[python(x,y)] python(x,y) source code and distributions

[Dan Savilonis]

Hi,

It seems my last couple attempts to post weren't successful, so I'm
trying again.

I've been able to download the python(x,y) source and see some of it
is versioned in mercurial, but only an old version (2.6.2.0) is
available for download. Can you upload the latest source?

Also, I've noticed the non-full editions seem to have disappeared from
the website. Do you plan to bring them back?

Regards,
Dan

--
You received this message because you are subscribed to the Google Groups "python(x,y)" group.
To post to this group, send email to pyth...@googlegroups.com.
To unsubscribe from this group, send email to pythonxy+u...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/pythonxy?hl=en.

[Pierre Raybaut]

Hi,

On Apr 23, 3:48 am, Dan Savilonis <d...@n-cube.org> wrote:
> Hi,
>
> It seems my last couple attempts to post weren't successful, so I'm
> trying again.
>
> I've been able to download the python(x,y) source and see some of it
> is versioned in mercurial, but only an old version (2.6.2.0) is
> available for download. Can you upload the latest source?

Version 2.6.2.0 is not so old... I've released v2.6.5.0 less than two
weeks ago!
I've just released v2.6.5.1 and I'm currently uploading the latest
source package (however, since I'm moving out today from my
appartment, if the upload is not complete when my internet provider
cut the the line, I won't be able to upload it before a week or so).

> Also, I've noticed the non-full editions seem to have disappeared from
> the website. Do you plan to bring them back?

Actually no, I don't. And that is only a question of time.
As you may have noticed, Python(x,y) website has moved from a non-free
host provider to Google Code (where some files were already hosted for
a long time). Since then, I've been unable to make the web installer
work and I had very little time to work this out so I chose to release
v2.6.5.0 without the web installer.

Pierre

[Dan Savilonis]

Thanks. I actually was originally looking for the source in order to
fix a problem I had with the web installer on Windows 7. Of course
this is moot if there is no longer a web installer.

However, I was thinking we could move the installer sources into a
public repository, separate from the binaries. The binaries could be
listed in a manifest which a script would use to recreate the full
build environment. Would you be interested in doing this? I'm willing
to help, I just wanted to wait for the latest sources before I
started.

Dan

[Pierre Raybaut]

On 23 avr, 18:14, Dan Savilonis <d...@n-cube.org> wrote:
> Thanks. I actually was originally looking for the source in order to
> fix a problem I had with the web installer on Windows 7. Of course
> this is moot if there is no longer a web installer.

It's still there in Python(x,y) sources but it's very buggy and badly
written (with script-based language "AutoIt"). Everything was wrote in
a hurry almost 2 years ago... things I'm really not proud of ;-)

> However, I was thinking we could move the installer sources into a
> public repository, separate from the binaries. The binaries could be
> listed in a manifest which a script would use to recreate the full
> build environment. Would you be interested in doing this? I'm willing
> to help, I just wanted to wait for the latest sources before I
> started.

Yes I'd be interested, that's a good idea. Things are not moving very
fast regarding Python(x,y) sources because I'm dedicating almost all
my free time to Spyder development for a year and a half now.

Latest sources are already available on the NTUA mirror site.
However I think I could do some clean up to make things easier for you
-- as I wrote above, everything was wrote in a hurry, meaning that
it's a complete mess. Some things are automated with AutoIt or (more
recenty) Python scripts. But so much more could be automated with not
too much efforts. And a lot of things need to be reorganized: when
looking at the sources, do not always try to find a reason of how
things are organized/done... because most of the time, choices were
made either to gain some time or to adapt old things to implement new
features. Anyway, it's a mess :-)

Now I won't be available at 100% in the next 2/3 weeks because I'm
moving out from my appartment and I'm concentrating on stabilizing
Spyder v1.1 (currently in beta test).

Thanks for your proposition.
Pierre

[Dan Savilonis]

Everyone thinks his code is a mess, but messy code that works is
always better than no code at all.

I was able to build pythonxy with only a minor modification and using
the following build requirements:

* AutoIt3 3.3.6.1
* NSIS 2.46
* NSIS 2.46-log executable

My first goal is to make it easy to contribute, so I've removed all
the binaries included and merged several repositories together:

* pythonxy
|
*-* autoit
|
* src
|
*-* eclipse (imported)
|
* other (imported)
|
* python (imported)
|
* pythonxy (renamed from Python(x,y), converted)
|
* update (imported)

I removed most files other than the nsi scripts in the eclipse, other
and python directories. I left out the wiki directory since it's not
directly related. I left some files that I wasn't sure about and
possibly deleted a few too many files :)

I've pushed the repository here:
http://bitbucket.org/djs/pythonxy

I spent some time looking for a way to recreate the full build
environment, but I didn't have a quick solution. I'm thinking the
easiest way to bootstrap the project now would be to just host the
binaries as a separate package (on the NTMU mirror, or elsewhere) and
download and extract via script.

Once the development environment is working so you can just pull from
bitbucket, then we could look into improvements. My next step is to
create a bundle of all the non-versioned files I left out.

Dan