*Newbie Question* --> Is it viable to distribute Python apps over the Internet?

[Alex Martelli]

Denis Agar wrote:
   ...
> Ok, try to ignore the thickness of my skull, but you're saying that it
> is possible to create progams using mfc42.dll, a GUI that most people
> have already? I have the win32 extensions and am in the process of

Yes, with Python and a simple add-on you can choose to reuse any DLL
you may have lying around.  win32all (PythonWin in particular) does
that for MFC specifically.  Other add-ons such as the oldie (but
goldie) calldll, and the newer (and more powerful) ctypes, let you
do that for ANY dll (with more Python-side work to adapt it).  Good
luck -- it IS, of course, lower-level programming than typical Python
work, since you want to use DLLs designed for lower-level languages.

> trying to understand them. (PS: Anyone got some (s)ample code they
> could throw my way?)
> But, as I understand it, this would be hard.

Of course it would.

> This isn't a request nor a demand nor a manifesto of any kind, but ---
> would it be possible to make a simple PythonCardish GUI shaping-type
> languagey thing (I'm not yet up on the lingo) that interfaces with the
> win32 extensions and dumbs it down to about my level? I'm just
> wondering about it is all. . .

The anygui project, anygui.sf.net, aims to design a uniform wrapper
that can use most any GUI toolkit as a back-end -- this would include
(at least potentially) MFC, or the bare Win32 API's, etc.  It's not
moving very fast, but there's nothing specific holding it back save
for big ambitions and scarce time being spent on it;-).  Once anygui
is stable, and there exists a working back-end for your chosen GUI
toolkit[s], it would surely be possible to write a screen-painter
front-end on top of it -- there is no conceptual block against that.

The pragmatical likelihood of this actually happening, like for any
other open-source software project, depends on how likely it is that
some active developers find that investing their time and energy into
it "scratches their itch".  anygui's slow progress so far seems to
suggest that, while intellectually interesting and stimulating, it
may not speak to an actual NEED of its developers -- so, it becomes a
"back-burner", "in our copious spare time" project, and those do not
tend to move forwards all that rapidly.


Alex