either. :-)
> > cp openbabel.pm blib/lib/Chemistry/openbabel.pm
> ...
> > t/example....Can't locate Chemistry/OpenBabel.pm in @INC (@INC
> 
> (smacks forehead) No, that's a different bug. Mac OS X has a case- 
> preserving, case-insensitive filesystem.
> 
> So Chemistry/openbabel.pm and Chemistry/OpenBabel.pm are the same  
> file on OS X. Not, of course on Linux.
> 
> For the Perl module, try "make clean; mv openbabel.pm OpenBabel.pm"  
> and repeat the original build.
> 
> Thanks,
> -Geoff

Re: [openbabel-scripting] Re: "did not load properly"

[Geoffrey H. <ge...@ge...>]

On Dec 1, 2005, at 10:01 AM, Noel O'Boyle wrote:

> Testing Perl...phew, doesn't work for Perl either. :-)
> cp openbabel.pm blib/lib/Chemistry/openbabel.pm
...
> t/example....Can't locate Chemistry/OpenBabel.pm in @INC (@INC

(smacks forehead) No, that's a different bug. Mac OS X has a case- 
preserving, case-insensitive filesystem.

So Chemistry/openbabel.pm and Chemistry/OpenBabel.pm are the same  
file on OS X. Not, of course on Linux.

For the Perl module, try "make clean; mv openbabel.pm OpenBabel.pm"  
and repeat the original build.

Thanks,
-Geoff

Re: [openbabel-scripting] Re: "did not load properly"

[Noel O'B. <no...@ca...>]

Testing Perl...phew, doesn't work for Perl either. :-)

~/Tools/OpenBabel/openbabel/scripts/perl$ perl Makefile.PL; make; make 

test
Checking if your kit is complete...
Looks good
Writing Makefile for Chemistry::OpenBabel
cp openbabel.pm blib/lib/Chemistry/openbabel.pm
g++ -c  -I../../src -D_REENTRANT -D_GNU_SOURCE -DTHREADS_HAVE_PIDS
-DDEBIAN -fno-strict-aliasing -I/usr/local/include -D_LARGEFILE_SOURCE
-D_FILE_OFFSET_BITS=64 -O2   -DVERSION=\"1.0.0\" -DXS_VERSION=\"1.0.0\"
-fPIC "-I/usr/lib/perl/5.8/CORE"   openbabel_perl.cpp
Running Mkbootstrap for Chemistry::OpenBabel ()
chmod 644 OpenBabel.bs
rm -f blib/arch/auto/Chemistry/OpenBabel/OpenBabel.so
LD_RUN_PATH="" g++  -shared -L/usr/local/lib
openbabel_perl.o ../../src/.libs/libopenbabel.a -lz  -o
blib/arch/auto/Chemistry/OpenBabel/OpenBabel.so
chmod 755 blib/arch/auto/Chemistry/OpenBabel/OpenBabel.so
cp OpenBabel.bs blib/arch/auto/Chemistry/OpenBabel/OpenBabel.bs
chmod 644 blib/arch/auto/Chemistry/OpenBabel/OpenBabel.bs
PERL_DL_NONLAZY=1 /usr/bin/perl "-MExtUtils::Command::MM" "-e"
"test_harness(0, 'blib/lib', 'blib/arch')" t/*.t
t/example....Can't locate Chemistry/OpenBabel.pm in @INC (@INC
contains: /home/no228/Tools/OpenBabel/openbabel/scripts/perl/blib/lib /home/no228/Tools/OpenBabel/openbabel/scripts/perl/blib/arch /etc/perl /usr/local/lib/perl/5.8.4 /usr/local/share/perl/5.8.4 /usr/lib/perl5 /usr/share/perl5 /usr/lib/perl/5.8 /usr/share/perl/5.8 /usr/local/lib/site_perl . /etc/perl /usr/local/lib/perl/5.8.4 /usr/local/share/perl/5.8.4 /usr/lib/perl5 /usr/share/perl5 /usr/lib/perl/5.8 /usr/share/perl/5.8 /usr/local/lib/site_perl .) at t/example.t line 11.
BEGIN failed--compilation aborted at t/example.t line 11.
t/example....dubious
        Test returned status 2 (wstat 512, 0x200)
DIED. FAILED tests 1-11
        Failed 11/11 tests, 0.00% okay
Failed Test Stat Wstat Total Fail  Failed  List of Failed
-------------------------------------------------------------------------------
t/example.t    2   512    11   21 190.91%  1-11
Failed 1/1 test scripts, 0.00% okay. 11/11 subtests failed, 0.00% okay.
make: *** [test_dynamic] Error 2




On Wed, 2005-11-30 at 10:42 -0500, Geoffrey Hutchison wrote:
> On Nov 30, 2005, at 4:51 AM, Noel O'Boyle wrote:
> 
> > On Tue, 2005-11-29 at 17:33 -0500, Geoffrey Hutchison wrote:
> >> On Nov 29, 2005, at 11:07 AM, S Joshua Swamidass wrote:
> >>
> >>> But we still get a problem. Here is the output when you run teh
> >>> exmaple file after changing the name of openbabel.so
> >> ...
> >>> /home/dock/linux/lib/openbabel/APIInterface.so did not load properly
> >>> /home/dock/linux/lib/openbabel/CSRformat.so did not load properly
> >> ...
> >
> > Same here.
> 
> OK, I have one more question... If you try running the tests for the  
> Perl wrapper, do they work?
> 
> "perl Makefile.PL; make; make test"
> 
> I'm trying to sort out if this is a Python-specific problem or not.  
> Haven't found much. :-(
> 
> Thanks,
> -Geoff
> 
> 
> -------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
> for problems?  Stop!  Download the new AJAX search engine that makes
> searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
> http://ads.osdn.com/?ad_id=7637&alloc_id=16865&op=click
> _______________________________________________
> OpenBabel-scripting mailing list
> Ope...@li...
> https://lists.sourceforge.net/lists/listinfo/openbabel-scripting

Re: [openbabel-scripting] Re: "did not load properly"

[Geoffrey H. <ge...@ge...>]

On Nov 30, 2005, at 1:57 PM, S Joshua Swamidass wrote:

> It will help knowing exactly how these delayed dynamic libray loads
> are made. Why aren't they all already part of the _openbabel.so?

The _openbabel.so is merely some "glue" code to allow the  
openbabel.py module to access the C++ library. So there are actually  
several dynamic library loads beyond just _openbabel.so. For example,  
libopenbabel.so is loaded.

The delayed loads are for the individual file format dynamic  
libraries. The OB-2.0 library structure separates them from the main  
library -- there are plenty of users who want use of the libopenbabel  
library but not necessarily all of the file format translators.

This library structure works just fine for C, C++, and the Perl  
wrapper AFAICT. And it works for me on Mac OS X with Python. I'm not  
sure what's different about Linux Python, or who can help with this  
necessarily.

> Do you think we coudl statically link them? Where is the code that  
> loads
> them dynamically?

You don't want to statically link them. In fact, there are problems  
with the static builds of Open Babel because the linking is tricky.

The C++ code that loads the file formats is in src/ 
dlhandler_unix.cpp. It has no error messages like the ones you  
report, which is why I don't know what to debug. :-(

-Geoff

Re: [openbabel-scripting] Re: "did not load properly"

[Geoffrey H. <ge...@ge...>]

On Nov 30, 2005, at 4:51 AM, Noel O'Boyle wrote:

> On Tue, 2005-11-29 at 17:33 -0500, Geoffrey Hutchison wrote:
>> On Nov 29, 2005, at 11:07 AM, S Joshua Swamidass wrote:
>>
>>> But we still get a problem. Here is the output when you run teh
>>> exmaple file after changing the name of openbabel.so
>> ...
>>> /home/dock/linux/lib/openbabel/APIInterface.so did not load properly
>>> /home/dock/linux/lib/openbabel/CSRformat.so did not load properly
>> ...
>
> Same here.

OK, I have one more question... If you try running the tests for the  
Perl wrapper, do they work?

"perl Makefile.PL; make; make test"

I'm trying to sort out if this is a Python-specific problem or not.  
Haven't found much. :-(

Thanks,
-Geoff

Re: [openbabel-scripting] Re: "did not load properly"

[Noel O'B. <no...@ca...>]

On Tue, 2005-11-29 at 17:33 -0500, Geoffrey Hutchison wrote:
> On Nov 29, 2005, at 11:07 AM, S Joshua Swamidass wrote:
> 
> > But we still get a problem. Here is the output when you run teh
> > exmaple file after changing the name of openbabel.so
> ...
> > /home/dock/linux/lib/openbabel/APIInterface.so did not load properly
> > /home/dock/linux/lib/openbabel/CSRformat.so did not load properly
> ...

Same here.

> I'm assuming that the babel binary itself loads the formats properly  
> and that these files do actually exist?

The files do actually exist. I'm not sure that all the formats are
loaded correctly as the test-suite fails pretty miserably...I've
attached the first five minutes [not sent to mailing list].

> Try setting the environment variable BABEL_LIBDIR:
> export BABEL_LIBDIR=/home/doc/linux/lib/openbabel

No effect.

> Other than that, it'd help to know what version of Linux (and
Python)  
> you're running. I'm not sure what it'd tell me, but it would give me  
> some info to try to find a similar machine for testing.

I have the same problem on Debian Sarge, for Python 2.3/2.4, even after
setting BABEL_LIBDIR=/usr/local/lib/openbabel. The files do exist: there
is an ".a",".la" and an ".so", for each.

Running 'file' on the .so:
/usr/local/lib/openbabel/xyzformat.so:         ELF 32-bit LSB shared
object, Intel 80386, version 1 (SYSV), not stripped

I also added /usr/local/lib/openbabel to LD_LIBRARY_PATH with no
improvement.

Regards,
Noel

[openbabel-scripting] Re: "did not load properly"

[Geoffrey H. <ge...@ge...>]

On Nov 29, 2005, at 11:07 AM, S Joshua Swamidass wrote:

> But we still get a problem. Here is the output when you run teh
> exmaple file after changing the name of openbabel.so
...
> /home/dock/linux/lib/openbabel/APIInterface.so did not load properly
> /home/dock/linux/lib/openbabel/CSRformat.so did not load properly
...

Well, it certainly works for me (or I wouldn't have shipped it).

I'm assuming that the babel binary itself loads the formats properly  
and that these files do actually exist?

Try setting the environment variable BABEL_LIBDIR:
export BABEL_LIBDIR=/home/doc/linux/lib/openbabel

Other than that, it'd help to know what version of Linux (and Python)  
you're running. I'm not sure what it'd tell me, but it would give me  
some info to try to find a similar machine for testing.

Cheers,
-Geoff

Re: [openbabel-scripting] import error

[S J. S. <swa...@gm...>]

You are getting the error because you still have openbabel.so in your
search path. so files have precidence over .py files.


Josh

On 29/11/05, Noel O'Boyle <no...@ca...> wrote:
> Dear all,
>
> After swapping "_openbabel.so" for "openbabel.so" in setup.py:
> ----------------------start of diff
> 7c7
> < obExtension =3D Extension('openbabel',
> ---
> > obExtension =3D Extension('_openbabel',
> ----------------------end of diff
>
> and after setting LD_LIBRARY_PATH to include /usr/local/lib, I still
> cannot import openbabel:
> ----------------------quote
> >>> import openbabel
> Traceback (most recent call last):
>  File "<stdin>", line 1, in ?
> ImportError: dynamic module does not define init function
> (initopenbabel)
> ----------------------unquote
>
> However, if I go to openbabel/scripts/python, and import the
> openbabel.py directly, I have no problems!
>
> Can anyone explain this?
>
> Regards,
> Noel
>
>
>
>
> -------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc. Do you grep through log fi=
les
> for problems?  Stop!  Download the new AJAX search engine that makes
> searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
> http://ads.osdn.com/?ad_id=3D7637&alloc_id=3D16865&op=3Dclick
> _______________________________________________
> OpenBabel-scripting mailing list
> Ope...@li...
> https://lists.sourceforge.net/lists/listinfo/openbabel-scripting
>

[openbabel-scripting] import error

[Noel O'B. <no...@ca...>]

Dear all,

After swapping "_openbabel.so" for "openbabel.so" in setup.py:
----------------------start of diff
7c7
< obExtension = Extension('openbabel',
---
> obExtension = Extension('_openbabel',
----------------------end of diff

and after setting LD_LIBRARY_PATH to include /usr/local/lib, I still
cannot import openbabel:
----------------------quote
>>> import openbabel
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
ImportError: dynamic module does not define init function
(initopenbabel)
----------------------unquote

However, if I go to openbabel/scripts/python, and import the
openbabel.py directly, I have no problems!

Can anyone explain this?

Regards,
Noel

Re: [openbabel-scripting] Re: ANNOUNCE: Open Babel 2.0 Released

[S J. S. <swa...@gm...>]

THis error is fixed by changing the name of openbabel.so to _openbabel.so.

But we still get a problem. Here is the output when you run teh
exmaple file after changing the name of openbabel.so

Any ideas?

Josh

=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D

> python examples.py
Should print 0 (atoms)
0
Should print 2 (atoms)
2
Should print 1 (bond)
1
/home/dock/linux/lib/openbabel/APIInterface.so did not load properly
/home/dock/linux/lib/openbabel/CSRformat.so did not load properly
/home/dock/linux/lib/openbabel/PQSformat.so did not load properly
/home/dock/linux/lib/openbabel/alchemyformat.so did not load properly
/home/dock/linux/lib/openbabel/amberformat.so did not load properly
/home/dock/linux/lib/openbabel/balstformat.so did not load properly
/home/dock/linux/lib/openbabel/bgfformat.so did not load properly
/home/dock/linux/lib/openbabel/boxformat.so did not load properly
/home/dock/linux/lib/openbabel/cacaoformat.so did not load properly
/home/dock/linux/lib/openbabel/cacheformat.so did not load properly
/home/dock/linux/lib/openbabel/carformat.so did not load properly
/home/dock/linux/lib/openbabel/cccformat.so did not load properly
/home/dock/linux/lib/openbabel/chem3dformat.so did not load properly
/home/dock/linux/lib/openbabel/chemdrawformat.so did not load properly
/home/dock/linux/lib/openbabel/chemtoolformat.so did not load properly
/home/dock/linux/lib/openbabel/copyformat.so did not load properly
/home/dock/linux/lib/openbabel/crkformat.so did not load properly
/home/dock/linux/lib/openbabel/cssrformat.so did not load properly
/home/dock/linux/lib/openbabel/dmolformat.so did not load properly
/home/dock/linux/lib/openbabel/fastsearchformat.so did not load properly
/home/dock/linux/lib/openbabel/featformat.so did not load properly
/home/dock/linux/lib/openbabel/fhformat.so did not load properly
/home/dock/linux/lib/openbabel/fingerprintformat.so did not load properly
/home/dock/linux/lib/openbabel/freefracformat.so did not load properly
/home/dock/linux/lib/openbabel/gamessformat.so did not load properly
/home/dock/linux/lib/openbabel/gaussformat.so did not load properly
/home/dock/linux/lib/openbabel/ghemicalformat.so did not load properly
/home/dock/linux/lib/openbabel/gromos96format.so did not load properly
/home/dock/linux/lib/openbabel/hinformat.so did not load properly
/home/dock/linux/lib/openbabel/inchiformat.so did not load properly
/home/dock/linux/lib/openbabel/jaguarformat.so did not load properly
/home/dock/linux/lib/openbabel/mdlformat.so did not load properly
/home/dock/linux/lib/openbabel/mmodformat.so did not load properly
/home/dock/linux/lib/openbabel/mpdformat.so did not load properly
/home/dock/linux/lib/openbabel/mol2format.so did not load properly
/home/dock/linux/lib/openbabel/mopacformat.so did not load properly
/home/dock/linux/lib/openbabel/mpqcformat.so did not load properly
/home/dock/linux/lib/openbabel/nwchemformat.so did not load properly
/home/dock/linux/lib/openbabel/pcmodelformat.so did not load properly
/home/dock/linux/lib/openbabel/pdbformat.so did not load properly
/home/dock/linux/lib/openbabel/povrayformat.so did not load properly
/home/dock/linux/lib/openbabel/qchemformat.so did not load properly
/home/dock/linux/lib/openbabel/reportformat.so did not load properly
/home/dock/linux/lib/openbabel/rxnformat.so did not load properly
/home/dock/linux/lib/openbabel/shelxformat.so did not load properly
/home/dock/linux/lib/openbabel/smilesformat.so did not load properly
/home/dock/linux/lib/openbabel/tinkerformat.so did not load properly
/home/dock/linux/lib/openbabel/turbomoleformat.so did not load properly
/home/dock/linux/lib/openbabel/unichemformat.so did not load properly
/home/dock/linux/lib/openbabel/viewmolformat.so did not load properly
/home/dock/linux/lib/openbabel/xedformat.so did not load properly
/home/dock/linux/lib/openbabel/xyzformat.so did not load properly
/home/dock/linux/lib/openbabel/yasaraformat.so did not load properly
/home/dock/linux/lib/openbabel/zindoformat.so did not load properly
Should print 5 (atoms)
0
Should print 9 (atoms) after adding hydrogens

=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D


On 29/11/05, Noel O'Boyle <no...@ca...> wrote:
> I'd just like to ditto the previous message. On Debian Sarge, after
> compiling everything and installing the python extensions (from CVS), I
> get the following message when trying to import openbabel:
>
> >>> import openbabel
> Traceback (most recent call last):
>  File "<stdin>", line 1, in ?
>  File "openbabel.py", line 4, in ?
>    import _openbabel
> ImportError: No module named _openbabel
>
> As Joshua points out, there is no _openbabel.so, only an openbabel.so.
>
> Regards,
> Noel
>
>
>
> -------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc. Do you grep through log fi=
les
> for problems?  Stop!  Download the new AJAX search engine that makes
> searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
> http://ads.osdn.com/?ad_id=3D7637&alloc_id=3D16865&op=3Dclick
> _______________________________________________
> OpenBabel-scripting mailing list
> Ope...@li...
> https://lists.sourceforge.net/lists/listinfo/openbabel-scripting
>

[openbabel-scripting] Re: ANNOUNCE: Open Babel 2.0 Released

[Noel O'B. <no...@ca...>]

I'd just like to ditto the previous message. On Debian Sarge, after
compiling everything and installing the python extensions (from CVS), I
get the following message when trying to import openbabel:

>>> import openbabel
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
  File "openbabel.py", line 4, in ?
    import _openbabel
ImportError: No module named _openbabel

As Joshua points out, there is no _openbabel.so, only an openbabel.so.

Regards,
Noel

Re: [openbabel-scripting] ANNOUNCE: Open Babel 2.0 Released

[Geoffrey H. <ge...@ge...>]

On Nov 26, 2005, at 9:47 PM, S Joshua Swamidass wrote:

> Looks good, but I found on small possible build/install bug. The
> python scripign libray compiles to:  openbabel.so isntead of
> _openbabel.so. Changing the name of the file fixes errors in loading
> the python module.

I'd be glad to get some feedback on building the openbabel-python  
module. If you have patches/suggestions for this, I'd really  
appreciate it. I'm familiar with the Perl packaging system, but the  
setup.py is still a bit out of my hands. Patches, complaints, etc.  
are quite welcome!

(I'm a Python newbie.)

In particular, I'd like to make sure the build works either for
* A user compiling from the openbabel source (i.e., setup.py should  
use ../../src/libopenbabel.*)
* A user building openbabel-python separately (i.e., using - 
lopenbabel and pkg-config)

My guess is that the scripting bindings will need faster releases  
than the main library for now. So #2 is important, but doesn't  
currently work.

Thanks,
-Geoff

[openbabel-scripting] ANNOUNCE: Open Babel 2.0 Released

[Geoffrey H. <ge...@ge...>]

I'm extremely proud to announce the release of Open Babel 2.0.0, the =20
latest stable version of the free chemistry file translation program =20
and chemistry software library. This release marks the fourth =20
"birthday" of the Open Babel project and a milestone for a stable, =20
flexible interface for developers and users alike.

OpenBabel is a project designed to pick up where Babel left off, as a =20=

cross-platform program and library designed to interconvert between =20
many file formats used in molecular modeling, computational chemistry =20=

and related areas.

Highlights of the 2.0 release include a new conversion framework =20
making it easier to develop new translators, dramatically improved =20
support for merging, splitting, and batch conversion, a framework for =20=

molecular fingerprints, similarity searching, a fast molecular =20
database format, support for Perl and Python scripting "wrappers," =20
automatic support for reading .gz (gzip) compressed files, support =20
for the new IUPAC/NIST InChI identifiers and more.

What's new? See the release notes at:
  http://openbabel.sourceforge.net/RELEASE.shtml

To download, see:
  http://prdownloads.sourceforge.net/openbabel/openbabel-2.0.0.tar.gz

For more information, see the project website at:
  http://openbabel.sourceforge.net/

Many thanks to all the contributors to Open Babel. Particular thanks =20
go to (in no particular order) Jean Br=E9fort, Nick England, David =20
Hoekman, Elmar Krieger, Christian Laggner, Chris Morley, and Chris =20
Swain for helping to get this release out the door.

Cheers,
-Geoff=

[openbabel-scripting] Re: [Open Babel] Cannot complie Python wrapper to the Open Babel .

[Geoffrey H. <ge...@ge...>]

On Oct 12, 2005, at 11:22 PM, Xiao Jianfeng wrote:

> Then I used the command "python setup.py build" to build the python  
> wrapper.
> I got  the follwing error message:

If you read the README for the Python wrapper, you'll see that you  
need to do a "make install" before you build the wrapper.  
Unfortunately, I don't know enough about the Python build scripts to  
have it search both the build directory and the standard system  
directories.

But if you run "make install" for the Open Babel library, then  
libopenbabel will be available system-wide, including for the Python  
wrapper.

Cheers,
-Geoff

[openbabel-scripting] Re: Scripting Progress

[Geoff H. <ge...@ge...>]

Hi folks,

Well, the 2005-07-01 snapshot is out, along with more recent  
scripting bindings for Python and Perl.

I can attest that the Perl module builds and loads correctly. I'm in  
the process of writing some tests and examples for it, and hopefully  
soon some work to help it integrate with PerlMol. In short, after  
building and installing the snapshot, "cd scripts/perl; perl  
Makefile.PL; make; make test" works perfectly.

I admit that I'm at a complete loss when building Python modules --  
it seems to compile correctly when I do it by hand, but the setup.py  
calls gcc instead of g++ for compilation. If anyone has suggestions  
on writing setup.py or working with Python distutils, please e-mail me.

I also would like some feedback on what functionality people want  
from a scripting wrapper. Clearly access to mol.h and obconversion.h  
(the new conversion framework) are key. Anything else?

Cheers,
-Geoff

[openbabel-scripting] Re: openbabel and swig

[Geoff H. <ge...@ge...>]

On Jun 13, 2005, at 5:49 PM, S Joshua Swamidass wrote:

> Okay, let me look through the new stuff. As i mentioned, the last time
> i looked at it a bit back. I'll start seeing what works, doesn't work
> and get back to you soon.

Well, I have good news on that front. I tried a few things last night  
and just committed some changes. The Python wrapper will compile,  
link and import into Python. (This is in the openbabel-new-framework  
CVS branch -- I'd be glad to send a snapshot if someone doesn't want  
to access the CVS version.)

At this point, I hope people can run with it? I don't know enough  
about Python or what functionality people want to write a decent test  
script. Hopefully there's now enough that people can get some work done.

% ob-newf: cd src/
% src: swig -c++ -modern -python -o openbabel_python.cpp openbabel.i
[various warnings]
% src: g++ -c openbabel_python.cpp -DHAVE_CONFIG_H -I/sw/include/ 
python2.4 -I/sw/lib/python2.4/config -I.
% src: g++ -bundle -undefined supress -flat_namespace .libs/*.o  
math/.libs/*.o openbabel_python.o -o _openbabel.so

[That last line is for linking on Mac OS X -- change as needed for  
your platform.]

% ob-newf: python
Python 2.4.1 (#1, May 11 2005, 10:46:27)
[GCC 4.0.0 20041026 (Apple Computer, Inc. build 4061)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
 >>> import openbabel
 >>>

Cheers,
-Geoff

Re: [openbabel-scripting] openbabel and swig

[Geoff H. <ge...@ge...>]

On Jun 13, 2005, at 1:43 PM, S Joshua Swamidass wrote:

> Let me point out I'm only familiar wiht python, not perl at all. So
> everything i say here has that slant.

That's fine. I think right now, there's more interest in Python  
wrappers than Perl.

> One sticky issue in particular is the '>>' and '<<' methods of
> ... Molecule input/output in openbable relies heavily on this
> feaure.

Not anymore. These are removed completely in the 2.0 development  
code. There are probably ways to bring them back, but honestly, it's  
nice to separate the conversion code completely from an OBMol.

Even in earlier versions, the << and >> operators were not  
extensively used by internal code.

> will allow open bable to read a molecule from any
> file-emulating-object of the scripting target language, not just real
> files (very important for python).

It's already there. If you pass me a C++ iostream, then I'll hand it  
off to the new OBConversion framework and it'll go to work. I would  
assume that should handle most cases for Python, though I'll admit I  
haven't extensively looked at the new SWIG wrappers.

> The other major problem i was stuck on was compiling the wrapper swig
> produced. I never got to the bottom of which i couldn't. I can go back
> and look again. In my (limited) experience, this is a sign swig
> misunderstood some more complex uses of c++ syntax.

No, it was a problem with the declarations for C replacement  
functions missing on some platforms. SWIG saw them in the OpenBabel  
namespace and had problems. First I updated the development version  
to use std:: so that SWIG would correctly ignore them. Now, they're  
out of mol.h entirely.

Please take a look at the latest CVS snapshot on the openbabel-new- 
framework branch or the CVS development snapshot. There's even a SWIG  
openbabel.i file there. :-)

At the very least, I can vouch that the wrapper compiles correctly.

Cheers,
-Geoff