Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.lang.python > #26252 > unrolled thread

py2c - an open source Python to C/C++ is looking for developers

Back to article view | Back to comp.lang.python

Contents

  py2c - an open source Python to C/C++ is looking for developers maniandram01@gmail.com - 2012-07-30 06:27 -0700
    Re: py2c - an open source Python to C/C++ is looking for developers andrea crotti <andrea.crotti.0@gmail.com> - 2012-07-30 16:08 +0100
    Re: py2c - an open source Python to C/C++ is looking for developers MaxTheMouse <maxthemouse@googlemail.com> - 2012-07-30 09:42 -0700
      Re: py2c - an open source Python to C/C++ is looking for developers alex23 <wuwei23@gmail.com> - 2012-07-30 17:16 -0700
        Re: py2c - an open source Python to C/C++ is looking for developers Stefan Behnel <stefan_ml@behnel.de> - 2012-07-31 07:23 +0200
        Re: py2c - an open source Python to C/C++ is looking for developers Stefan Behnel <stefan_ml@behnel.de> - 2012-07-31 14:46 +0200
          Re: py2c - an open source Python to C/C++ is looking for developers Ramchandra Apte <maniandram01@gmail.com> - 2012-09-01 20:15 -0700
            Re: py2c - an open source Python to C/C++ is looking for developers Michael Torrie <torriem@gmail.com> - 2012-09-01 23:27 -0600
              Re: py2c - an open source Python to C/C++ is looking for developers Ramchandra Apte <maniandram01@gmail.com> - 2012-09-01 23:10 -0700
                Re: py2c - an open source Python to C/C++ is looking for developers Stefan Behnel <stefan_ml@behnel.de> - 2012-09-02 08:27 +0200
              Re: py2c - an open source Python to C/C++ is looking for developers Ramchandra Apte <maniandram01@gmail.com> - 2012-09-01 23:10 -0700
              Re: py2c - an open source Python to C/C++ is looking for developers Ramchandra Apte <maniandram01@gmail.com> - 2012-09-01 23:16 -0700
                Re: py2c - an open source Python to C/C++ is looking for developers Mark Lawrence <breamoreboy@yahoo.co.uk> - 2012-09-02 09:43 +0100
              Re: py2c - an open source Python to C/C++ is looking for developers Ramchandra Apte <maniandram01@gmail.com> - 2012-09-01 23:16 -0700
            Re: py2c - an open source Python to C/C++ is looking for developers Serhiy Storchaka <storchaka@gmail.com> - 2012-09-02 10:23 +0300
              Re: py2c - an open source Python to C/C++ is looking for developers Ramchandra Apte <maniandram01@gmail.com> - 2012-09-02 06:46 -0700
              Re: py2c - an open source Python to C/C++ is looking for developers Ramchandra Apte <maniandram01@gmail.com> - 2012-09-02 06:46 -0700
          Re: py2c - an open source Python to C/C++ is looking for developers Ramchandra Apte <maniandram01@gmail.com> - 2012-09-01 20:15 -0700

#26252 — py2c - an open source Python to C/C++ is looking for developers

I created py2c ( http://code.google.com/p/py2c )- an open source Python to C/C++ translator!
py2c is looking for developers!
To join create a posting in the py2c-discuss Google Group or email me!
Thanks
PS:I hope this is the appropiate group for this message.

[toc] | [next] | [standalone]

#26260

2012/7/30  <maniandram01@gmail.com>:
> I created py2c ( http://code.google.com/p/py2c )- an open source Python to C/C++ translator!
> py2c is looking for developers!
> To join create a posting in the py2c-discuss Google Group or email me!
> Thanks
> PS:I hope this is the appropiate group for this message.
> --
> http://mail.python.org/mailman/listinfo/python-list

It looks like a very very hard task, and really useful or for exercise?

The first few lines I've seen there are the dangerous * imports and
LazyStrin looks like a typo..

from ast import *
import functools
from c_types import *
from lazystring import *
#constant data
empty = LazyStrin
ordertuple = ((Or,),(And

[toc] | [prev] | [next] | [standalone]

#26265

On Jul 30, 7:27 am, maniandra...@gmail.com wrote:
> I created py2c (http://code.google.com/p/py2c)- an open source Python to C/C++ translator!
> py2c is looking for developers!
> To join create a posting in the py2c-discuss Google Group or email me!
> Thanks
> PS:I hope this is the appropiate group for this message.

Out of curiosity.
What is the difference between this and Shedskin? Shedskin being a
(restricted) python-to-C++ compiler. (http://code.google.com/p/
shedskin/) Is the goal to be able to handle any python code or a
subset?

Cheers,
Adam

[toc] | [prev] | [next] | [standalone]

#26277

On Jul 31, 2:42 am, MaxTheMouse <maxthemo...@googlemail.com> wrote:
> What is the difference between this and Shedskin? Shedskin being a
> (restricted) python-to-C++ compiler. (http://code.google.com/p/
> shedskin/) Is the goal to be able to handle any python code or a
> subset?

There's also Nuitka, which is an unrestricted compiler, I believe:
http://nuitka.net/pages/overview.html

Is this a completely independent project, or are there plans to
leverage off of PyPy's toolchain, for example?

[toc] | [prev] | [next] | [standalone]

#26292

alex23, 31.07.2012 02:16:
> On Jul 31, 2:42 am, MaxTheMouse wrote:
>> What is the difference between this and Shedskin? Shedskin being a
>> (restricted) python-to-C++ compiler. (http://code.google.com/p/
>> shedskin/) Is the goal to be able to handle any python code or a
>> subset?
> 
> There's also Nuitka, which is an unrestricted compiler, I believe:
> http://nuitka.net/pages/overview.html

Not to forget Cython, which is the only Python-to-C compiler that is in
widespread use.


> Is this a completely independent project, or are there plans to
> leverage off of PyPy's toolchain, for example?

>From a look at the source code, it seems hard to bring it together with
anything. It looks very monolithic.

Stefan

[toc] | [prev] | [next] | [standalone]

#26306

Stefan Behnel, 31.07.2012 07:23:
> From a look at the source code, it seems hard to bring it together with
> anything. It looks very monolithic.

Hmm, sorry, I mixed it up with "2c.py", which is yet another of those
Python-to-C compilers with an all too similar name.

https://code.google.com/p/2c-python/

There are a couple of others here:

http://wiki.python.org/moin/PythonImplementations

Seeing the number of those compilers, almost none of which is commonly used
and/or still alive as a project, the question really is: why another one?

I mean, it's totally fine as a hobby educational project, sure, and I
really don't want to discourage anyone from going through this to have fun.

But apart from "just for fun", what is the goal that would make this
particular compiler different from the others? And also different enough to
merit its own source base, instead of basing it on one of the existing
projects? I don't consider "source is harder to read than to write" a good
answer to this in general.

Stefan

[toc] | [prev] | [next] | [standalone]

#28233

On Tuesday, 31 July 2012 18:16:27 UTC+5:30, Stefan Behnel  wrote:
> Stefan Behnel, 31.07.2012 07:23:
> 
> > From a look at the source code, it seems hard to bring it together with
> 
> > anything. It looks very monolithic.
> 
> 
> 
> Hmm, sorry, I mixed it up with "2c.py", which is yet another of those
> 
> Python-to-C compilers with an all too similar name.
> 
> 
> 
> https://code.google.com/p/2c-python/
> 
> 
> 
> There are a couple of others here:
> 
> 
> 
> http://wiki.python.org/moin/PythonImplementations
> 
> 
> 
> Seeing the number of those compilers, almost none of which is commonly used
> 
> and/or still alive as a project, the question really is: why another one?
> 
> 
> 
> I mean, it's totally fine as a hobby educational project, sure, and I
> 
> really don't want to discourage anyone from going through this to have fun.
> 
> 
> 
> But apart from "just for fun", what is the goal that would make this
> 
> particular compiler different from the others? And also different enough to
> 
> merit its own source base, instead of basing it on one of the existing
> 
> projects? I don't consider "source is harder to read than to write" a good
> 
> answer to this in general.

> 
> 
> Stefan

It converts to *pure* C/C++ *without* using Python or its API so that it can be the same speed as C/C++

[toc] | [prev] | [next] | [standalone]

#28236

On 09/01/2012 09:15 PM, Ramchandra Apte wrote:
> It converts to *pure* C/C++ *without* using Python or its API so that it can be the same speed as C/C++

Sounds like a fun project for you.  I hope you learn a lot doing it.
That's reason enough for it.  Do you plan to port all the standard
python modules as well, though?  Because Python modules, both in the
standard library and third-party, are the main reasons that I use
python.  For example, PyGTK.  Python is a great glue language.  Since it
can seamlessly interact with C and be extended in C, speed has never
really been an issue for me.

[toc] | [prev] | [next] | [standalone]

#28238

>That's reason enough for it.  Do you plan to port all the standard 
python modules as well, though?

Well, it should be quite easy because most of the _modules are either C accelerator (which there is no need to port) or a C wrapper (which should be trivial to port)

[toc] | [prev] | [next] | [standalone]

#28243

Ramchandra Apte, 02.09.2012 08:10:
>> That's reason enough for it.  Do you plan to port all the standard 
> python modules as well, though?
> 
> Well, it should be quite easy because most of the _modules are either C accelerator (which there is no need to port) or a C wrapper (which should be trivial to port)

Nope, not at all. They use the CPython C-API internally, so in order to
port them, you'll have to reimplement that first. That's a huge amount of
work, as proven by the incompleteness of all other Python implementations
in that regard. If you think you can do better here then IronPython or
PyPy, please go ahead.

Stefan

[toc] | [prev] | [next] | [standalone]

#28239

>That's reason enough for it.  Do you plan to port all the standard 
python modules as well, though?

Well, it should be quite easy because most of the _modules are either C accelerator (which there is no need to port) or a C wrapper (which should be trivial to port)

[toc] | [prev] | [next] | [standalone]

#28240

On Sunday, 2 September 2012 11:18:38 UTC+5:30, Michael Torrie  wrote:
> On 09/01/2012 09:15 PM, Ramchandra Apte wrote:
> 
> > It converts to *pure* C/C++ *without* using Python or its API so that it can be the same speed as C/C++
> 
> 
> 
> Sounds like a fun project for you.  I hope you learn a lot doing it.
> 
> That's reason enough for it.  Do you plan to port all the standard
> 
> python modules as well, though?  Because Python modules, both in the
> 
> standard library and third-party, are the main reasons that I use
> 
> python.  For example, PyGTK.  Python is a great glue language.  Since it
> 
> can seamlessly interact with C and be extended in C, speed has never
> 
> really been an issue for me.

I am writing py2c for my OS written in Python (planned) .
>Sounds like a fun project for you.
More fun with more developers. (I need developers)

[toc] | [prev] | [next] | [standalone]

#28248

On 02/09/2012 07:16, Ramchandra Apte wrote:
> On Sunday, 2 September 2012 11:18:38 UTC+5:30, Michael Torrie  wrote:
>> On 09/01/2012 09:15 PM, Ramchandra Apte wrote:
>>
>>> It converts to *pure* C/C++ *without* using Python or its API so that it can be the same speed as C/C++
>>
>>
>>
>> Sounds like a fun project for you.  I hope you learn a lot doing it.
>>
>> That's reason enough for it.  Do you plan to port all the standard
>>
>> python modules as well, though?  Because Python modules, both in the
>>
>> standard library and third-party, are the main reasons that I use
>>
>> python.  For example, PyGTK.  Python is a great glue language.  Since it
>>
>> can seamlessly interact with C and be extended in C, speed has never
>>
>> really been an issue for me.
>
> I am writing py2c for my OS written in Python (planned) .
>> Sounds like a fun project for you.
> More fun with more developers. (I need developers)
>

Your earlier quote "Well, it should be quite easy because most of the 
_modules are either C accelerator (which there is no need to port) or a 
C wrapper (which should be trivial to port)" means that I won't be 
touching the project with a 10 foot long disinfected barge pole.

-- 
Cheers.

Mark Lawrence.

[toc] | [prev] | [next] | [standalone]

#28241

On Sunday, 2 September 2012 11:18:38 UTC+5:30, Michael Torrie  wrote:
> On 09/01/2012 09:15 PM, Ramchandra Apte wrote:
> 
> > It converts to *pure* C/C++ *without* using Python or its API so that it can be the same speed as C/C++
> 
> 
> 
> Sounds like a fun project for you.  I hope you learn a lot doing it.
> 
> That's reason enough for it.  Do you plan to port all the standard
> 
> python modules as well, though?  Because Python modules, both in the
> 
> standard library and third-party, are the main reasons that I use
> 
> python.  For example, PyGTK.  Python is a great glue language.  Since it
> 
> can seamlessly interact with C and be extended in C, speed has never
> 
> really been an issue for me.

I am writing py2c for my OS written in Python (planned) .
>Sounds like a fun project for you.
More fun with more developers. (I need developers)

[toc] | [prev] | [next] | [standalone]

#28244

On 02.09.12 06:15, Ramchandra Apte wrote:
> It converts to *pure* C/C++ *without* using Python or its API so that it can be the same speed as C/C++

How is it implemented long integers?

[toc] | [prev] | [next] | [standalone]

#28270

On Sunday, 2 September 2012 12:53:35 UTC+5:30, Serhiy Storchaka  wrote:
> On 02.09.12 06:15, Ramchandra Apte wrote:
> 
> > It converts to *pure* C/C++ *without* using Python or its API so that it can be the same speed as C/C++
> 
> 
> 
> How is it implemented long integers?

py2c  has been just started (long integer multiplacation might use the Python internal multiplication code) (it uses the karatsuba algorithm)

[toc] | [prev] | [next] | [standalone]

#28271

On Sunday, 2 September 2012 12:53:35 UTC+5:30, Serhiy Storchaka  wrote:
> On 02.09.12 06:15, Ramchandra Apte wrote:
> 
> > It converts to *pure* C/C++ *without* using Python or its API so that it can be the same speed as C/C++
> 
> 
> 
> How is it implemented long integers?

py2c  has been just started (long integer multiplacation might use the Python internal multiplication code) (it uses the karatsuba algorithm)

[toc] | [prev] | [next] | [standalone]

#28234

On Tuesday, 31 July 2012 18:16:27 UTC+5:30, Stefan Behnel  wrote:
> Stefan Behnel, 31.07.2012 07:23:
> 
> > From a look at the source code, it seems hard to bring it together with
> 
> > anything. It looks very monolithic.
> 
> 
> 
> Hmm, sorry, I mixed it up with "2c.py", which is yet another of those
> 
> Python-to-C compilers with an all too similar name.
> 
> 
> 
> https://code.google.com/p/2c-python/
> 
> 
> 
> There are a couple of others here:
> 
> 
> 
> http://wiki.python.org/moin/PythonImplementations
> 
> 
> 
> Seeing the number of those compilers, almost none of which is commonly used
> 
> and/or still alive as a project, the question really is: why another one?
> 
> 
> 
> I mean, it's totally fine as a hobby educational project, sure, and I
> 
> really don't want to discourage anyone from going through this to have fun.
> 
> 
> 
> But apart from "just for fun", what is the goal that would make this
> 
> particular compiler different from the others? And also different enough to
> 
> merit its own source base, instead of basing it on one of the existing
> 
> projects? I don't consider "source is harder to read than to write" a good
> 
> answer to this in general.

> 
> 
> Stefan

It converts to *pure* C/C++ *without* using Python or its API so that it can be the same speed as C/C++

[toc] | [prev] | [standalone]

Back to top | Article view | comp.lang.python

csiph-web