rdkit-devel — Developers information and discussion.

Re: [Rdkit-devel] More build fun with RHEL 6

[Greg L. <gre...@gm...>]

I'd like to have it work on EL6; that's still pretty common.
I will make some time to investigate either this evening or tomorrow.

Thanks for narrowing it down so much.
-greg


On Thu, Apr 14, 2016 at 12:18 PM, Gianluca Sforna <gi...@gm...> wrote:

> For the records, Build with RHEL/CentOS 7 succeded, this leaves us
> with only one platform where it fails; if we think the older boost
> version cannot be supported, I can just stick to an older release for
> EL6
>
> On Thu, Apr 14, 2016 at 11:47 AM, Gianluca Sforna <gi...@gm...>
> wrote:
> > On Thu, Apr 14, 2016 at 11:37 AM, Greg Landrum <gre...@gm...>
> wrote:
> >> Which version of boost are you using?
> >
> > RPMs are built against the packages in the repositories (stock+updates).
> >
> > Right now it means 1.41.0 for RHEL6, 1.53 for RHEL 7 (building now,
> > will let you know how it goes)
>
>
>
> --
> Gianluca Sforna
>
> http://plus.google.com/+gianlucasforna - http://twitter.com/giallu
> Tinker Garage - http://tinkergarage.it
>

Re: [Rdkit-devel] More build fun with RHEL 6

[Gianluca S. <gi...@gm...>]

For the records, Build with RHEL/CentOS 7 succeded, this leaves us
with only one platform where it fails; if we think the older boost
version cannot be supported, I can just stick to an older release for
EL6

On Thu, Apr 14, 2016 at 11:47 AM, Gianluca Sforna <gi...@gm...> wrote:
> On Thu, Apr 14, 2016 at 11:37 AM, Greg Landrum <gre...@gm...> wrote:
>> Which version of boost are you using?
>
> RPMs are built against the packages in the repositories (stock+updates).
>
> Right now it means 1.41.0 for RHEL6, 1.53 for RHEL 7 (building now,
> will let you know how it goes)



-- 
Gianluca Sforna

http://plus.google.com/+gianlucasforna - http://twitter.com/giallu
Tinker Garage - http://tinkergarage.it

Re: [Rdkit-devel] More build fun with RHEL 6

[Gianluca S. <gi...@gm...>]

On Thu, Apr 14, 2016 at 11:37 AM, Greg Landrum <gre...@gm...> wrote:
> Which version of boost are you using?

RPMs are built against the packages in the repositories (stock+updates).

Right now it means 1.41.0 for RHEL6, 1.53 for RHEL 7 (building now,
will let you know how it goes)

Re: [Rdkit-devel] More build fun with RHEL 6

[Greg L. <gre...@gm...>]

Gianluca,

Which version of boost are you using?

-greg


On Thu, Apr 14, 2016 at 11:33 AM, Gianluca Sforna <gi...@gm...> wrote:

> The RPM build for Fedora worked so I moved to RHEL/CentOS and got a new
> error:
>
>
> In file included from /usr/include/boost/python/to_python_indirect.hpp:10,
>                  from
> /usr/include/boost/python/converter/arg_to_python.hpp:10,
>                  from /usr/include/boost/python/call.hpp:15,
>                  from /usr/include/boost/python/object_core.hpp:12,
>                  from /usr/include/boost/python/args.hpp:25,
>                  from /usr/include/boost/python.hpp:11,
>                  from
>
> /builddir/build/BUILD/rdkit-Release_2016_03_1b1/python2/Code/RDBoost/python.h:2,
>                  from
>
> /builddir/build/BUILD/rdkit-Release_2016_03_1b1/python2/Code/GraphMol/FilterCatalog/Wrap/FilterCatalog.cpp:31:
> /usr/include/boost/python/object/pointer_holder.hpp: In member
> function 'void* boost::python::objects::pointer_holder<Pointer,
> Value>::holds
> (boost::python::type_info, bool) [with Pointer =
> boost::shared_ptr<const RDKit::FilterCatalogEntry>, Value = const
> RDKit::FilterCatalogEntry
> ]':
>
> /builddir/build/BUILD/rdkit-Release_2016_03_1b1/python2/Code/GraphMol/FilterCatalog/Wrap/FilterCatalog.cpp:463:
>   instantiated from here
> /usr/include/boost/python/object/pointer_holder.hpp:145: error:
> invalid conversion from 'const void*' to 'void*'
> /usr/include/boost/python/object/pointer_holder.hpp:145: error:
> initializing argument 1 of 'void*
> boost::python::objects::find_dynamic_type(void*,
> boost::python::type_info, boost::python::type_info)'
> /usr/include/boost/python/object/pointer_holder.hpp:145: error:
> invalid conversion from 'const void*' to 'void*'
>
> anything we can do to make it work on this older compiler?
>
> Talking about this, I'd like to note the compilation is pretty noisy
> because of some pragmas not supported by GCC until version 4.6. I
> applied the attached patch to silence them.
>
>
>
> --
> Gianluca Sforna
>
> http://plus.google.com/+gianlucasforna - http://twitter.com/giallu
> Tinker Garage - http://tinkergarage.it
>
>
> ------------------------------------------------------------------------------
> Find and fix application performance issues faster with Applications
> Manager
> Applications Manager provides deep performance insights into multiple
> tiers of
> your business applications. It resolves application problems quickly and
> reduces your MTTR. Get your free trial!
> https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
> _______________________________________________
> Rdkit-devel mailing list
> Rdk...@li...
> https://lists.sourceforge.net/lists/listinfo/rdkit-devel
>
>

[Rdkit-devel] More build fun with RHEL 6

[Gianluca S. <gi...@gm...>]

The RPM build for Fedora worked so I moved to RHEL/CentOS and got a new error:


In file included from /usr/include/boost/python/to_python_indirect.hpp:10,
                 from /usr/include/boost/python/converter/arg_to_python.hpp:10,
                 from /usr/include/boost/python/call.hpp:15,
                 from /usr/include/boost/python/object_core.hpp:12,
                 from /usr/include/boost/python/args.hpp:25,
                 from /usr/include/boost/python.hpp:11,
                 from
/builddir/build/BUILD/rdkit-Release_2016_03_1b1/python2/Code/RDBoost/python.h:2,
                 from
/builddir/build/BUILD/rdkit-Release_2016_03_1b1/python2/Code/GraphMol/FilterCatalog/Wrap/FilterCatalog.cpp:31:
/usr/include/boost/python/object/pointer_holder.hpp: In member
function 'void* boost::python::objects::pointer_holder<Pointer,
Value>::holds
(boost::python::type_info, bool) [with Pointer =
boost::shared_ptr<const RDKit::FilterCatalogEntry>, Value = const
RDKit::FilterCatalogEntry
]':
/builddir/build/BUILD/rdkit-Release_2016_03_1b1/python2/Code/GraphMol/FilterCatalog/Wrap/FilterCatalog.cpp:463:
  instantiated from here
/usr/include/boost/python/object/pointer_holder.hpp:145: error:
invalid conversion from 'const void*' to 'void*'
/usr/include/boost/python/object/pointer_holder.hpp:145: error:
initializing argument 1 of 'void*
boost::python::objects::find_dynamic_type(void*,
boost::python::type_info, boost::python::type_info)'
/usr/include/boost/python/object/pointer_holder.hpp:145: error:
invalid conversion from 'const void*' to 'void*'

anything we can do to make it work on this older compiler?

Talking about this, I'd like to note the compilation is pretty noisy
because of some pragmas not supported by GCC until version 4.6. I
applied the attached patch to silence them.



-- 
Gianluca Sforna

http://plus.google.com/+gianlucasforna - http://twitter.com/giallu
Tinker Garage - http://tinkergarage.it

Re: [Rdkit-devel] Beta of Q1 2016 release now up

[Greg L. <gre...@gm...>]

On Wed, Apr 13, 2016 at 8:43 PM, Gianluca Sforna <gi...@gm...> wrote:

> On Wed, Apr 13, 2016 at 12:27 PM, Greg Landrum <gre...@gm...>
> wrote:
> > Wow, that's a "typing in the wrong buffer" bug.
> > Great catch.
>
> :)
>
> Another couple points.
> I noticed the RDKit_Overview.pdf file is not there anymore. Was it
> integrated in the main manual, moved somewhere else or just removed?
>

The information in the pdf and the odb source for it were pretty out of
date, they've been removed. Docs/Book/Overview.md and the other contents of
the Docs/Book directory are a better information source.


> Regarding PostgreSQL cartridge, I expected the build to fail since I
> knew it was ported to cmake; instead, the older recipe just worked, I
> don't know if this is intended or not.
>

I didn't see a reason to immediately remove the old cartridge build make
system and break things for people who are used to using it. It will be
deprecated and removed in a future release.


> Finally, Docs were not building anymore, it seems some stray char was
> left in a file, so I needed to apply this:
>

Fixed.

-greg

Re: [Rdkit-devel] Beta of Q1 2016 release now up

[Gianluca S. <gi...@gm...>]

On Wed, Apr 13, 2016 at 12:27 PM, Greg Landrum <gre...@gm...> wrote:
> Wow, that's a "typing in the wrong buffer" bug.
> Great catch.

:)

Another couple points.
I noticed the RDKit_Overview.pdf file is not there anymore. Was it
integrated in the main manual, moved somewhere else or just removed?

Regarding PostgreSQL cartridge, I expected the build to fail since I
knew it was ported to cmake; instead, the older recipe just worked, I
don't know if this is intended or not.

Finally, Docs were not building anymore, it seems some stray char was
left in a file, so I needed to apply this:

diff --git a/Docs/Book/Install.md b/Docs/Book/Install.md
index cd660cb..9c838f4 100644
--- a/Docs/Book/Install.md
+++ b/Docs/Book/Install.md
@@ -388,4 +388,4 @@ This document is copyright (C) 2012-2015 by Greg Landrum

 This work is licensed under the Creative Commons
Attribution-ShareAlike 3.0 License. To view a copy of this license,
visit <http://creativecommons.org/licenses/by-sa/3.0/> or send a
letter to Creative Commons, 543 Howard Street, 5th Floor, San
Francisco, California, 94105, USA.

-The intent of this license is similar to that of the RDKit itself. In
simple words: "Do whatever you want with it, but please give us some
credit."<9D>
+The intent of this license is similar to that of the RDKit itself. In
simple words: "Do whatever you want with it, but please give us some
credit."


Regards

G.



-- 
Gianluca Sforna

http://plus.google.com/+gianlucasforna - http://twitter.com/giallu
Tinker Garage - http://tinkergarage.it

Re: [Rdkit-devel] Beta of Q1 2016 release now up

[Greg L. <gre...@gm...>]

Wow, that's a "typing in the wrong buffer" bug.
Great catch.

I will check in the fix in the next couple of minutes.

On Wed, Apr 13, 2016 at 12:09 PM, Gianluca Sforna <gi...@gm...> wrote:

> I bisected with git the issue, it popped up after commit b76d1a9 last
> december.
>
> I know almost nothing about flex & bison, but I noticed a suspicious
> change in the diff, and indeed the following patch fixes the build:
>
>  diff --git a/Code/GraphMol/SmilesParse/smiles.ll
> b/Code/GraphMol/SmilesParse/smiles.ll
> index 3a2346e..1f1cd27 100644
> --- a/Code/GraphMol/SmilesParse/smiles.ll
> +++ b/Code/GraphMol/SmilesParse/smiles.ll
> @@ -1,4 +1,4 @@
> -%setpropoption reentrant
> +%option reentrant
>  %option bison-bridge
>  %option noyywrap
>
> Was it correct?
>
>
>
> On Wed, Apr 13, 2016 at 10:44 AM, Greg Landrum <gre...@gm...>
> wrote:
> > Hi Gianluca,
> >
> > On Tue, Apr 12, 2016 at 1:09 PM, Gianluca Sforna <gi...@gm...>
> wrote:
> >>
> >>
> >> But no joy still....
> >>
> >>
> >> [ 36%] Building CXX object
> >> Code/GraphMol/SmilesParse/CMakeFiles/SmilesParse.dir/lex.yysmarts.cpp.o
> >> cd
> >>
> /builddir/build/BUILD/rdkit-Release_2016_03_1b1/python2/Code/GraphMol/SmilesParse
> >> && /usr/bin/c++   -DRDK_64BIT_BUILD -DRDK_TEST_MULTITHR
> >> EADED -DRDK_USE_STRICT_ROTOR_DEFINITION -DSmilesParse_EXPORTS
> >> -I/usr/include/python2.7
> >> -I/builddir/build/BUILD/rdkit-Release_2016_03_1b1/pyt
> >> hon2/Code  -O2 -g -pipe -Wall -Werror=format-security
> >> -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector-strong
> >> --param=ssp-buffer-size=
> >> 4 -grecord-gcc-switches -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1
> >> -m64 -mtune=generic  -mpopcnt -Wno-deprecated -Wno-unused-function -f
> >> no-strict-aliasing -fPIC -Wall -Wextra -O2 -g -DNDEBUG -fPIC
> >> -DRDK_THREADSAFE_SSS -DBOOST_ALL_NO_LIB -o
> >> CMakeFiles/SmilesParse.dir/lex.yys
> >> marts.cpp.o -c
> >>
> /builddir/build/BUILD/rdkit-Release_2016_03_1b1/python2/Code/GraphMol/SmilesParse/lex.yysmarts.cpp
> >> smiles.ll:40:52: error: 'yyscan_t' has not been declared
> >>  size_t setup_smiles_string(const std::string &text,yyscan_t yyscanner){
> >>                                                     ^
> >> smiles.ll: In function 'size_t setup_smiles_string(const string&, int)':
> >> smiles.ll:49:46: error: too many arguments to function 'void*
> >> yysmiles_alloc(yy_size_t)'
> >>    buf = (char *) yysmiles_alloc(n ,yyscanner );
> >>                                               ^
> >>
> >>
> /builddir/build/BUILD/rdkit-Release_2016_03_1b1/python2/Code/GraphMol/SmilesParse/lex.yysmiles.cpp:326:7:
> >> note: declared here
> >>  void *yysmiles_alloc (yy_size_t  );
> >>        ^
> >> smiles.ll:69:45: error: too many arguments to function
> >> 'yy_buffer_state* yysmiles__scan_buffer(char*, yy_size_t)'
> >>    b = yysmiles__scan_buffer(buf,n ,yyscanner);
> >>                                              ^
> >>
> >>
> /builddir/build/BUILD/rdkit-Release_2016_03_1b1/python2/Code/GraphMol/SmilesParse/lex.yysmiles.cpp:322:17:
> >> note: declared here
> >>  YY_BUFFER_STATE yysmiles__scan_buffer (char *base,yy_size_t size  );
> >
> >
> >
> > Something odd is going on with the output of flex. The easiest way to
> "fix"
> > this is just to avoid the problem by not running flex/bison yourself.
> >
> > If that would be ok, you can just change the cmake command line argment
> > -D RDK_USE_FLEXBISON:BOOL=ON
> > to
> > -D RDK_USE_FLEXBISON:BOOL=OFF
> >
> > -greg
> >
>
>
>
> --
> Gianluca Sforna
>
> http://plus.google.com/+gianlucasforna - http://twitter.com/giallu
> Tinker Garage - http://tinkergarage.it
>

Re: [Rdkit-devel] Beta of Q1 2016 release now up

[Gianluca S. <gi...@gm...>]

I bisected with git the issue, it popped up after commit b76d1a9 last december.

I know almost nothing about flex & bison, but I noticed a suspicious
change in the diff, and indeed the following patch fixes the build:

 diff --git a/Code/GraphMol/SmilesParse/smiles.ll
b/Code/GraphMol/SmilesParse/smiles.ll
index 3a2346e..1f1cd27 100644
--- a/Code/GraphMol/SmilesParse/smiles.ll
+++ b/Code/GraphMol/SmilesParse/smiles.ll
@@ -1,4 +1,4 @@
-%setpropoption reentrant
+%option reentrant
 %option bison-bridge
 %option noyywrap

Was it correct?



On Wed, Apr 13, 2016 at 10:44 AM, Greg Landrum <gre...@gm...> wrote:
> Hi Gianluca,
>
> On Tue, Apr 12, 2016 at 1:09 PM, Gianluca Sforna <gi...@gm...> wrote:
>>
>>
>> But no joy still....
>>
>>
>> [ 36%] Building CXX object
>> Code/GraphMol/SmilesParse/CMakeFiles/SmilesParse.dir/lex.yysmarts.cpp.o
>> cd
>> /builddir/build/BUILD/rdkit-Release_2016_03_1b1/python2/Code/GraphMol/SmilesParse
>> && /usr/bin/c++   -DRDK_64BIT_BUILD -DRDK_TEST_MULTITHR
>> EADED -DRDK_USE_STRICT_ROTOR_DEFINITION -DSmilesParse_EXPORTS
>> -I/usr/include/python2.7
>> -I/builddir/build/BUILD/rdkit-Release_2016_03_1b1/pyt
>> hon2/Code  -O2 -g -pipe -Wall -Werror=format-security
>> -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector-strong
>> --param=ssp-buffer-size=
>> 4 -grecord-gcc-switches -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1
>> -m64 -mtune=generic  -mpopcnt -Wno-deprecated -Wno-unused-function -f
>> no-strict-aliasing -fPIC -Wall -Wextra -O2 -g -DNDEBUG -fPIC
>> -DRDK_THREADSAFE_SSS -DBOOST_ALL_NO_LIB -o
>> CMakeFiles/SmilesParse.dir/lex.yys
>> marts.cpp.o -c
>> /builddir/build/BUILD/rdkit-Release_2016_03_1b1/python2/Code/GraphMol/SmilesParse/lex.yysmarts.cpp
>> smiles.ll:40:52: error: 'yyscan_t' has not been declared
>>  size_t setup_smiles_string(const std::string &text,yyscan_t yyscanner){
>>                                                     ^
>> smiles.ll: In function 'size_t setup_smiles_string(const string&, int)':
>> smiles.ll:49:46: error: too many arguments to function 'void*
>> yysmiles_alloc(yy_size_t)'
>>    buf = (char *) yysmiles_alloc(n ,yyscanner );
>>                                               ^
>>
>> /builddir/build/BUILD/rdkit-Release_2016_03_1b1/python2/Code/GraphMol/SmilesParse/lex.yysmiles.cpp:326:7:
>> note: declared here
>>  void *yysmiles_alloc (yy_size_t  );
>>        ^
>> smiles.ll:69:45: error: too many arguments to function
>> 'yy_buffer_state* yysmiles__scan_buffer(char*, yy_size_t)'
>>    b = yysmiles__scan_buffer(buf,n ,yyscanner);
>>                                              ^
>>
>> /builddir/build/BUILD/rdkit-Release_2016_03_1b1/python2/Code/GraphMol/SmilesParse/lex.yysmiles.cpp:322:17:
>> note: declared here
>>  YY_BUFFER_STATE yysmiles__scan_buffer (char *base,yy_size_t size  );
>
>
>
> Something odd is going on with the output of flex. The easiest way to "fix"
> this is just to avoid the problem by not running flex/bison yourself.
>
> If that would be ok, you can just change the cmake command line argment
> -D RDK_USE_FLEXBISON:BOOL=ON
> to
> -D RDK_USE_FLEXBISON:BOOL=OFF
>
> -greg
>



-- 
Gianluca Sforna

http://plus.google.com/+gianlucasforna - http://twitter.com/giallu
Tinker Garage - http://tinkergarage.it

Re: [Rdkit-devel] Beta of Q1 2016 release now up

[Greg L. <gre...@gm...>]

Hi Gianluca,

On Tue, Apr 12, 2016 at 1:09 PM, Gianluca Sforna <gi...@gm...> wrote:

>
> But no joy still....
>
>
> [ 36%] Building CXX object
> Code/GraphMol/SmilesParse/CMakeFiles/SmilesParse.dir/lex.yysmarts.cpp.o
> cd
> /builddir/build/BUILD/rdkit-Release_2016_03_1b1/python2/Code/GraphMol/SmilesParse
> && /usr/bin/c++   -DRDK_64BIT_BUILD -DRDK_TEST_MULTITHR
> EADED -DRDK_USE_STRICT_ROTOR_DEFINITION -DSmilesParse_EXPORTS
> -I/usr/include/python2.7
> -I/builddir/build/BUILD/rdkit-Release_2016_03_1b1/pyt
> hon2/Code  -O2 -g -pipe -Wall -Werror=format-security
> -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector-strong
> --param=ssp-buffer-size=
> 4 -grecord-gcc-switches -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1
> -m64 -mtune=generic  -mpopcnt -Wno-deprecated -Wno-unused-function -f
> no-strict-aliasing -fPIC -Wall -Wextra -O2 -g -DNDEBUG -fPIC
> -DRDK_THREADSAFE_SSS -DBOOST_ALL_NO_LIB -o
> CMakeFiles/SmilesParse.dir/lex.yys
> marts.cpp.o -c
> /builddir/build/BUILD/rdkit-Release_2016_03_1b1/python2/Code/GraphMol/SmilesParse/lex.yysmarts.cpp
> smiles.ll:40:52: error: 'yyscan_t' has not been declared
>  size_t setup_smiles_string(const std::string &text,yyscan_t yyscanner){
>                                                     ^
> smiles.ll: In function 'size_t setup_smiles_string(const string&, int)':
> smiles.ll:49:46: error: too many arguments to function 'void*
> yysmiles_alloc(yy_size_t)'
>    buf = (char *) yysmiles_alloc(n ,yyscanner );
>                                               ^
>
> /builddir/build/BUILD/rdkit-Release_2016_03_1b1/python2/Code/GraphMol/SmilesParse/lex.yysmiles.cpp:326:7:
> note: declared here
>  void *yysmiles_alloc (yy_size_t  );
>        ^
> smiles.ll:69:45: error: too many arguments to function
> 'yy_buffer_state* yysmiles__scan_buffer(char*, yy_size_t)'
>    b = yysmiles__scan_buffer(buf,n ,yyscanner);
>                                              ^
>
> /builddir/build/BUILD/rdkit-Release_2016_03_1b1/python2/Code/GraphMol/SmilesParse/lex.yysmiles.cpp:322:17:
> note: declared here
>  YY_BUFFER_STATE yysmiles__scan_buffer (char *base,yy_size_t size  );



Something odd is going on with the output of flex. The easiest way to "fix"
this is just to avoid the problem by not running flex/bison yourself.

If that would be ok, you can just change the cmake command line argment
-D RDK_USE_FLEXBISON:BOOL=ON
to
-D RDK_USE_FLEXBISON:BOOL=OFF

-greg

Re: [Rdkit-devel] Beta of Q1 2016 release now up

[Gianluca S. <gi...@gm...>]

On Tue, Apr 12, 2016 at 11:51 AM, Gianluca Sforna <gi...@gm...> wrote:
> On Mon, Apr 11, 2016 at 1:20 PM, Greg Landrum <gre...@gm...> wrote:
>> This could be the issue fixed by this PR from Matt Swain:
>> https://github.com/rdkit/rdkit/pull/856
>
> Yeah, that's it. Thanks!

But no joy still....


[ 36%] Building CXX object
Code/GraphMol/SmilesParse/CMakeFiles/SmilesParse.dir/lex.yysmarts.cpp.o
cd /builddir/build/BUILD/rdkit-Release_2016_03_1b1/python2/Code/GraphMol/SmilesParse
&& /usr/bin/c++   -DRDK_64BIT_BUILD -DRDK_TEST_MULTITHR
EADED -DRDK_USE_STRICT_ROTOR_DEFINITION -DSmilesParse_EXPORTS
-I/usr/include/python2.7
-I/builddir/build/BUILD/rdkit-Release_2016_03_1b1/pyt
hon2/Code  -O2 -g -pipe -Wall -Werror=format-security
-Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector-strong
--param=ssp-buffer-size=
4 -grecord-gcc-switches -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1
-m64 -mtune=generic  -mpopcnt -Wno-deprecated -Wno-unused-function -f
no-strict-aliasing -fPIC -Wall -Wextra -O2 -g -DNDEBUG -fPIC
-DRDK_THREADSAFE_SSS -DBOOST_ALL_NO_LIB -o
CMakeFiles/SmilesParse.dir/lex.yys
marts.cpp.o -c /builddir/build/BUILD/rdkit-Release_2016_03_1b1/python2/Code/GraphMol/SmilesParse/lex.yysmarts.cpp
smiles.ll:40:52: error: 'yyscan_t' has not been declared
 size_t setup_smiles_string(const std::string &text,yyscan_t yyscanner){
                                                    ^
smiles.ll: In function 'size_t setup_smiles_string(const string&, int)':
smiles.ll:49:46: error: too many arguments to function 'void*
yysmiles_alloc(yy_size_t)'
   buf = (char *) yysmiles_alloc(n ,yyscanner );
                                              ^
/builddir/build/BUILD/rdkit-Release_2016_03_1b1/python2/Code/GraphMol/SmilesParse/lex.yysmiles.cpp:326:7:
note: declared here
 void *yysmiles_alloc (yy_size_t  );
       ^
smiles.ll:69:45: error: too many arguments to function
'yy_buffer_state* yysmiles__scan_buffer(char*, yy_size_t)'
   b = yysmiles__scan_buffer(buf,n ,yyscanner);
                                             ^
/builddir/build/BUILD/rdkit-Release_2016_03_1b1/python2/Code/GraphMol/SmilesParse/lex.yysmiles.cpp:322:17:
note: declared here
 YY_BUFFER_STATE yysmiles__scan_buffer (char *base,yy_size_t size  );




-- 
Gianluca Sforna

http://plus.google.com/+gianlucasforna - http://twitter.com/giallu
Tinker Garage - http://tinkergarage.it

Re: [Rdkit-devel] Beta of Q1 2016 release now up

[Gianluca S. <gi...@gm...>]

On Mon, Apr 11, 2016 at 1:20 PM, Greg Landrum <gre...@gm...> wrote:
> This could be the issue fixed by this PR from Matt Swain:
> https://github.com/rdkit/rdkit/pull/856

Yeah, that's it. Thanks!



-- 
Gianluca Sforna

http://plus.google.com/+gianlucasforna - http://twitter.com/giallu
Tinker Garage - http://tinkergarage.it

Re: [Rdkit-devel] Beta of Q1 2016 release now up

[Greg L. <gre...@gm...>]

Gianluca,

This could be the issue fixed by this PR from Matt Swain:
https://github.com/rdkit/rdkit/pull/856

-greg


On Mon, Apr 11, 2016 at 12:32 PM, Gianluca Sforna <gi...@gm...> wrote:

> On Thu, Apr 7, 2016 at 6:24 AM, Greg Landrum <gre...@gm...>
> wrote:
> > I have tagged a beta of the next RDKit release here:
> > https://github.com/rdkit/rdkit/releases/tag/Release_2016_03_1b1
>
>
> Tried to rebuild the RPMs but got this right on cmake:
>
> -- Found INCHI: /usr/lib64/libinchi.so
> CMake Error at External/INCHI-API/CMakeLists.txt:64 (install):
>   install TARGETS given target "Inchi" which does not exist in this
>   directory.
>
>
> cmake invocation as follow:
>
> /usr/bin/cmake -DCMAKE_C_FLAGS_RELEASE:STRING=-DNDEBUG
> -DCMAKE_CXX_FLAGS_RELEASE:STRING=-DNDEBUG
> -DCMAKE_Fortran_FLAGS_RELEASE:STRING=-DNDEBUG
> -DCMAKE_VERBOSE_MAKEFILE:BOOL=ON -DCMAKE_INSTALL_PREFIX:PATH=/usr
> -DINCLUDE_INSTALL_DIR:PATH=/usr/include
> -DLIB_INSTALL_DIR:PATH=/usr/lib64 -DSYSCONF_INSTALL_DIR:PATH=/etc
> -DSHARE_INSTALL_PREFIX:PATH=/usr/share -DLIB_SUFFIX=64
> -DBUILD_SHARED_LIBS:BOOL=ON -D RDK_INSTALL_INTREE=OFF -D
> CMAKE_BUILD_TYPE=RelWithDebInfo -D RDK_BUILD_INCHI_SUPPORT:BOOL=ON -D
> RDK_BUILD_THREADSAFE_SSS:BOOL=ON -D RDK_BUILD_CAIRO_SUPPORT:BOOL=ON -D
> RDK_INSTALL_STATIC_LIBS:BOOL=OFF -D RDK_USE_FLEXBISON:BOOL=ON -D
> RDK_TEST_MULTITHREADED:BOOL=ON -D
> PYTHON_EXECUTABLE:FILEPATH=/usr/bin/python2 .
>
> I'll dig in it later, in the meanwhile if you have ideas they will be
> very welcome.
>

Re: [Rdkit-devel] Beta of Q1 2016 release now up

[Gianluca S. <gi...@gm...>]

On Thu, Apr 7, 2016 at 6:24 AM, Greg Landrum <gre...@gm...> wrote:
> I have tagged a beta of the next RDKit release here:
> https://github.com/rdkit/rdkit/releases/tag/Release_2016_03_1b1


Tried to rebuild the RPMs but got this right on cmake:

-- Found INCHI: /usr/lib64/libinchi.so
CMake Error at External/INCHI-API/CMakeLists.txt:64 (install):
  install TARGETS given target "Inchi" which does not exist in this
  directory.


cmake invocation as follow:

/usr/bin/cmake -DCMAKE_C_FLAGS_RELEASE:STRING=-DNDEBUG
-DCMAKE_CXX_FLAGS_RELEASE:STRING=-DNDEBUG
-DCMAKE_Fortran_FLAGS_RELEASE:STRING=-DNDEBUG
-DCMAKE_VERBOSE_MAKEFILE:BOOL=ON -DCMAKE_INSTALL_PREFIX:PATH=/usr
-DINCLUDE_INSTALL_DIR:PATH=/usr/include
-DLIB_INSTALL_DIR:PATH=/usr/lib64 -DSYSCONF_INSTALL_DIR:PATH=/etc
-DSHARE_INSTALL_PREFIX:PATH=/usr/share -DLIB_SUFFIX=64
-DBUILD_SHARED_LIBS:BOOL=ON -D RDK_INSTALL_INTREE=OFF -D
CMAKE_BUILD_TYPE=RelWithDebInfo -D RDK_BUILD_INCHI_SUPPORT:BOOL=ON -D
RDK_BUILD_THREADSAFE_SSS:BOOL=ON -D RDK_BUILD_CAIRO_SUPPORT:BOOL=ON -D
RDK_INSTALL_STATIC_LIBS:BOOL=OFF -D RDK_USE_FLEXBISON:BOOL=ON -D
RDK_TEST_MULTITHREADED:BOOL=ON -D
PYTHON_EXECUTABLE:FILEPATH=/usr/bin/python2 .

I'll dig in it later, in the meanwhile if you have ideas they will be
very welcome.

[Rdkit-devel] Beta of Q1 2016 release now up

[Greg L. <gre...@gm...>]

Dear all,

I have tagged a beta of the next RDKit release here:
https://github.com/rdkit/rdkit/releases/tag/Release_2016_03_1b1

The relevant section of the release notes is below. Note that the list of
people in the acknowledgement section was mostly created automatically from
github. If you feel like I missed you, or if I got your name wrong, please
let me know so that I can update the release notes before the actual
release.

Unless someone specifically requests them, I don't plan to do binaries for
the beta.

Unless major problems are found, I plan to do the actual release late next
week. Between now and then I hope to get a few more bugs crushed, but there
shouldn't be any additional feature additions.

Best,
-greg


# Release_2016.03.1
(Changes relative to Release_2015.09.2)

## Important
In order to build the RDKit, it is now necessary to have at least v1.7 of
numpy installed.

## Acknowledgements:
Note: The RDKit has the wonderful "problem" that there are a lot of
contributors and it's tough for me to capture them all to put together
release
notes. I don't even know many of the contributors (which is *awesome!*)
The names here come largely from what I pull in an automated way from
github.
In cases where there's no real name listed in github, I either guessed
or used just the github alias in quotes. If I got it wrong, please let me
know!

Josep Arus, Nik Bates-Haus, Andrew Dalke, 'DoliathGavid', 'elcaceres', Peter
Gedeck, James Jeffryes, Brian Kelley, Rich Lewis, Daniel Lowe, 'maddogcz',
Kozo
Nishida, Michal Nowotka, Axel Pahl, Steven Roughley, Alexander Savelyev,
Nadine
Schneider, Teague Sterling, Nik Stiefl, Matt Swain, Eric Ting, Paolo Tosco,
Samo
Turk, JL Varjo, Riccardo Vianello

## Highlights:
- Improvements to the build system: it's now much easier to build with
InChI and/or Avalon support since cmake now knows how to fetch the
appropriate source code for you. Building the PostgreSQL cartridge is now
integrated into normal build process.
- Some improvements to molecule rendering and Jupyter notebook integration:
The new `Draw.PrepareMolForDrawing()` function takes care of standard tasks
like wedging bonds, kekulization, and adding chiral Hs.
`Draw.MolsToGridImage()` can generate SVGs and uses the new molecular
drawing code for PNGs when possible. The Jupyter notebook integration uses
the new drawing code when possible.
- Error and warning messages from the C++ core can now be displayed in the
Jupyter notebook

## Bug Fixes:
  - Sanitizer rejects higher valency halides
 (github issue #115 from dan2097)
  - Bad E/Z assignment from ctab
 (github issue #188 from greglandrum)
  - bad E/Z assignment from ctab
 (github issue #192 from greglandrum)
  - Documentation is still python2 specific.
 (github issue #374 from greglandrum)
  - SVG export - Python 3 support
 (github issue #398 from maddogcz)
  - FragmentOnBonds() producing incorrect chirality
 (github issue #511 from greglandrum)
  - Rings containing all dummy atoms with single bonds are flagged as
aromatic
 (github issue #518 from greglandrum)
  - IPython integration broken with latest Jupyter
 (github issue #666 from samoturk)
  - Added missing include/forward declarations
 (github pull #668 from ptosco)
  - Fixes a memory leak in fragmentMol
 (github pull #669 from bp-kelley)
  - resetVect option being ignored by reaccsToFingerprint()
 (github issue #671 from greglandrum)
  - failure in AddHs when addCoords is true and coords are all zero
 (github issue #678 from greglandrum)
  - 404 error for the link to Installation instructions
 (github issue #679 from EricTing)
  - Fix java8 build
 (github pull #681 from greglandrum)
  - Smiles containing "[as]" do not parse.
 (github issue #682 from greglandrum)
  - SMARTS reaction triggers invariant violation on chiral compounds
 (github issue #685 from JamesJeffryes)
  - partially specified chiral substructure queries don't work properly
 (github issue #688 from bp-kelley)
  - ExactMolWt ignoring the mass of the electron
 (github issue #694 from greglandrum)
  - Bad 1-4 bounds matrix elements in highly constrained system
 (github issue #696 from greglandrum)
  - More ChEMBL molecules that fail bounds smoothing
 (github issue #697 from greglandrum)
  - Molecule serialization doesn't read/write atomic numbers above 128
 (github issue #713 from greglandrum)
  - AddHs cip rank is declared <int> should be unsigned int?
 (github issue #717 from bp-kelley)
  - ensure line endings are handled consistently for all users
 (github pull #729 from rvianello)
  - Fixes return type of operator[] (fails on later clangs)
 (github pull #733 from bp-kelley)
  - Fix/thread safe localeswitcher line endings
 (github pull #743 from bp-kelley)
  - Fixes Boost 1.46 issues with type traits
 (github pull #748 from bp-kelley)
  - PR #749 causes seg faults on windows
 (github issue #750 from greglandrum)
  - Fixes notebook problems with newer jupyter installs
 (github pull #753 from bp-kelley)
  - Double bond geometry loss on calling removeHs
 (github issue #754 from sroughley)
  - Bug fix to getShortestPath
 (github pull #757 from JLVarjo)
  - reversed stereochemistry with sulfoxides and ring closures
 (github issue #760 from greglandrum)
  - libRDBoost.so.1: undefined symbol
 (github issue #762 from kozo2)
  - Removed -Xdoclint:none flag when packing org.RDKitDoc.jar
 (github pull #763 from undeadpixel)
  - AnyBond specification treated as single when joining rings in SMARTS
 (github issue #766 from teaguesterling)
  - CanonicalRankAtomsInFragment() leaks when called from Python
 (github issue #769 from greglandrum)
  - MolCanvas2D drawing upside down
 (github issue #774 from greglandrum)
  - Drawing single-atom molecules hangs.
 (github issue #781 from greglandrum)
  - chiral lexical order for ring closure after branch
 (github issue #786 from adalke)
  - surface -> self.surface
 (github pull #787 from mnowotka)
  - Chem.MolToSmarts param misnomer
 (github issue #792 from elcaceres)
  - Fixes MolToSmarts python docs
 (github pull #793 from bp-kelley)
  - npscorer.py: Py3 compat and importable from other locations
 (github #801 from apahl)
  - Pre-condition Violation: bad bond type
 (github issue #805 from nbateshaus)
  - rooted atom fingerprint non identical for the same molecules
 (github issue #811 from nisti74)
  - test60RunSingleReactant() not being run
 (github issue #825 from greglandrum)
  - PostgreSQL bug fixes
 (github pull #835 from ptosco)
  - Return value of NumRadicalElectrons and NumValenceElectrons should be
integer
 (github issue #846 from gedeck)

## New Features and Enhancements:
  - switch to using new version of avalon toolkit
 (github issue #382 from greglandrum)
  - MolDraw2D: Expand basic drawing api
 (github issue #417 from greglandrum)
  - MolDraw2D: add options
 (github issue #424 from greglandrum)
  - fixed FutureWarning in PeriodicTable.py
 (github pull #665 from richlewis42)
  - first pass, using google style
 (github pull #672 from greglandrum)
  - Use sets instead of and map. Minor comments cleanup.
 (github pull #675 from DoliathGavid)
  - Dev/squash msvc14 warnings
 (github pull #684 from bp-kelley)
  - Fix/stop unnecessary filtercatalog updates
 (github pull #690 from bp-kelley)
  - Add RDK_USE_BOOST_SERIALIZATION configure option (On by default)
 (github pull #691 from bp-kelley)
  - Minor optimizations of the force field minimization code, fix for issue
696
 (github pull #693 from greglandrum)
  - Include cis/trans stereochemistry when useChirality=true with the
morgan fingerprints
 (github issue #695 from greglandrum)
  - Fixed a couple of compilation warnings in Resonance.cpp/Resonance.h
 (github pull #701 from ptosco)
  - Dev/run single reactant
 (github pull #705 from bp-kelley)
  - Updates CMAKE_SOURCE_DIR to CMAKE_CURRENT_SOURCE_DIR
 (github pull #707 from bp-kelley)
  - Make LocaleSwitcher threadsafe
 (github issue #710 from greglandrum)
  - Exposes Get/Set Double, Int, Uint and bool props to molecules
 (github pull #711 from bp-kelley)
  - Speed up molblock generation
 (github pull #712 from greglandrum)
  - Expose generateOneProductSet?
 (github issue #721 from DoliathGavid)
  - Add a reader for FPB files (still experimental)
 (github pull #724 from greglandrum)
  - replace std::map::at with std::map::operator[]
 (github pull #730 from rvianello)
  - Fix/get double prop get props asdict
 (github pull #734 from bp-kelley)
  - Add support for Tversky similarity to the FPB reader
 (github pull #735 from greglandrum)
  - Fix ConformerParser to use const std::string &
 (github pull #737 from mcs07)
  - Fix/expose invariant exception
 (github pull #740 from bp-kelley)
  - Support CTABs where the second letter in atom symbols is capitalized
 (github issue #741 from greglandrum)
  - Adds support for capturing RDLogs in Python StdErr streams
 (github pull #749 from bp-kelley)
  - Allow adding Hs only to atoms matching a query operator
 (github issue #758 from greglandrum)
  - Add argument to addHs allowing only certain Hs to be considered
 (github pull #759 from greglandrum)
  - avoid the multiple definition of rdkitVersion/boostVersion
 (github pull #761 from rvianello)
  - cleanup possible pythonObjectToVect leaks in python wrappers
 (github issue #764 from greglandrum)
  - Stop possible core leaks in pythonObjectToVect()
 (github pull #770 from greglandrum)
  - Add C++ function to prepare mol for rendering
 (github issue #771 from greglandrum)
  - Prefer wedging bonds to Hs
 (github issue #772 from greglandrum)
  - Add prepareMolForDrawing() function to C++
 (github pull #775 from greglandrum)
  - Support blanks in MolsToGridImage()
 (github issue #776 from greglandrum)
  - A number of small additions and features to the drawing code
 (github pull #802 from greglandrum)
  - Support larger isotope deltas in the chirality assignment
 (github issue #803 from greglandrum)
  - Adds option RDK_USE_COMPLEX_ROTOR_DEFINITION
 (github pull #810 from bp-kelley)
  - add Draw.MolsToSVGGrid()
 (github pull #817 from greglandrum)
  - make Hs black instead of gray
 (github pull #819 from greglandrum)
  - Fix alignMols so that it takes into account of QueryAtoms and QueryBonds
 (github pull #821 from DoliathGavid)
  - feat/github831: Add getText() static method.
 (github pull #832 from greglandrum)
  - Add an unfolded count-based version of the RDKFingerprint
 (github pull #838 from NadineSchneider)
  - Add some utils functions to ChemReactions
 (github pull #840 from NadineSchneider)
  - Autodetect boost c++ library and compile with matching one
 (github pull #845 from bp-kelley)

## New Database Cartridge Features:
  - support providing InChI (or InChI key) generation options in cartridge
 (github pull #755 from greglandrum)
  - building the cartridge is now integrated with the cmake build system
 (github pull #785 from ptosco)

## New Java Wrapper Features:
  - Add a bit more control over the lazy MaxMin picker to the java layer
 (github pull #791 from greglandrum)
  - Ensure reduceProductToSideChains exposed in Java/Swig
 (github issue #744 from bp-kelley)

## Deprecated code (to be removed in next release):

## Removed code:

## Contrib updates:

## Other:

Re: [Rdkit-devel] Pharm2D -- 4-point pharmacophores

[Ondrej M. <mit...@gm...>]

Hi Greg,

> There's an illustration of the representation here:
> http://rdkit.org/docs/RDKit_Book.html#representation-of-pharmacophore-fingerprints
> Some discussion in the code of the terminology used:
> https://github.com/rdkit/rdkit/blob/master/rdkit/Chem/Pharm2D/Generate.py#L11
> and an illustration of the way the distances are defined:
> https://github.com/rdkit/rdkit/blob/master/Docs/Code/Chem/Pharm2D.triangles.jpg

thank you, but I found these myself, I was looking for something more 
thorough, since the code isn't quite self-explanatory.

> It's been a while since I really looked at that code, but I don'd think
> there's really any reason why support for more than 3 features couldn't
> be enabled.

So far, I've found this line
https://github.com/rdkit/rdkit/blob/master/rdkit/Chem/Pharm2D/SigFactory.py#L231
which might be the problem when using n >= 4, but I have absolutely no 
idea, why is it there. Any suggestions?
Ondrej

Re: [Rdkit-devel] Pharm2D -- 4-point pharmacophores

[Greg L. <gre...@gm...>]

Hi Ondrej,

There's an illustration of the representation here:
http://rdkit.org/docs/RDKit_Book.html#representation-of-pharmacophore-fingerprints
Some discussion in the code of the terminology used:
https://github.com/rdkit/rdkit/blob/master/rdkit/Chem/Pharm2D/Generate.py#L11
and an illustration of the way the distances are defined:
https://github.com/rdkit/rdkit/blob/master/Docs/Code/Chem/Pharm2D.triangles.jpg

It's been a while since I really looked at that code, but I don'd think
there's really any reason why support for more than 3 features couldn't be
enabled.

-greg


On Tue, Feb 9, 2016 at 10:02 PM, Ondrej Micka <mit...@gm...> wrote:

> Hello,
> I'd like to implement four-point (or generally N-point) pharmacophores
> into rdkit's  Pharm2D module. It seems that same parts of the module can
> support more-than-three-point pharmacophores without any modification,
> while some other can't. Is there any more thorough documentation than
> the one generated by epydoc which could help me uderstand how the
> generation of 2D pharmacophore fingerprints works?
> Thanks,
> Ondrej
>
>
> ------------------------------------------------------------------------------
> Site24x7 APM Insight: Get Deep Visibility into Application Performance
> APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
> Monitor end-to-end web transactions and take corrective actions now
> Troubleshoot faster and improve end-user experience. Signup Now!
> http://pubads.g.doubleclick.net/gampad/clk?id=272487151&iu=/4140
> _______________________________________________
> Rdkit-devel mailing list
> Rdk...@li...
> https://lists.sourceforge.net/lists/listinfo/rdkit-devel
>

[Rdkit-devel] Pharm2D -- 4-point pharmacophores

[Ondrej M. <mit...@gm...>]

Hello,
I'd like to implement four-point (or generally N-point) pharmacophores 
into rdkit's  Pharm2D module. It seems that same parts of the module can 
support more-than-three-point pharmacophores without any modification, 
while some other can't. Is there any more thorough documentation than 
the one generated by epydoc which could help me uderstand how the 
generation of 2D pharmacophore fingerprints works?
Thanks,
Ondrej

Re: [Rdkit-devel] RDKit binaries available for Win32 .Net?

[Greg L. <gre...@gm...>]

Hi Tom,

There's never really been demand for such a thing. It ought to be doable. I
will investigate and see what I can come up with.

-greg


On Sun, Feb 7, 2016 at 7:27 PM, Tom Hagadone <tom...@gm...> wrote:

> Hello All:
>
> We have a need to be able to call RDKit from an existing Win32 C#.Net app.
> I've been trying to build from source using the existing RDKit SWIG wrapper
> for C#. I haven't had any success yet after two days of trying to sort out
> various SWIG/Boost-related issues. Does anyone know if there is a prebuilt
> Win32.Net RDKit binary (similar to the Win32.Java binary) available
> anywhere? If not, I fear my only option will be to implement a MSVC C++
> Interop solution to bridge between C# and C++.
>
> Thanks,
> Tom Hagadone
> Eli Lilly and Company
>
>
> ------------------------------------------------------------------------------
> Site24x7 APM Insight: Get Deep Visibility into Application Performance
> APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
> Monitor end-to-end web transactions and take corrective actions now
> Troubleshoot faster and improve end-user experience. Signup Now!
> http://pubads.g.doubleclick.net/gampad/clk?id=272487151&iu=/4140
> _______________________________________________
> Rdkit-devel mailing list
> Rdk...@li...
> https://lists.sourceforge.net/lists/listinfo/rdkit-devel
>
>

[Rdkit-devel] RDKit binaries available for Win32 .Net?

[Tom H. <tom...@gm...>]

Hello All:

We have a need to be able to call RDKit from an existing Win32 C#.Net app.
I've been trying to build from source using the existing RDKit SWIG wrapper
for C#. I haven't had any success yet after two days of trying to sort out
various SWIG/Boost-related issues. Does anyone know if there is a prebuilt
Win32.Net RDKit binary (similar to the Win32.Java binary) available
anywhere? If not, I fear my only option will be to implement a MSVC C++
Interop solution to bridge between C# and C++.

Thanks,
Tom Hagadone
Eli Lilly and Company

Re: [Rdkit-devel] failing with rdkit.Chem.PyMol.MolViewer()

[Greg L. <gre...@gm...>]

Hi Zhenquan,

Did you start PyMol with the "-R" flag so that it is running the xmlrpc
server?
You can verify this by looking in the PyMol console for text like this:
"xml-rpc server running on host localhost, port 9123"

-greg



On Wed, Jan 6, 2016 at 2:16 AM, Zhenquan Hu <zhq...@gm...> wrote:

> Dear all,
>
> Recently I tried to install RDKit on ubuntu-14.04/CentOS7.1. First I tried
> with the pre-build version, then compiled with the git source code. Neither
> could call the module Chem.PyMol.MolViewer().
> It looks like something wrong with xmlrpc, but I have no idea about the
> detail.
>
> Anyone knows why?
>
> best,
> Zhenquan
>
> test.py
> #==========================
> from rdkit.Chem import PyMol
> v = PyMol.MolViewer()
> #==========================
>
> python test.py
> #===================================================
> Traceback (most recent call last):
>   File "test.py", line 3, in <module>
>     v=PyMol.MolViewer()
>   File "/opt/rdkit/rdkit/Chem/PyMol.py", line 35, in __init__
>     serv.ping()
>   File "/usr/lib64/python2.7/xmlrpclib.py", line 1233, in __call__
>     return self.__send(self.__name, args)
>   File "/usr/lib64/python2.7/xmlrpclib.py", line 1587, in __request
>     verbose=self.__verbose
>   File "/usr/lib64/python2.7/xmlrpclib.py", line 1273, in request
>     return self.single_request(host, handler, request_body, verbose)
>   File "/usr/lib64/python2.7/xmlrpclib.py", line 1301, in single_request
>     self.send_content(h, request_body)
>   File "/usr/lib64/python2.7/xmlrpclib.py", line 1448, in send_content
>     connection.endheaders(request_body)
>   File "/usr/lib64/python2.7/httplib.py", line 975, in endheaders
>     self._send_output(message_body)
>   File "/usr/lib64/python2.7/httplib.py", line 835, in _send_output
>     self.send(msg)
>   File "/usr/lib64/python2.7/httplib.py", line 797, in send
>     self.connect()
>   File "/usr/lib64/python2.7/httplib.py", line 778, in connect
>     self.timeout, self.source_address)
>   File "/usr/lib64/python2.7/socket.py", line 571, in create_connection
>     raise err
> socket.error: [Errno 111] Connection refused
> #=====================================================
>
>
>
> ------------------------------------------------------------------------------
>
> _______________________________________________
> Rdkit-devel mailing list
> Rdk...@li...
> https://lists.sourceforge.net/lists/listinfo/rdkit-devel
>
>

[Rdkit-devel] failing with rdkit.Chem.PyMol.MolViewer()

[Zhenquan Hu <zhq...@gm...>]

Dear all,

Recently I tried to install RDKit on ubuntu-14.04/CentOS7.1. First I tried
with the pre-build version, then compiled with the git source code. Neither
could call the module Chem.PyMol.MolViewer().
It looks like something wrong with xmlrpc, but I have no idea about the
detail.

Anyone knows why?

best,
Zhenquan

test.py
#==========================
from rdkit.Chem import PyMol
v = PyMol.MolViewer()
#==========================

python test.py
#===================================================
Traceback (most recent call last):
  File "test.py", line 3, in <module>
    v=PyMol.MolViewer()
  File "/opt/rdkit/rdkit/Chem/PyMol.py", line 35, in __init__
    serv.ping()
  File "/usr/lib64/python2.7/xmlrpclib.py", line 1233, in __call__
    return self.__send(self.__name, args)
  File "/usr/lib64/python2.7/xmlrpclib.py", line 1587, in __request
    verbose=self.__verbose
  File "/usr/lib64/python2.7/xmlrpclib.py", line 1273, in request
    return self.single_request(host, handler, request_body, verbose)
  File "/usr/lib64/python2.7/xmlrpclib.py", line 1301, in single_request
    self.send_content(h, request_body)
  File "/usr/lib64/python2.7/xmlrpclib.py", line 1448, in send_content
    connection.endheaders(request_body)
  File "/usr/lib64/python2.7/httplib.py", line 975, in endheaders
    self._send_output(message_body)
  File "/usr/lib64/python2.7/httplib.py", line 835, in _send_output
    self.send(msg)
  File "/usr/lib64/python2.7/httplib.py", line 797, in send
    self.connect()
  File "/usr/lib64/python2.7/httplib.py", line 778, in connect
    self.timeout, self.source_address)
  File "/usr/lib64/python2.7/socket.py", line 571, in create_connection
    raise err
socket.error: [Errno 111] Connection refused
#=====================================================

Re: [Rdkit-devel] last test failing with python3

[Gianluca S. <gi...@gm...>]

On Thu, Dec 3, 2015 at 4:31 PM, Greg Landrum <gre...@gm...> wrote:
> If both pycairo and cairocffi are installed, the code will prefer cairocffi
> with python3.

Yeah, that I also figured by looking at rdkit/Chem/Draw/cairoCanvas.py.
However, adding python3-cairocffi as dependency did not produce the
desired result, so I investigated further and found a bug in the
python3-cairocffi fedora package [1]

Added a workaround and all is good now!

Thanks again

G.

[1] https://bugzilla.redhat.com/show_bug.cgi?id=1288627

-- 
Gianluca Sforna

http://morefedora.blogspot.com
http://plus.google.com/+gianlucasforna - http://twitter.com/giallu

Re: [Rdkit-devel] last test failing with python3

[Greg L. <gre...@gm...>]

Hi Gianluca,

It looks like that function is not available with pycairo for Python3:
http://cairographics.org/documentation/pycairo/3/reference/surfaces.html#cairo.ImageSurface.create_for_data
It is (or should be) available in cairocffi:
https://pythonhosted.org/cairocffi/api.html#cairocffi.ImageSurface.create_for_data

If both pycairo and cairocffi are installed, the code will prefer cairocffi
with python3.

-greg


On Thu, Dec 3, 2015 at 4:13 PM, Gianluca Sforna <gi...@gm...> wrote:

> Hi all,
> my attempt to build and test python3 bindings in the rpm package is
> one tiny test away from success...
>
> Specifically, I see test 101 fails with:
>
> 101: ======================================================================
> 101: ERROR: testCairoImage (__main__.TestCase)
> 101: ----------------------------------------------------------------------
> 101: Traceback (most recent call last):
> 101:   File "UnitTestDraw.py", line 91, in testCairoImage
> 101:     img=Draw.MolToImage(self.mol,size=(300,300))
> 101:   File
> "/builddir/build/BUILD/rdkit-Release_2015_09_2/python3/rdkit/Chem/Draw/__init__.py",
> line 94, in MolToImage
> 101:     img,canvas=_createCanvas(size)
> 101:   File
> "/builddir/build/BUILD/rdkit-Release_2015_09_2/python3/rdkit/Chem/Draw/__init__.py",
> line 50, in _createCanvas
> 101:     canvas = Canvas(img)
> 101:   File
> "/builddir/build/BUILD/rdkit-Release_2015_09_2/python3/rdkit/Chem/Draw/cairoCanvas.py",
> line 86, in __init__
> 101:     image.size[0], image.size[1], stride)
> 101: NotImplementedError: Surface.create_for_data: Not Implemented yet.
>
> Any idea how this could happen? If you need any additional info, just
> let me know
>
>
> --
> Gianluca Sforna
>
> http://morefedora.blogspot.com
> http://plus.google.com/+gianlucasforna - http://twitter.com/giallu
>
>
> ------------------------------------------------------------------------------
> Go from Idea to Many App Stores Faster with Intel(R) XDK
> Give your users amazing mobile app experiences with Intel(R) XDK.
> Use one codebase in this all-in-one HTML5 development environment.
> Design, debug & build mobile apps & 2D/3D high-impact games for multiple
> OSs.
> http://pubads.g.doubleclick.net/gampad/clk?id=254741911&iu=/4140
> _______________________________________________
> Rdkit-devel mailing list
> Rdk...@li...
> https://lists.sourceforge.net/lists/listinfo/rdkit-devel
>

[Rdkit-devel] last test failing with python3

[Gianluca S. <gi...@gm...>]

Hi all,
my attempt to build and test python3 bindings in the rpm package is
one tiny test away from success...

Specifically, I see test 101 fails with:

101: ======================================================================
101: ERROR: testCairoImage (__main__.TestCase)
101: ----------------------------------------------------------------------
101: Traceback (most recent call last):
101:   File "UnitTestDraw.py", line 91, in testCairoImage
101:     img=Draw.MolToImage(self.mol,size=(300,300))
101:   File "/builddir/build/BUILD/rdkit-Release_2015_09_2/python3/rdkit/Chem/Draw/__init__.py",
line 94, in MolToImage
101:     img,canvas=_createCanvas(size)
101:   File "/builddir/build/BUILD/rdkit-Release_2015_09_2/python3/rdkit/Chem/Draw/__init__.py",
line 50, in _createCanvas
101:     canvas = Canvas(img)
101:   File "/builddir/build/BUILD/rdkit-Release_2015_09_2/python3/rdkit/Chem/Draw/cairoCanvas.py",
line 86, in __init__
101:     image.size[0], image.size[1], stride)
101: NotImplementedError: Surface.create_for_data: Not Implemented yet.

Any idea how this could happen? If you need any additional info, just
let me know


-- 
Gianluca Sforna

http://morefedora.blogspot.com
http://plus.google.com/+gianlucasforna - http://twitter.com/giallu