pygccxml-development

Re: [pygccxml-development] What am I doing wrong?

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

On 4/18/06, Neal Becker <ndb...@gm...> wrote:
>
> I replaced scanner.py in pygccxml source, ran python setup.py build + ins=
tall,
> but the same problem is still there.  I confirmed that scanner.py is upda=
ted.
> parsing files - started

I run all tests I have.
Can you run unittests? You have unittests dir. Within this directory
there is file "autoconfig.py"
Please adjust the settings and run next command:

python test_all.py

Only 2 tests should fail.

Also can you debug the problem?

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

Re: [pygccxml-development] What am I doing wrong?

[Neal B. <ndb...@gm...>]

On Tuesday 18 April 2006 4:10 am, Roman Yakovenko wrote:
> Neal, I found the problem.
> Description: GCC-XML changed serialization of unnamed declarations (
> enums + namespaces )
> In previous version of GCC-XML, an element, that describes the
> declaration has attribute "name" with some content. Now it does not
> have this attribute.
>
> By the way, it was nothing wrong with SAX, but with my code. pygccxml
> uses SAX to
> parse XML files. It gaves huge speed improvement.
>
> The fix is simple:
> file pygccxml/parser/scanner.py
>
> search and replace next string
> attrs[ XML_AN_NAME ]
> with
> attrs.get( XML_AN_NAME, '' )
>
> Or you can replace the file with one I attach with this mail.
>

Thanks!

I replaced scanner.py in pygccxml source, ran python setup.py build + install, 
but the same problem is still there.  I confirmed that scanner.py is updated.
parsing files - started
Reading project files: file by file
Reading source file: [hello_world.hpp].
File has not been found in cache, parsing...
 Command line for GCC-XML: /usr/bin/gccxml  -I"." -I"/usr/local/src/boost.cvs"   
"hello_world.hpp" -fxml="/tmp/tmpDRYXEY.xml"
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
  File "/usr/tmp/python-4tAz49.py", line 17, in ?
    mb = module_builder.module_builder_t(
  File 
"/usr/lib/python2.4/site-packages/pyplusplus/module_builder/builder.py", line 
76, in __init__
    , cache )
  File 
"/usr/lib/python2.4/site-packages/pyplusplus/module_builder/builder.py", line 
101, in __parse_declarations
    decls = reader.read_files( files, compilation_mode )
  File "/usr/lib/python2.4/site-packages/pygccxml/parser/project_reader.py", 
line 149, in read_files
    return self.__parse_file_by_file(files)
  File "/usr/lib/python2.4/site-packages/pygccxml/parser/project_reader.py", 
line 172, in __parse_file_by_file
    decls = reader.read_file( header )
  File "/usr/lib/python2.4/site-packages/pygccxml/parser/source_reader.py", 
line 164, in read_file
    raise error
KeyError: 'name'

Re: [pygccxml-development] What am I doing wrong?

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

Neal, I found the problem.
Description: GCC-XML changed serialization of unnamed declarations (
enums + namespaces )
In previous version of GCC-XML, an element, that describes the
declaration has attribute "name" with some content. Now it does not
have this attribute.

By the way, it was nothing wrong with SAX, but with my code. pygccxml
uses SAX to
parse XML files. It gaves huge speed improvement.

The fix is simple:
file pygccxml/parser/scanner.py

search and replace next string
attrs[ XML_AN_NAME ]
with
attrs.get( XML_AN_NAME, '' )

Or you can replace the file with one I attach with this mail.

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

Re: [pygccxml-development] What am I doing wrong?

[Neal B. <ndb...@gm...>]

On Monday 17 April 2006 5:41 pm, Roman Yakovenko wrote:
> On 4/18/06, Neal Becker <ndb...@gm...> wrote:
> > OK, back to my previous test.  gccxml cvs is now updated so the problem I
> > reported last week is fixed (thanks!)
> >
> > But still no luck.  Any hints?  (I have to say pyplusplus error messages
> > could probably be more helpful)
>
> Well, you are right messages could be much better.  In most cases they are.
> In your case, I did not expect this code to fail. I guess this is
> because of your
> version of GCCXML. I will look at it.
>
> Thanks for bug reporting.
>
> --

It is a problem in 
(Pdb) n
> /usr/lib64/python2.4/site-packages/_xmlplus/sax/__init__.py(31)parse()
-> parser.parse(filename_or_stream)
(Pdb) n
KeyError: <exceptions.KeyError instance at 0x2aaab05c7dd0>

Something sax doesn't like.  Not sure how to debug it, one thing does look 
strange in the xml file:
  <File id="f55" name="/usr/include/c++/4.1.0/cstring"/>
  <File id="f56" name="&lt;internal&gt;"/>
  <File id="f57" name="/usr/include/c++/4.1.0/ext/new_allocator.h"/>

Anyway I put the xml here:
http://nbecker.dyndns.org:8080/tmpTHoLWR.xml.gz

Re: [pygccxml-development] What am I doing wrong?

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

On 4/18/06, Neal Becker <ndb...@gm...> wrote:
> OK, back to my previous test.  gccxml cvs is now updated so the problem I
> reported last week is fixed (thanks!)
>
> But still no luck.  Any hints?  (I have to say pyplusplus error messages =
could
> probably be more helpful)

Well, you are right messages could be much better.  In most cases they are.
In your case, I did not expect this code to fail. I guess this is
because of your
version of GCCXML. I will look at it.

Thanks for bug reporting.

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

[pygccxml-development] What am I doing wrong?

[Neal B. <ndb...@gm...>]

OK, back to my previous test.  gccxml cvs is now updated so the problem I 
reported last week is fixed (thanks!)

But still no luck.  Any hints?  (I have to say pyplusplus error messages could 
probably be more helpful)

gccxml now runs without errors.

But I get this:
parsing files - started
Reading project files: file by file
Reading source file: [hello_world.hpp].
File has not been found in cache, parsing...
 Command line for GCC-XML: /usr/bin/gccxml  -I"." -I"/usr/local/src/boost.cvs"   
"hello_world.hpp" -fxml="/tmp/tmp_TUhxm.xml"
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
  File "/usr/tmp/python-W1AaN3.py", line 17, in ?
    mb = module_builder.module_builder_t(
  File 
"/usr/lib/python2.4/site-packages/pyplusplus/module_builder/builder.py", line 
76, in __init__
    , cache )
  File 
"/usr/lib/python2.4/site-packages/pyplusplus/module_builder/builder.py", line 
101, in __parse_declarations
    decls = reader.read_files( files, compilation_mode )
  File "/usr/lib/python2.4/site-packages/pygccxml/parser/project_reader.py", 
line 149, in read_files
    return self.__parse_file_by_file(files)
  File "/usr/lib/python2.4/site-packages/pygccxml/parser/project_reader.py", 
line 172, in __parse_file_by_file
    decls = reader.read_file( header )
  File "/usr/lib/python2.4/site-packages/pygccxml/parser/source_reader.py", 
line 164, in read_file
    raise error
KeyError: 'name'

This is from 'generate_code.py', which is this:
import os
#from environment import settings
from pyplusplus import module_builder

mb = module_builder.module_builder_t(
        
files=['hello_world.hpp'],include_paths=['/usr/local/src/boost.cvs'],compilation_verbose=True
)

Here is hello_world.hpp:

[pygccxml-development] Phisical (re)design

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

Hi. As decided, development of pyplusplus is going to switch to SVN.

Here is a proposed phisical design for the packages. Please comment it.
The main goal of redesign is to use Python distutils package.

pygccxml_dev/
|    |----setup.py
|    |----docs/
|    |    |----apidocs/
|    |----pygccxml/
|    |    |----declarations/
|    |    |----parser/
|    |    |----utils/
|    |----unittests/
|    |    |----data/
pyplusplus_dev/
|    |----environment.py #containes paths to boost  and pygccxml
|    |----setup.py
|    |----docs/
|    |    |----apidocs/
|    |    |----examples/
|    |    |----tutorials/
|    |----pyplusplus/
|    |    |----code_creators/
|    |    |----code_repository/
|    |    |----decl_wrappers/
|    |    |----file_writers/
|    |    |----gui/
|    |    |----_logging_/
|    |    |----module_builder/
|    |    |----module_creator/
|    |    |----scripts/
|    |    |----utils/
|    |----unittests/
|    |     |----data/
|    |     |----generated/
|    |----examples/
|    |    |----pyboost_dev/
|    |    |    |----setup.py
|    |    |    |----pyboost/
|    |    |    |    |----date_time/
|    |    |    |    |    |----generated/
|    |    |    |    |----....
|    |    |----pyeasybmp_dev/
|    |    |    |----setup.py
|    |    |    |----environment.py
|    |    |    |----pyeasybmp/
|    |    |    |    |----generated/
|    |    |----custom_code_creator/
|    |    |----unittests/
|    |----tutorials/
|    |    |----module_builder/
|    |    |----...

What do you think?

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

[pygccxml-development] New release

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

Good evening.

First of all it is really good, because I just uploaded to SourceForge
the final version
of pygccxml-0.7.2 and pyplusplus-0.7.1.

Second, I want to say "thank you" to Matthias Baas. He contributed so much =
to
both projects! Here is some short list of what he did:
1. He created high level API design!
2. He wrote a lot of documentation strings!
3. He improved pygccxml cache performance!
4. He added logging functionality in some places!
5. ... ( I am sure he did something else, I just don't remember )

Third, I want to say "thank you" to Allen Bierbaum, too. He also contribute=
d
for both projects: documentation, performance turning and ideas.

Changes:
1. Code released IS NOT different from last release
    ( pygccxml - 0.7.1 and pyplusplus - 0.7.0 )
    This release adds documentation, tutorials.

2. What has been changed from pyplusplus 0.6.0
    http://www.language-binding.net/pygccxml/history/history.html
    http://www.language-binding.net/pyplusplus/history/history.html
    The short version: pyplusplus is now much user friendly! It has
documentation
    and tutorials:
    http://www.language-binding.net/pyplusplus/tutorials/tutorials.html

Future:
    1. TODO list is open
    2. Wishes, comments,  spell fixes, ideas, FAQs, examples and ...
are welcome.
    3. This is last release we use CVS. I hope, that before next
release we will switch
        to SVN.

Enjoy.

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

Re: [pygccxml-development] first try - more info

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

On 4/14/06, Neal Becker <ndb...@gm...> wrote:
> I removed gccxml got a fresh copy from cvs, rebuilt, and re-installed.  S=
ame
> problem:
>  ...

I saw your posting on GCCXML development list. I think you are right,
the problem
is with GCCXML.

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

Re: [pygccxml-development] first try - more info

[Neal B. <ndb...@gm...>]

I removed gccxml got a fresh copy from cvs, rebuilt, and re-installed.  Same 
problem:

gccxml hello_world.hpp -fxml=/dev/null --debug -I /usr/local/src/boost.cvs
Using "/usr/bin/gccxml_cc1plus" as GCC-XML executable.
Using the following arguments to GCC-XML executable:
  "-quiet"
  "-fsyntax-only"
  "-w"
  "-o"
  "/dev/null"
  "-nostdinc"
  "-D__GCCXML__=700"
  "-D__GCCXML_GNUC__=3"
  "-D__GCCXML_GNUC_MINOR__=3"
  "-D__GCCXML_GNUC_PATCHLEVEL__=2"
  "hello_world.hpp"
  "-fxml=/dev/null"
  "-I"
  "/usr/local/src/boost.cvs"
  "-D__DBL_MIN_EXP__=(-1021)"
  "-D__FLT_MIN__=1.17549435e-38F"
  "-D__CHAR_BIT__=8"
  "-D__WCHAR_MAX__=2147483647"
  "-D__DBL_DENORM_MIN__=4.9406564584124654e-324"
  "-D__FLT_EVAL_METHOD__=0"
  "-D__DBL_MIN_10_EXP__=(-307)"
  "-D__FINITE_MATH_ONLY__=0"
  "-D__LP64__=1"
  "-D__SHRT_MAX__=32767"
  "-D__LDBL_MAX__=1.18973149535723176502e+4932L"
  "-D__UINTMAX_TYPE__=long unsigned int"
  "-D__linux=1"
  "-D__unix=1"
  "-D__linux__=1"
  "-D__SCHAR_MAX__=127"
  "-D__USER_LABEL_PREFIX__="
  "-D__STDC_HOSTED__=1"
  "-D__DBL_DIG__=15"
  "-D__FLT_EPSILON__=1.19209290e-7F"
  "-D__GXX_WEAK__=1"
  "-D__LDBL_MIN__=3.36210314311209350626e-4932L"
  "-D__unix__=1"
  "-D__DECIMAL_DIG__=21"
  "-D__gnu_linux__=1"
  "-D__LDBL_HAS_QUIET_NAN__=1"
  "-D__GNUC__=4"
  "-D__MMX__=1"
  "-D__DBL_MAX__=1.7976931348623157e+308"
  "-D__DBL_HAS_INFINITY__=1"
  "-D__cplusplus=1"
  "-D__DEPRECATED=1"
  "-D__DBL_MAX_EXP__=1024"
  "-D__SSE2_MATH__=1"
  "-D__amd64=1"
  "-D__GNUG__=4"
  "-D__LONG_LONG_MAX__=9223372036854775807LL"
  "-D__GXX_ABI_VERSION=1002"
  "-D__FLT_MIN_EXP__=(-125)"
  "-D__x86_64=1"
  "-D__DBL_MIN__=2.2250738585072014e-308"
  "-D__FLT_MIN_10_EXP__=(-37)"
  "-D__DBL_HAS_QUIET_NAN__=1"
  "-D__REGISTER_PREFIX__="
  "-D__NO_INLINE__=1"
  "-D__FLT_MANT_DIG__=24"
  "-D__VERSION__="4.1.0 20060304 (Red Hat 4.1.0-3)""
  "-Dunix=1"
  "-D__SIZE_TYPE__=long unsigned int"
  "-D__ELF__=1"
  "-D__FLT_RADIX__=2"
  "-D__LDBL_EPSILON__=1.08420217248550443401e-19L"
  "-D__GNUC_RH_RELEASE__=3"
  "-D__k8=1"
  "-D__LDBL_DIG__=18"
  "-D__x86_64__=1"
  "-D__FLT_HAS_QUIET_NAN__=1"
  "-D__FLT_MAX_10_EXP__=38"
  "-D__LONG_MAX__=9223372036854775807L"
  "-D__FLT_HAS_INFINITY__=1"
  "-Dlinux=1"
  "-D__EXCEPTIONS=1"
  "-D__LDBL_MANT_DIG__=64"
  "-D__k8__=1"
  "-D__WCHAR_TYPE__=int"
  "-D__FLT_DIG__=6"
  "-D__INT_MAX__=2147483647"
  "-D__FLT_MAX_EXP__=128"
  "-D__DBL_MANT_DIG__=53"
  "-D__WINT_TYPE__=unsigned int"
  "-D__SSE__=1"
  "-D__LDBL_MIN_EXP__=(-16381)"
  "-D__amd64__=1"
  "-D__LDBL_MAX_EXP__=16384"
  "-D__LDBL_MAX_10_EXP__=4932"
  "-D__DBL_EPSILON__=2.2204460492503131e-16"
  "-D_LP64=1"
  "-D__GNUC_PATCHLEVEL__=0"
  "-D__LDBL_HAS_INFINITY__=1"
  "-D__INTMAX_MAX__=9223372036854775807L"
  "-D__FLT_DENORM_MIN__=1.40129846e-45F"
  "-D__FLT_MAX__=3.40282347e+38F"
  "-D__SSE2__=1"
  "-D__INTMAX_TYPE__=long int"
  "-D__GNUC_MINOR__=1"
  "-D__DBL_MAX_10_EXP__=308"
  "-D__LDBL_DENORM_MIN__=3.64519953188247460253e-4951L"
  "-D__STDC__=1"
  "-D__PTRDIFF_TYPE__=long int"
  "-D__LDBL_MIN_10_EXP__=(-4931)"
  "-D__SSE_MATH__=1"
  "-D_GNU_SOURCE=1"
  "-iwrapper/usr/share/gccxml-0.7/GCC/4.1"
  "-I/usr/lib/gcc/x86_64-redhat-linux/4.1.0/../../../../include/c++/4.1.0"
  
"-I/usr/lib/gcc/x86_64-redhat-linux/4.1.0/../../../../include/c++/4.1.0/x86_64-redhat-linux"
  
"-I/usr/lib/gcc/x86_64-redhat-linux/4.1.0/../../../../include/c++/4.1.0/backward"
  "-I/usr/local/include"
  "-I/usr/lib/gcc/x86_64-redhat-linux/4.1.0/include"
  "-I/usr/include"
  "-include"
  "/usr/share/gccxml-0.7/GCC/4.1/gccxml_builtins.h"
xml_find_template_parm encountered unsupported type 
bound_template_template_parm

Re: [pygccxml-development] first try - no good

[Neal B. <ndb...@gm...>]

OK.  I am using gccxml from cvs as of today.

gccxml hello_world.hpp -fxml=/dev/null
xml_find_template_parm encountered unsupported type 
bound_template_template_parm

Problem is with gccxml I guess.

Again, here is hello_world.hpp:

Re: [pygccxml-development] first try - no good

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

On 4/14/06, Neal Becker <ndb...@gm...> wrote:
> A simple test:
>
> g++ -fsyntax-only -xc++ hello_world.hpp produces no error.
>
> parsing files - started
> Traceback (most recent call last):
>   File "<stdin>", line 1, in ?
>   File "/usr/tmp/python-ByATVu.py", line 17, in ?
>     mb =3D module_builder.module_builder_t(
>   File
> "/usr/lib/python2.4/site-packages/pyplusplus/module_builder/builder.py", =
line
> 52, in __init__
>     , cache )
>   File
> "/usr/lib/python2.4/site-packages/pyplusplus/module_builder/builder.py", =
line
> 72, in __parse_declarations
>     decls =3D reader.read_files( files, compilation_mode )
>   File "/usr/lib/python2.4/site-packages/pygccxml/parser/project_reader.p=
y",
> line 149, in read_files
>     return self.__parse_file_by_file(files)
>   File "/usr/lib/python2.4/site-packages/pygccxml/parser/project_reader.p=
y",
> line 172, in __parse_file_by_file
>     decls =3D reader.read_file( header )
>   File "/usr/lib/python2.4/site-packages/pygccxml/parser/source_reader.py=
",
> line 164, in read_file
>     raise error
> pygccxml.parser.source_reader.gccxml_runtime_error_t: Error occured while
> running GCC-XML: xml_find_template_parm encountered unsupported type
> bound_template_template_parm
> xml_find_template_parm encountered unsupported type
> bound_template_template_parm
> [repeat...]

Lets start.

I am using Debian 3.1. Almost latests CVS version  of boost. I think this o=
ne
will become next release. I am using GCC-XML cvs version month or two month
old. Current version of cvs( =3D=3D preview ) pygccxml/pyplusplus.

I run GUI version of pypluspus, where I added to include paths boost locati=
on.
And here is the code generated by pyplusplus:

// std directories: []
// user defined directories: ['/home/roman/boost_cvs', '/home/roman/tmp']

#include "boost/python.hpp"

#include "hw.hpp"

namespace bp =3D boost::python;

BOOST_PYTHON_MODULE(pyplusplus){
    bp::def( "Copy_int"
            , &Copy_int
            , ( bp::arg("in") )
            , bp::default_call_policies() );

    bp::def( "Copy"
            , &Copy
            , ( bp::arg("in") )
            , bp::default_call_policies() );
}

So, can you try to describe your problem, one more time?

Thanks.

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

[pygccxml-development] first try - no good

[Neal B. <ndb...@gm...>]

A simple test:

g++ -fsyntax-only -xc++ hello_world.hpp produces no error.

parsing files - started
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
  File "/usr/tmp/python-ByATVu.py", line 17, in ?
    mb = module_builder.module_builder_t(
  File 
"/usr/lib/python2.4/site-packages/pyplusplus/module_builder/builder.py", line 
52, in __init__
    , cache )
  File 
"/usr/lib/python2.4/site-packages/pyplusplus/module_builder/builder.py", line 
72, in __parse_declarations
    decls = reader.read_files( files, compilation_mode )
  File "/usr/lib/python2.4/site-packages/pygccxml/parser/project_reader.py", 
line 149, in read_files
    return self.__parse_file_by_file(files)
  File "/usr/lib/python2.4/site-packages/pygccxml/parser/project_reader.py", 
line 172, in __parse_file_by_file
    decls = reader.read_file( header )
  File "/usr/lib/python2.4/site-packages/pygccxml/parser/source_reader.py", 
line 164, in read_file
    raise error
pygccxml.parser.source_reader.gccxml_runtime_error_t: Error occured while 
running GCC-XML: xml_find_template_parm encountered unsupported type 
bound_template_template_parm
xml_find_template_parm encountered unsupported type 
bound_template_template_parm
[repeat...]

[pygccxml-development] Re: MJ: CVS still a problem

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

On 4/13/06, Janes, Matthias <mat...@si...> wrote:
>  hi all,
>
> As I wrote on the 7th it seems that the CVS is not updating correctly.
>
> Neal Becker also mentioned it.
>
>
> I think in the meantime we can just stay with the official file releases =
from the April 8, 2006.
> This is what I'm using.

Yes, this release is latest good release.

> I think after the next release roman might switch to SVN. Maybe this can =
solve some of the problems (and create others ;-)

I already discussed this move with Matthias Baas. He does not have objectio=
n.
So I think that after this release we will move to SVN.


> Greetings
>
> MJ


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

Re: [pygccxml-development] cvs up to date?

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

On 4/13/06, Neal Becker <ndb...@gm...> wrote:
> I'm confused.  Is the pygccxml cvs up-to-date or not?

CVS that provides anonymous access is not up-to-date. If I understood right=
,
this is due to SF hardware failure.

You can find more information here: https://sourceforge.net/docs/A04/

Also here you can download latest sources of pygccxml and pyplusplus.
http://sourceforge.net/project/showfiles.php?group_id=3D118209
Please use *-preview packages. They contains same source as in CVS, the onl=
y
difference is documentation. CVS has much more doc string.

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

[pygccxml-development] MJ: CVS still a problem

[Janes, M. <mat...@si...>]

 hi all,

As I wrote on the 7th it seems that the CVS is not updating correctly.

Neal Becker also mentioned it.


I think in the meantime we can just stay with the official file releases =
from the April 8, 2006.
This is what I'm using.

I think after the next release roman might switch to SVN. Maybe this can =
solve some of the problems (and create others ;-)

Greetings

MJ



> -----Urspr=FCngliche Nachricht-----
> Von: Roman Yakovenko [mailto:rom...@gm...]=20
> Gesendet: Freitag, 07. April 2006 19:24
> An: Janes, Matthias
> Cc: rom...@us...;=20
> pyg...@li...
> Betreff: Re: MJ: not sure about CVS - see just old versions
>=20
> On 4/7/06, Janes, Matthias <mat...@si...> wrote:
> >
> >
> >
> >  hi roman,
> >
> > I noticed today that in the  CVS tree (Browse CVS =20
> Repository) I see just old > versions.
>=20
> It could take few days before CVS is updated and every=20
> developer can see same
> source code. Same is applied on CVS from web
>=20
> Any way thanks
>=20
>=20
>=20
>=20
> --
> Roman Yakovenko
> C++ Python language binding
> http://www.language-binding.net/
>=20

[pygccxml-development] cvs up to date?

[Neal B. <ndb...@gm...>]

I'm confused.  Is the pygccxml cvs up-to-date or not?

Re: [pygccxml-development] array_t vs pointer_t

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

On 4/11/06, Matthias Baas <ba...@ir...> wrote:
> Roman Yakovenko wrote:
> >> I'm wrapping a method that takes one single argument of type "double
> >> scale[3]". In the argument list of the corresponding declaration objec=
t
> >> this appears as a pointer_t object with a double_t object as base. Wha=
t
> >> I'm missing is the '3'.
> >> I noticed that pyplusplus defines a type array_t. Shouldn't the above
> >> argument be an array_t instead of a pointer_t to a double? Is this a b=
ug
> >> in pyplusplus, in gccxml or something else?
> >
> > I could be wrong, but this is how C/C++ works. Array is passed by point=
ers.
> > Am I wrong? Even if I wrong, GCC-XML reports this.
>
> That's too bad. :( The array size would have been a valuable piece of
> information for creating the wrappers. Without them I have to wade
> through ~250 methods manually to check if the arguments are really plain
> pointers or fixed size arrays as above (which seems to be the majority
> of cases in the Maya SDK)...

:-(.

First of all I will ask on GCC-XML list what could be done.

Second it is possible to implement solution using pygccxml only.
We can implement it and see how well it will work.
Description:
    1. Please take a look on declarations/call_invocation.py module.
        ( There is some doc strings within it )
    2. Please take a look on parser/patcher.py module.
        Basically GCC-XML has few bugs/features that pygccxml corrects.
        patch_it function is invoked from
        source_reader_t.__parse_gccxml_created_file method for every parsed
        file.

The idea is next: you can create your own patcher and rewrite/replace patch=
_it
with an other one. Every calldef declaration has location property initiali=
zed.
Thus you can get file path and line of declaration. Using type_traits
you can find
whether function has relevant type of arguments or not. After this you can =
go to
file, read original declaration, using call_invocation parser to split
it's arguments
and then find out whether it is array or not. After this you can replace ty=
pe of
argument to be an array of relevant type.

I don't say that this is a perfect method. But I think it will work in
most cases.

An other way to solve this issue is to create small parser, that will parse=
 all
header files and will create table with relevant information. Then you
can just use
that information from your script. I use this trick with tnfox. Take a look=
 on
examples/tnfox/declarations_to_exclude.py/find_deprecated function.

> In which case does pygccxml create an array_t object then?

When you have member/free variable declared as array:

const int items[23];

> - Matthias -

Hope this will help you.

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

Re: [pygccxml-development] How to customize log messages?

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

On 4/11/06, Matthias Baas <ba...@ir...> wrote:
> Roman Yakovenko wrote:
> > Can you start some to do list to next release? Small text file under
> > pyplusplus/docs will be good enough.
>
> Done.
Thanks.

> Well, I had a look at _logging_ and noticed that it's not quite true
> that one cannot set the verbosity level. As the standard logging module
> is used I can obtain the logger and adjust the level (I don't know if
> setting the level to DEBUG by default is such a good idea...).

I did not think about this a lot. But I prefer to leave it as is. You
are an experience
user of pyplusplus, other developers are not. So I think they will be glad
to see so many log messages at least one week. After this they will increas=
e
log messages level.

> The presence of the _logging_ module suggests that you already have some
> plans concerning log messages in pygccxml/pyplusplus. What plans do you
> have how logging should be done? (or if there are no plans, why didn't
> you use the logging module directly?)

Tool chain. I would like to see people start integrate code generation with=
in
their build process. Thus dedicating some logger to pygccxml/pyplusplus wil=
l
allow them better control on process.

As I already said, in future we will add under main/root project
logger, loggers for
some specific packages. Thus you will have only one point to configure
all loggers
for whole code generator.

Does it make sense? Do you agree with this. If so, I will add to "todo.txt"=
 this
explanation. So someone will be ale to implement this.

> - Matthias -

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

Re: [pygccxml-development] array_t vs pointer_t

[Matthias B. <ba...@ir...>]

Roman Yakovenko wrote:
>> I'm wrapping a method that takes one single argument of type "double
>> scale[3]". In the argument list of the corresponding declaration object
>> this appears as a pointer_t object with a double_t object as base. What
>> I'm missing is the '3'.
>> I noticed that pyplusplus defines a type array_t. Shouldn't the above
>> argument be an array_t instead of a pointer_t to a double? Is this a bug
>> in pyplusplus, in gccxml or something else?
> 
> I could be wrong, but this is how C/C++ works. Array is passed by pointers.
> Am I wrong? Even if I wrong, GCC-XML reports this.

That's too bad. :( The array size would have been a valuable piece of 
information for creating the wrappers. Without them I have to wade 
through ~250 methods manually to check if the arguments are really plain 
pointers or fixed size arrays as above (which seems to be the majority 
of cases in the Maya SDK)...

In which case does pygccxml create an array_t object then?

- Matthias -

Re: [pygccxml-development] How to customize log messages?

[Matthias B. <ba...@ir...>]

Roman Yakovenko wrote:
> Can you start some to do list to next release? Small text file under
> pyplusplus/docs will be good enough.

Done.

> Okay. Right now you can not. 

Well, I had a look at _logging_ and noticed that it's not quite true 
that one cannot set the verbosity level. As the standard logging module 
is used I can obtain the logger and adjust the level (I don't know if 
setting the level to DEBUG by default is such a good idea...).

The presence of the _logging_ module suggests that you already have some 
plans concerning log messages in pygccxml/pyplusplus. What plans do you 
have how logging should be done? (or if there are no plans, why didn't 
you use the logging module directly?)

- Matthias -

Re: [pygccxml-development] array_t vs pointer_t

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

On 4/11/06, Matthias Baas <ba...@ir...> wrote:
> Hi again,
>
> I'm wrapping a method that takes one single argument of type "double
> scale[3]". In the argument list of the corresponding declaration object
> this appears as a pointer_t object with a double_t object as base. What
> I'm missing is the '3'.
> I noticed that pyplusplus defines a type array_t. Shouldn't the above
> argument be an array_t instead of a pointer_t to a double? Is this a bug
> in pyplusplus, in gccxml or something else?

I could be wrong, but this is how C/C++ works. Array is passed by pointers.
Am I wrong? Even if I wrong, GCC-XML reports this.

I will think what can be done.

> - Matthias -

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

Re: [pygccxml-development] How to customize log messages?

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

On 4/11/06, Matthias Baas <ba...@ir...> wrote:
> Hi,
>
> how can I customize the log messages generated by pyplusplus? Is there a
> switch somewhere to set the verbosity level? Currently, when writing the
> files I get a lot of output for each single file which prevents me from
> noticing the "important" messages. Can I disable those messages (and
> maybe only get a summary) or can I customize the actual log message (so
> that I could replace the newline with a linefeed?).

:-)
Can you start some to do list to next release? Small text file under
pyplusplus/docs will be
good enough.

Okay. Right now you can not. For next release we can create different
loggers for different
components. Also we( I mean you :-) ) will define for every message
type: ( debug, info ... )
Thus you will be able to trigger messages that you want to see or even
disable whole
components.


> - Matthias -

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

[pygccxml-development] array_t vs pointer_t

[Matthias B. <ba...@ir...>]

Hi again,

I'm wrapping a method that takes one single argument of type "double 
scale[3]". In the argument list of the corresponding declaration object 
this appears as a pointer_t object with a double_t object as base. What 
I'm missing is the '3'.
I noticed that pyplusplus defines a type array_t. Shouldn't the above 
argument be an array_t instead of a pointer_t to a double? Is this a bug 
in pyplusplus, in gccxml or something else?

- Matthias -

[pygccxml-development] How to customize log messages?

[Matthias B. <ba...@ir...>]

Hi,

how can I customize the log messages generated by pyplusplus? Is there a 
switch somewhere to set the verbosity level? Currently, when writing the 
files I get a lot of output for each single file which prevents me from 
noticing the "important" messages. Can I disable those messages (and 
maybe only get a summary) or can I customize the actual log message (so 
that I could replace the newline with a linefeed?).

- Matthias -