pygccxml-development

Re: [pygccxml-development] Wrapping abstract C++ classes

[Roman Y. <rom...@gm...>]

On 5/5/07, Gustavo Carneiro <gjc...@gm...> wrote:
> On 05/05/07, Roman Yakovenko <rom...@gm...> wrote:
> > On 5/5/07, Gustavo Carneiro <gjc...@gm...> wrote:
> > >   Is it possible to wrap a C++ class that has pure virtual functions?  I
> am
> > > getting a compilation error below; what should I do?
> > >
> > > PS: those pure virtual methods are being exclud()ed by me.
> >
> > If you exclude them, than I guess you don't want Python user to create
> > instances of the class and derive from it. If I am right, than you
> > should use no_init:
> >
> > http://boost.org/libs/python/doc/v2/class.html.
> >
> > The only way to tweak Py++ to generate no_init for the class is to
> > exclude all class constructors.
>
>   Well, assuming the code below is correct, it doesn't seem to make any
> difference:
>
>     for meth in Queue.constructors():
>         meth.exclude()
>
>  But, anyway, I just white-listed the pure virtual methods and now it
> compiles.  They won't work in runtime because they handle unregistered
> parameters, but that is a lesser evil.

If you had some free time, could you submit a complete example that
show a problem?

Thanks

-- 
Roman Yakovenko
C++ Python language binding
http://www.language-binding.net/

Re: [pygccxml-development] Wrapping abstract C++ classes

[Gustavo C. <gjc...@gm...>]

On 05/05/07, Roman Yakovenko <rom...@gm...> wrote:
>
> On 5/5/07, Gustavo Carneiro <gjc...@gm...> wrote:
> >   Is it possible to wrap a C++ class that has pure virtual functions?  I
> am
> > getting a compilation error below; what should I do?
> >
> > PS: those pure virtual methods are being exclud()ed by me.
>
> If you exclude them, than I guess you don't want Python user to create
> instances of the class and derive from it. If I am right, than you
> should use no_init:
>
> http://boost.org/libs/python/doc/v2/class.html.
>
> The only way to tweak Py++ to generate no_init for the class is to
> exclude all class constructors.


  Well, assuming the code below is correct, it doesn't seem to make any
difference:

    for meth in Queue.constructors():
        meth.exclude()

 But, anyway, I just white-listed the pure virtual methods and now it
compiles.  They won't work in runtime because they handle unregistered
parameters, but that is a lesser evil.

  Thanks anyway,

-- 
Gustavo J. A. M. Carneiro
"The universe is always one step beyond logic." -- Frank Herbert

Re: [pygccxml-development] __convenience.pypp.hpp error calling 'len'

[Roman Y. <rom...@gm...>]

On 5/5/07, Allen Bierbaum <al...@vr...> wrote:
> If you can point me at where the file is being introduced and what logic
> I should check against to know not to include it, then I could try
> something out.

check out the latest version, I removed the dependency.

-- 
Roman Yakovenko
C++ Python language binding
http://www.language-binding.net/

Re: [pygccxml-development] Fixed list (Re: Py++ IDE - suggested improvements)

[Roman Y. <rom...@gm...>]

On 5/4/07, Alexander Eisenhuth <st...@st...> wrote:
> > I did not have time and your agreement to change the interface, so I did
> > it this way.
>
> I give you now "my agreement" to change the code whatever you tink it's
> worth/needed to change. I think you have further plans with the IDE. I see my
> contribution in  the projectmanagement and some basic BOA stuff, and hope
> my work is useful for the future of the IDE.

Definitely.

>
> If you find further complaints feel free to post them.

Thank you very much.

-- 
Roman Yakovenko
C++ Python language binding
http://www.language-binding.net/

Re: [pygccxml-development] Wrapping abstract C++ classes

[Roman Y. <rom...@gm...>]

On 5/5/07, Gustavo Carneiro <gjc...@gm...> wrote:
>   Is it possible to wrap a C++ class that has pure virtual functions?  I am
> getting a compilation error below; what should I do?
>
> PS: those pure virtual methods are being exclud()ed by me.

If you exclude them, than I guess you don't want Python user to create
instances of the class and derive from it. If I am right, than you
should use no_init:

http://boost.org/libs/python/doc/v2/class.html.

The only way to tweak Py++ to generate no_init for the class is to
exclude all class constructors.

HTH.

-- 
Roman Yakovenko
C++ Python language binding
http://www.language-binding.net/

[pygccxml-development] Wrapping abstract C++ classes

[Gustavo C. <gjc...@gm...>]

  Is it possible to wrap a C++ class that has pure virtual functions?  I am
getting a compilation error below; what should I do?

PS: those pure virtual methods are being exclud()ed by me.

/usr/include/boost/python/object/value_holder.hpp:66: error: cannot declare
field 'boost::python::objects::value_holder<Queue_wrapper>::m_held' to be of
abstract type 'Queue_wrapper'
debug/bindings/python/ns3.cpp:58: note:   because the following virtual
functions are pure within 'Queue_wrapper':
./debug/ns3/queue.h:91: note:   virtual bool ns3::Queue::DoEnqueue(const
ns3::Packet&)
./debug/ns3/queue.h:92: note:   virtual bool
ns3::Queue::DoDequeue(ns3::Packet&)
./debug/ns3/queue.h:93: note:   virtual bool
ns3::Queue::DoPeek(ns3::Packet&)
Compilation failed

-- 
Gustavo J. A. M. Carneiro
"The universe is always one step beyond logic." -- Frank Herbert

Re: [pygccxml-development] __convenience.pypp.hpp error calling 'len'

[Roman Y. <rom...@gm...>]

On 5/5/07, Allen Bierbaum <al...@vr...> wrote:
> With the latest version of pypp from svn I am getting this error while
> using boost 1.33.1 with fedora core 6.
>
> src/osg_module/generated/__convenience.pypp.hpp: In function 'void
> pyplusplus::convenience::ensure_uniform_sequence(boost::python::api::object,
> pyplusplus::convenience::index_type)':
> src/osg_module/generated/__convenience.pypp.hpp:46: error: 'len' is not
> a member of 'boost::python'
> src/osg_module/generated/__convenience.pypp.hpp: In function 'void
> pyplusplus::convenience::copy_sequence(const
> boost::python::api::object&, Inserter)':
> src/osg_module/generated/__convenience.pypp.hpp:76: error: 'len' is not
> a member of 'boost::python'
> src/osg_module/generated/__convenience.pypp.hpp: In function 'void
> pyplusplus::convenience::copy_sequence(const
> boost::python::api::object&, Inserter, boost::type<Target>)':
> src/osg_module/generated/__convenience.pypp.hpp:84: error: 'len' is not
> a member of 'boost::python'
>
> Is this a case where the helper methods are using capabilities from the
> development version of boost?

Yes, can you create a patch for this?

-- 
Roman Yakovenko
C++ Python language binding
http://www.language-binding.net/

[pygccxml-development] __convenience.pypp.hpp error calling 'len'

[Allen B. <al...@vr...>]

With the latest version of pypp from svn I am getting this error while 
using boost 1.33.1 with fedora core 6.

src/osg_module/generated/__convenience.pypp.hpp: In function 'void 
pyplusplus::convenience::ensure_uniform_sequence(boost::python::api::object, 
pyplusplus::convenience::index_type)':
src/osg_module/generated/__convenience.pypp.hpp:46: error: 'len' is not 
a member of 'boost::python'
src/osg_module/generated/__convenience.pypp.hpp: In function 'void 
pyplusplus::convenience::copy_sequence(const 
boost::python::api::object&, Inserter)':
src/osg_module/generated/__convenience.pypp.hpp:76: error: 'len' is not 
a member of 'boost::python'
src/osg_module/generated/__convenience.pypp.hpp: In function 'void 
pyplusplus::convenience::copy_sequence(const 
boost::python::api::object&, Inserter, boost::type<Target>)':
src/osg_module/generated/__convenience.pypp.hpp:84: error: 'len' is not 
a member of 'boost::python'

Is this a case where the helper methods are using capabilities from the 
development version of boost?

-Allen

[pygccxml-development] Fixed list (Re: Py++ IDE - suggested improvements)

[Alexander E. <st...@st...>]

Roman Yakovenko schrieb:
> On 5/3/07, *Alexander Eisenhuth* <st...@st... 
> <mailto:st...@st...>> wrote:
> 
>     Roman Yakovenko schrieb:
>      > Good evening Alexander. First of all thank you for the huge work
>     you do.
>      >
>      > While working on adding "generate xxx" functionality I decided to
>     write
>      > here every problem I have. This is a list.
>      >
>      > 1. Header file - the dialog should be opened in current working
>     directory
Done

> 
>      > 2. The dialog should not filter any C or C++ files. Right now it
>     filters
>      > out everything, except "*.h"
>     On Windows you have a drop down to change the filter to "*". Isn't
>     that under
>     ubuntu the same? I can change the default to "*" if you like that
>     better. Waht
>     would be a useful filter?
> 
> 
> I can change this too. I was not clear, the default value should be 
> extensions you listed.
> 
>     Suggestion:
>     (*.h,*.hpp, *.hxx, *.c,*.cpp, *.cxx)
> 
Done

>      > 3. The extension of saved project file should not be XML. XML is
>     just a
>      > format. I suggest pypp, but you are free to propose something else.
>     Ok, will be changed
> 
Done

>      > 5. It is not clear how to add\remove "Include Path" or Macros. I
>     think
>      > two buttons "add" and "remove" will provide better interface. May
>     be you
>      > have nice icons? Boa comes with good collection of icons.
>     I used a contex menue to save the extra space needed for buttons. I
>     think users
>     who use Py++ IDE have the knowledge of the right mouse button ;-)

It's not a quick thing to do that in BOA. You need new sizers. At the moment I'm 
short of time, but I think this would be a good practice to get into BOA Designer :)

>      > 6. I think the title of the window should contain name of the
>     project
>      > and the path to it.
>     That's the way how I implemented it. Did you try to save the project?
> 
> 
> I could be wrong, but it contains the full path to the project

I hope I got it. -> Done

> 
>      > 7. Output warning\info should use Python logging package, where
>     handlers
>      > write to specific window.
>     Feel free to add that functionality. Maybe you can write the handler
>     interface
>     and I'll integrate it in the IDE code. OK? 
> 
> 
> Okey. I already did it for some of my projects.

Waiting for your interface to logging

> 
>      > 8. gen_xml, gen_cpp, gen_pypp should use pygccxml defined classes in
>      > their interfaces:
>     Why is this a should? Why must it be created outside of
>     code_generator.py
> 
> 
> Good question, I hope I will provide good answer:
> 
> 1. We are not developing another Boa or Eclipse
I agree

> 2. The code will be much smaller: code reuse
Now we reuse python lists

> 3. The interfaces are clear.
For you, Ok.


>      > 11. Code editor is luck of horizontal scroll bar.
>     Ok, I'll include one.
Done

> 
> I did not have time and your agreement to change the interface, so I did 
> it this way.  

I give you now "my agreement" to change the code whatever you tink it's 
worth/needed to change. I think you have further plans with the IDE. I see my 
contribution in  the projectmanagement and some basic BOA stuff, and hope my 
work is useful for the future of the IDE.

If you find further complaints feel free to post them.

Best greetings
Alexander

Re: [pygccxml-development] Py++ IDE - suggested improvements

[Roman Y. <rom...@gm...>]

On 5/3/07, Alexander Eisenhuth <st...@st...> wrote:
>
> Roman Yakovenko schrieb:
> > Good evening Alexander. First of all thank you for the huge work you do.
> >
> > While working on adding "generate xxx" functionality I decided to write
> > here every problem I have. This is a list.
> >
> > 1. Header file - the dialog should be opened in current working
> directory
> I'll fix that.


Thanks

> 2. The dialog should not filter any C or C++ files. Right now it filters
> > out everything, except "*.h"
> On Windows you have a drop down to change the filter to "*". Isn't that
> under
> ubuntu the same? I can change the default to "*" if you like that better.
> Waht
> would be a useful filter?


I can change this too. I was not clear, the default value should be
extensions you listed.

Suggestion:
> (*.h,*.hpp, *.hxx, *.c,*.cpp, *.cxx)


I agree

> 3. The extension of saved project file should not be XML. XML is just a
> > format. I suggest pypp, but you are free to propose something else.
> Ok, will be changed


Thanks

> 4. Macros - may be we should change to "defines"?
> I'm not shure weather this is the correct expression. Macro is,afaik , the
> expression. DEFINE is the keyword of C language. But if you insist I'll
> change that.


No I don't

> 5. It is not clear how to add\remove "Include Path" or Macros. I think
> > two buttons "add" and "remove" will provide better interface. May be you
> > have nice icons? Boa comes with good collection of icons.
> I used a contex menue to save the extra space needed for buttons. I think
> users
> who use Py++ IDE have the knowledge of the right mouse button ;-)


Yes, but this is not intuitive. Do you read JoelOnSoftware blog? If not
,stop coding and go to read it (http://www.joelonsoftware.com/)

> 6. I think the title of the window should contain name of the project
> > and the path to it.
> That's the way how I implemented it. Did you try to save the project?


I could be wrong, but it contains the full path to the project

> 7. Output warning\info should use Python logging package, where handlers
> > write to specific window.
> Feel free to add that functionality. Maybe you can write the handler
> interface
> and I'll integrate it in the IDE code. OK?


Okey. I already did it for some of my projects.

> 8. gen_xml, gen_cpp, gen_pypp should use pygccxml defined classes in
> > their interfaces:
> Why is this a should? Why must it be created outside of code_generator.py


Good question, I hope I will provide good answer:

1. We are not developing another Boa or Eclipse
2. The code will be much smaller: code reuse
3. The interfaces are clear.

> 9. All previous functions do not get small but very important parameter
> > - name of the file they work on :-) - patched
> Ok
>
> > 10. You don't need "verbose" argument - use logging package instead and
> > play with logger settings.
> Related to logging, see above
>
> > 11. Code editor is luck of horizontal scroll bar.
> Ok, I'll include one.


Thanks


> 12. I attached few screen-shots. I am sure you can make the dialogs to
> > look better :-)
> >
> See below
>
> > Wish list: can you use Scite editor? Does it bring additional
> dependencies?
> I don't know Scite. At the moment we keep it on the wish list?


Yes of course.

>
> > Now I have few question:
> > 1. Why gen_xxx functions have such interface? What is wrong with simple
> >     def gen_xxx( [file_configuration_t], gccxml_configuration ):
> >          ...
> >          return string
> >
>
> A) It runs in a own thread, because otherwise it will block the event
> handler of
> the GUI. (Reason of queues)


I don't agree. You can use the queues out-side of the functions.

B) I didn't use the file_configuration_t, gccxml_configuration because I
> didn't
> know them. I'm no experienced py++ user.


Ok.

>     Error happens, when exception is raised, the caller will "except" it
> > and report to the user.
>
> I'm not shure waht you mean. As I looked on the code, you implement it the
> way
> it was intended.


I did not have time and your agreement to change the interface, so I did it
this way.

>
> > 2. I don't understand why did you define such complex interface for
> > ProjectSettings ? Can you simplify it?
> - To have a generic interface for parameters
> - To use existing code
>
> What do you need / miss ?


Py++ code generator has pretty limited requirements from the
ProjectSettings:
1. compiler (gccxml) configuration
2. source files
3. In future it will ask access for module_builder_t instance, that's all

By the way I think we can define save\load functions on pygccxml classes I
mentioned. what do you think?

>
> > I added "generate xml" and "generate code" functionality. Please update
> > your sources. I also added "_dev_run_ide_.py" script. It allows to run
> > IDE from the source tree.
> >
> > Nevertheless of my complains, the IDE looks promising. Please keep up
> > working on this project.
> I'm not shure weather you're lucky of my approach / SW-Design.


Yes I am , the things  I am complaining about are minors. Your design makes
clear to user how to use the IDE and it also responsive. I mean all hard
work is done in another thread. This is what important to user, hence to me.

>
> > Thanks
>
> On ubuntu there seems a other behaviour of the gui system.  I'm a bit
> surprised
> of that fact.
>
> I'm having access to ubuntu 6.06. Is it possible to install python 2.5 +
> wx
> 2.8.1.1 and don't destroy the 6.06 installation? I think I need a dev
> environment, otherwise I've no chance to reproduce your screenshots.
>
> Some words on Screenshot.png would be helpful.


Don't  bother yourself. I am in the middle of the release process of
pygccxml and Py++. After I release them I will modify the IDE by myself.

-- 
Roman Yakovenko
C++ Python language binding
http://www.language-binding.net/

Re: [pygccxml-development] Py++ IDE - suggested improvements

[Alexander E. <st...@st...>]

Roman Yakovenko schrieb:
> Good evening Alexander. First of all thank you for the huge work you do.
> 
> While working on adding "generate xxx" functionality I decided to write 
> here every problem I have. This is a list.
> 
> 1. Header file - the dialog should be opened in current working directory
I'll fix that.

> 2. The dialog should not filter any C or C++ files. Right now it filters 
> out everything, except "*.h"
On Windows you have a drop down to change the filter to "*". Isn't that under
ubuntu the same? I can change the default to "*" if you like that better. Waht 
would be a useful filter?

Suggestion:
(*.h,*.hpp, *.hxx, *.c,*.cpp, *.cxx)

> 3. The extension of saved project file should not be XML. XML is just a 
> format. I suggest pypp, but you are free to propose something else.
Ok, will be changed

> 4. Macros - may be we should change to "defines"?
I'm not shure weather this is the correct expression. Macro is,afaik , the 
expression. DEFINE is the keyword of C language. But if you insist I'll change that.

> 5. It is not clear how to add\remove "Include Path" or Macros. I think 
> two buttons "add" and "remove" will provide better interface. May be you 
> have nice icons? Boa comes with good collection of icons.
I used a contex menue to save the extra space needed for buttons. I think users
who use Py++ IDE have the knowledge of the right mouse button ;-)

> 6. I think the title of the window should contain name of the project 
> and the path to it.
That's the way how I implemented it. Did you try to save the project?

> 7. Output warning\info should use Python logging package, where handlers 
> write to specific window.
Feel free to add that functionality. Maybe you can write the handler interface
and I'll integrate it in the IDE code. OK?

> 8. gen_xml, gen_cpp, gen_pypp should use pygccxml defined classes in 
> their interfaces:
Why is this a should? Why must it be created outside of code_generator.py

> 9. All previous functions do not get small but very important parameter 
> - name of the file they work on :-) - patched
Ok

> 10. You don't need "verbose" argument - use logging package instead and 
> play with logger settings.
Related to logging, see above

> 11. Code editor is luck of horizontal scroll bar.
Ok, I'll include one.

> 12. I attached few screen-shots. I am sure you can make the dialogs to 
> look better :-)
> 
See below

> Wish list: can you use Scite editor? Does it bring additional dependencies?
I don't know Scite. At the moment we keep it on the wish list?

> 
> Now I have few question:
> 1. Why gen_xxx functions have such interface? What is wrong with simple
>     def gen_xxx( [file_configuration_t], gccxml_configuration ):
>          ...
>          return string
> 

A) It runs in a own thread, because otherwise it will block the event handler of
the GUI. (Reason of queues)

B) I didn't use the file_configuration_t, gccxml_configuration because I didn't
know them. I'm no experienced py++ user.

>     Error happens, when exception is raised, the caller will "except" it 
> and report to the user.

I'm not shure waht you mean. As I looked on the code, you implement it the way 
it was intended.

> 
> 2. I don't understand why did you define such complex interface for 
> ProjectSettings ? Can you simplify it?
- To have a generic interface for parameters
- To use existing code

What do you need / miss ?

> 
> I added "generate xml" and "generate code" functionality. Please update 
> your sources. I also added "_dev_run_ide_.py" script. It allows to run 
> IDE from the source tree.
> 
> Nevertheless of my complains, the IDE looks promising. Please keep up 
> working on this project.
I'm not shure weather you're lucky of my approach / SW-Design.

> 
> Thanks

On ubuntu there seems a other behaviour of the gui system.  I'm a bit surprised 
of that fact.

I'm having access to ubuntu 6.06. Is it possible to install python 2.5 + wx 
2.8.1.1 and don't destroy the 6.06 installation? I think I need a dev 
environment, otherwise I've no chance to reproduce your screenshots.

Some words on Screenshot.png would be helpful.

Re: [pygccxml-development] Next Commit of Py++ IDE

[Roman Y. <rom...@gm...>]

On 5/2/07, Alexander Eisenhuth <new...@st...> wrote:
> Roman,
>
> today i comitted a new version of the ide. I added everything (also a little
> documentation), except the code generation. Can you add it ?
>
> The interface to the code generation is located in:
>
>         <IDE_DIR>/model/code_generator.py
>
> There you can find three functions:
> - gen_xml
> - gen_cpp
> - gen_pypp
>
> Can you please test the gui under your ubuntu and send me a report?

Yes, I will try to do this today.

Thank you very much.


-- 
Roman Yakovenko
C++ Python language binding
http://www.language-binding.net/

[pygccxml-development] Next Commit of Py++ IDE

[Alexander E. <new...@st...>]

Roman,

today i comitted a new version of the ide. I added everything (also a little 
documentation), except the code generation. Can you add it ?

The interface to the code generation is located in:

	<IDE_DIR>/model/code_generator.py

There you can find three functions:
- gen_xml
- gen_cpp
- gen_pypp

Can you please test the gui under your ubuntu and send me a report?

Greetings
Alexander

[pygccxml-development] Py++ - code freeze

[Roman Y. <rom...@gm...>]

Hi all. I am going to freeze source code and to prepare it for another
release. I believe it will take some time. The code itself in a good shape -
it passes almost all unit tests and few external big projects. There were a
lot of changes and I will try to document all of them.

Recently I worked on performance. I believe Py++ is now 30-50% faster.

The first improvement comes from pygccxml - standard XML SAX parser was
replaced with cElementTree iterparse functionality. The old parser is still
available and will be used if cElementTree is not available. Here you can
find some information about cElementTree benchmarks:
http://effbot.org/zone/celementtree.htm#benchmarks<http://effbot.org/zone/celementtree.htm#downloads>

The second one comes from caching some intermediate result during code
generation + creating a repository of md5 sum of generated file. Thus next
time you generate code, Py++ can use the repository and compare md5 sums,
instead of file content. You have to explicitly to enable this
functionality:

    mb = module_builder_t( ... )
    ...
    mb.split_module( ..., use_files_sum_repository=True )


I ask you to try SVN version of Py++ and report any problem with it.

Thanks

-- 
Roman Yakovenko
C++ Python language binding
http://www.language-binding.net/

[pygccxml-development] Performance improvements

[Roman Y. <rom...@gm...>]

Hi. I just committed few changes to pygccxml, which gave really good
performance optimization. In my case, I get more than 20% speed up.

In order to enjoy the improvement you have to use Python 2.5 or to install
cElementTree package.

Please try it and report your result.

P.S. The funny thing is that I changed very small amount of code.

-- 
Roman Yakovenko
C++ Python language binding
http://www.language-binding.net/

[pygccxml-development] better "include" directive control

[Roman Y. <rom...@gm...>]

Hi. I made some changes to Py++, which allows better control and code
generation of code related to include directives:

1. System ( Py++, boost.python ) header files is only included if
needed/used. From now, Py++ defined header files will be only included from
files that need them.

2. I added new feature which allows user to specify additional include files
for file in which relevant declaration generated code will be
placed in:

  mb = module_builder_t( ... )
  my_class = mb.class_( ... )
  my_class.include_files.append( "vector" )

File "vector" will be included only from file, where "my_class" will be
generated.

Both this feature can and will drop compilation the compilation time. They
also allow to tweak generated code in a such way that only really "changed"
code will compile. I mean if some class in exposed project was changed, than
only single file ( the class wrapper ) will be compiled.

I also made few really simple optimization while working on this feature. On
my project I've got something like 8-9% of speed improvement. ( The time
dropped from 6 minutes to 5.20 ).

This is a last feature for this release. I begin to work on documentation.
Please try latest SVN sources and report any problem you has.

Thanks.

-- 
Roman Yakovenko
C++ Python language binding
http://www.language-binding.net/

[pygccxml-development] unit tests

[Roman Y. <rom...@gm...>]

Hi. I would like to use py.test package as unit tests framework. Does
anyone has experience with it?

Basically what I need is its "boxed" mode - to run every unit test in
its own process. It should solve collisions within Boost.Python
registry.

Thanks
-- 
Roman Yakovenko
C++ Python language binding
http://www.language-binding.net/

Re: [pygccxml-development] Do function transformers require a customized version of boost?

[Roman Y. <rom...@gm...>]

On 4/1/07, Allen Bierbaum <al...@vr...> wrote:
>
>
> >
> http://boost.cvs.sourceforge.net/boost/boost/boost/python/object.hpp?view=markup
> >      >
> >      > As you can see this functions will be released as a part of 1.34.
> >     True, but 1.34 isn't released yet so my users don't have it.  Is
> there
> >     anyway to keep features like this (FT.output) working for 1.33 until
> >     1.34 is out and in more wide spread use?
> >
> >
> > Yes I will fix this too
>
> Thank you.
>
> After thinking about this more, I was wondering what you may think of
> adding support for a py++ generation script to specify a "target
> software set".  For example the user could tell the module builder that
> they are targetting boost 1.33.1 without the indexing suite 2 or
> conversely that they are targeting boost HEAD and indexing suite 2.
> Then the code could use this for two things:
>
>     1) Telling the user if they try to use a feature that is not
> supported by their target
>     2) Letting the system make decisions in code generation based on
> available features.
>
> What does everyone think, would this be a useful feature?


I think this is a cool feature and it will take me 0( zero) minutes to
implement it :-)

http://language-binding.net/pyplusplus/documentation/apidocs/pyplusplus.code_creators.target_configuration.target_configuration_t-class.html
http://language-binding.net/pyplusplus/documentation/apidocs/pyplusplus.module_builder.builder.module_builder_t-class.html#build_code_creator

The problem with this feature is that I cannot test it :-(


-- 
Roman Yakovenko
C++ Python language binding
http://www.language-binding.net/

Re: [pygccxml-development] Do function transformers require a customized version of boost?

[Allen B. <al...@vr...>]

>     http://boost.cvs.sourceforge.net/boost/boost/boost/python/object.hpp?view=markup
>      >
>      > As you can see this functions will be released as a part of 1.34.
>     True, but 1.34 isn't released yet so my users don't have it.  Is there
>     anyway to keep features like this (FT.output) working for 1.33 until
>     1.34 is out and in more wide spread use?
> 
> 
> Yes I will fix this too

Thank you.

After thinking about this more, I was wondering what you may think of 
adding support for a py++ generation script to specify a "target 
software set".  For example the user could tell the module builder that 
they are targetting boost 1.33.1 without the indexing suite 2 or 
conversely that they are targeting boost HEAD and indexing suite 2. 
Then the code could use this for two things:

    1) Telling the user if they try to use a feature that is not 
supported by their target
    2) Letting the system make decisions in code generation based on 
available features.

What does everyone think, would this be a useful feature?

-Allen

Re: [pygccxml-development] Do function transformers require a customized version of boost?

[Roman Y. <rom...@gm...>]

On 4/1/07, Allen Bierbaum <al...@vr...> wrote:
>
> Roman Yakovenko wrote:
> >
> >
> > On 4/1/07, *Allen Bierbaum* <al...@vr...
> > <mailto:al...@vr...>> wrote:
> >
> >     [...]
> >     >> Yes. I will try to address the issue in a few days.
> >     >
> >     > Allen I separated functionality dependent on BP Index V2 to a
> >     > different file.
> >     > Can you test this?
> >     It didn't work.  I still get compile errors:
> >
> >     src/python/ben_module/generated/__convenience.pypp.hpp: In function
> >     'void
> >
> pyplusplus::convenience::ensure_uniform_sequence(boost::python::api::object,
> >     pyplusplus::convenience::index_type)':
> >     src/python/ben_module/generated/__convenience.pypp.hpp:46: error:
> >     'len'
> >     is not a member of 'boost::python'
> >     src/python/ben_module/generated/__convenience.pypp.hpp: In function
> >     'void pyplusplus::convenience::copy_sequence(const
> >     boost::python::api::object&, Inserter)':
> >     src/python/ben_module/generated/__convenience.pypp.hpp:76: error:
> >     'len'
> >     is not a member of 'boost::python'
> >     src/python/ben_module/generated/__convenience.pypp.hpp: In function
> >     'void pyplusplus::convenience::copy_sequence(const
> >     boost::python::api::object&, Inserter, boost::type<Target>)':
> >     src/python/ben_module/generated/__convenience.pypp.hpp:84: error:
> >     'len'
> >     is not a member of 'boost::python'
> >
> >
> >
> http://boost.cvs.sourceforge.net/boost/boost/boost/python/object.hpp?view=markup
> >
> > As you can see this functions will be released as a part of 1.34.
> True, but 1.34 isn't released yet so my users don't have it.  Is there
> anyway to keep features like this (FT.output) working for 1.33 until
> 1.34 is out and in more wide spread use?
>

Yes I will fix this too

-- 
Roman Yakovenko
C++ Python language binding
http://www.language-binding.net/

Re: [pygccxml-development] Do function transformers require a customized version of boost?

[Allen B. <al...@vr...>]

Roman Yakovenko wrote:
>
>
> On 4/1/07, *Allen Bierbaum* <al...@vr... 
> <mailto:al...@vr...>> wrote:
>
>     [...]
>     >> Yes. I will try to address the issue in a few days.
>     >
>     > Allen I separated functionality dependent on BP Index V2 to a
>     > different file.
>     > Can you test this?
>     It didn't work.  I still get compile errors:
>
>     src/python/ben_module/generated/__convenience.pypp.hpp: In function
>     'void
>     pyplusplus::convenience::ensure_uniform_sequence(boost::python::api::object,
>     pyplusplus::convenience::index_type)':
>     src/python/ben_module/generated/__convenience.pypp.hpp:46: error:
>     'len'
>     is not a member of 'boost::python'
>     src/python/ben_module/generated/__convenience.pypp.hpp: In function
>     'void pyplusplus::convenience::copy_sequence(const
>     boost::python::api::object&, Inserter)':
>     src/python/ben_module/generated/__convenience.pypp.hpp:76: error:
>     'len'
>     is not a member of 'boost::python'
>     src/python/ben_module/generated/__convenience.pypp.hpp: In function
>     'void pyplusplus::convenience::copy_sequence(const
>     boost::python::api::object&, Inserter, boost::type<Target>)':
>     src/python/ben_module/generated/__convenience.pypp.hpp:84: error:
>     'len'
>     is not a member of 'boost::python'
>
>
> http://boost.cvs.sourceforge.net/boost/boost/boost/python/object.hpp?view=markup
>
> As you can see this functions will be released as a part of 1.34.
True, but 1.34 isn't released yet so my users don't have it.  Is there 
anyway to keep features like this (FT.output) working for 1.33 until 
1.34 is out and in more wide spread use?

-Allen

>
> -- 
> Roman Yakovenko
> C++ Python language binding
> http://www.language-binding.net/
> ------------------------------------------------------------------------
>
> -------------------------------------------------------------------------
> Take Surveys. Earn Cash. Influence the Future of IT
> Join SourceForge.net's Techsay panel and you'll get the chance to share your
> opinions on IT & business topics through brief surveys-and earn cash
> http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
> ------------------------------------------------------------------------
>
> _______________________________________________
> pygccxml-development mailing list
> pyg...@li...
> https://lists.sourceforge.net/lists/listinfo/pygccxml-development
>   

Re: [pygccxml-development] Do function transformers require a customized version of boost?

[Roman Y. <rom...@gm...>]

On 4/1/07, Allen Bierbaum <al...@vr...> wrote:
>
> [...]
> >> Yes. I will try to address the issue in a few days.
> >
> > Allen I separated functionality dependent on BP Index V2 to a
> > different file.
> > Can you test this?
> It didn't work.  I still get compile errors:
>
> src/python/ben_module/generated/__convenience.pypp.hpp: In function
> 'void
>
> pyplusplus::convenience::ensure_uniform_sequence(boost::python::api::object,
> pyplusplus::convenience::index_type)':
> src/python/ben_module/generated/__convenience.pypp.hpp:46: error: 'len'
> is not a member of 'boost::python'
> src/python/ben_module/generated/__convenience.pypp.hpp: In function
> 'void pyplusplus::convenience::copy_sequence(const
> boost::python::api::object&, Inserter)':
> src/python/ben_module/generated/__convenience.pypp.hpp:76: error: 'len'
> is not a member of 'boost::python'
> src/python/ben_module/generated/__convenience.pypp.hpp: In function
> 'void pyplusplus::convenience::copy_sequence(const
> boost::python::api::object&, Inserter, boost::type<Target>)':
> src/python/ben_module/generated/__convenience.pypp.hpp:84: error: 'len'
> is not a member of 'boost::python'


http://boost.cvs.sourceforge.net/boost/boost/boost/python/object.hpp?view=markup

As you can see this functions will be released as a part of 1.34.

-- 
Roman Yakovenko
C++ Python language binding
http://www.language-binding.net/

Re: [pygccxml-development] Do function transformers require a customized version of boost?

[Allen B. <al...@vr...>]

[...]
>> Yes. I will try to address the issue in a few days.
>
> Allen I separated functionality dependent on BP Index V2 to a 
> different file.
> Can you test this?
It didn't work.  I still get compile errors:

src/python/ben_module/generated/__convenience.pypp.hpp: In function 
'void 
pyplusplus::convenience::ensure_uniform_sequence(boost::python::api::object, 
pyplusplus::convenience::index_type)':
src/python/ben_module/generated/__convenience.pypp.hpp:46: error: 'len' 
is not a member of 'boost::python'
src/python/ben_module/generated/__convenience.pypp.hpp: In function 
'void pyplusplus::convenience::copy_sequence(const 
boost::python::api::object&, Inserter)':
src/python/ben_module/generated/__convenience.pypp.hpp:76: error: 'len' 
is not a member of 'boost::python'
src/python/ben_module/generated/__convenience.pypp.hpp: In function 
'void pyplusplus::convenience::copy_sequence(const 
boost::python::api::object&, Inserter, boost::type<Target>)':
src/python/ben_module/generated/__convenience.pypp.hpp:84: error: 'len' 
is not a member of 'boost::python'

Any ideas?

-Allen

Re: [pygccxml-development] Do function transformers require a customized version of boost?

[Roman Y. <rom...@gm...>]

On 3/30/07, Roman Yakovenko <rom...@gm...> wrote:
> On 3/30/07, Allen Bierbaum <al...@vr...> wrote:
> > Last night I added function transformers to my py++ code generation
> > script.  But unfortunately it now looks like I can't use this feature
> > with boost 1.33.1.  It gives this error on compiling:
> >
> > In file included from src/python/ben_module/generated/_ben.main.cpp:7:
> > src/python/ben_module/generated/__call_policies.pypp.hpp:14:58: error:
> > boost/python/suite/indexing/iterator_range.hpp: No such file or directory
> > src/python/ben_module/generated/__convenience.pypp.hpp: In function
> > 'void
> > pyplusplus::convenience::ensure_uniform_sequence(boost::python::api::object,
> > pyplusplus::convenience::index_type)':
> > src/python/ben_module/generated/__convenience.pypp.hpp:46: error: 'len'
> > is not a member of 'boost::python'
> >
> > I am assuming this means that it requires the a customized version of
> > boost with the new indexing suite that is not part of the standard
> > boost.python.  Is this correct?  Is there any way around this?
> >
> > It would be very painful for me to tell all my users that they can't use
> > the standard version of boost that comes with their OS or that they can
> > get from boost.org.  Unless the indexing suite is accepted into
> > boost.python I don't really think it will ever become an option for me
> > to use it or any py++ code that requires it. :(
> >
> > In the case of the code I am generating, I am only using the FT.output
> > transformation anyway, so it doesn't even need the indexing suite.  Is
> > there any way that py++ could be a bit smarter and not include the
> > indexing suite files unless they are needed?
>
> Yes. I will try to address the issue in a few days.

Allen I separated functionality dependent on BP Index V2 to a different file.
Can you test this?

Thanks

-- 
Roman Yakovenko
C++ Python language binding
http://www.language-binding.net/

Re: [pygccxml-development] First commit of ide

[Roman Y. <rom...@gm...>]

On 3/30/07, Alexander Eisenhuth <new...@st...> wrote:
> Roman,
>
> I commited a first version of the ide. There was no funcionality added.
>
> I checked the gui with ubuntu 6.06. There is by default python 2.4 an wx 2.6 and
> it has a strange look. What versions are you using?
>
> Can we define as requirement python 2.5 and wx 2.8 ?

You can define whatever requirements you want - just specify them
explicitly somewhere.

The reasons is simple: in addition to source code we will distribute
binary(frozen) version. So users will not have to install any package
from the dependencies.

-- 
Roman Yakovenko
C++ Python language binding
http://www.language-binding.net/