[IronPython] Is any one use IronPython in your project?
sh at defuze.org
Thu Nov 23 19:32:21 CET 2006
I think the topic has been somehow polluted by the little exchange this
I'd like to set things straight. I've never suggested the IP team had
done a bad job. If I thought so I would not share my views today.
No the problem I think comes from a somehow wrong feeling (as it appears
only from my part) that because IP has gone beyond a 1.0 version it
would safely be used when coming from the Python land. I'm not saying
this is the message the IP team has passed but I don't think they have
explained enough the fact that IP was more oriented towards .NET
developers wishing to have a try at Python rather than providing a
stable implementation of Python that Python developers could use safely
to extend their own applications. [By the way the page indicating
differences between the implementations is broken on codeplex and none
of the link works.]
Unlike what Fredrik kept repeating ad libitum this morning this is not a
pray. Simply an explanation of why I got frustrated at IP. Because of
the wrong impression that the environment was complete enough to port my
own code or others'. Fredrik and I haven't the same purpose at using IP
and we could not understand each other.
> What you say is true, but I don't think the job of the IP team is to
> port all the python libraries.
> They've done an excellent job writing the whole language in C#, and some
> But there are also some users who are working hard to port many useful
> libraries too.
> Do not forget that this is also a community project, where everybody can
> Look at Seo Sangyeong work for example.
Yes I have and I do use IPCE rather than IP but that actually backs-up
my point. Have you looked at the number of modules Seo has added to IPCE
to make it useful enough for Python developers wishing to move slowly to
.NET while keeping many modules they've been using for so long? You
cannot expect developers will learn the whole .NET framework that
quickly or even like it, can you? There is a demand.
Again I assume that it would also attract some .NET developers to play
with TurboGears or Django as much it could attract Python developers to
incorporate the huge .NET framework functionalities into their application.
When IP 1.0.1 was released I naively thought it wouldn't be such a long
way to get those Python packages working (they don't by the way). Maybe
a few of them to tweak but not much more. But I was wrong. My fault
maybe but this thread was about to say whether or not we were using IP.
Currently I'd love to but I can't because my goal is not to re-write my
entire code. Fredrik was inviting me to use IP for new code rather than
bothering with existing one. Sure this would work but I have no will to
drop them, instead I'd like to propose them to the .NET developers and
maybe learn from their point of views.
> Is there any library you would like to see ported to .Net?
> Well, why don't you work on it? Nothing prevents you from doing it.
> In fact, it's just a matter of time until every useful library is ported
> to .NET.
I had started a few month ago based on Seo's work but there was so many
things preventing me to succeed that I had to stop for the time being
until IP would get a little more mature.
Besides my time is as limited as anyone else and I already work on many
projects. Does that I mean I have to shut it down?
> The real value of IP is that it (will) let you use the best of both
> worlds, always using the same nice syntax we like so much.
> If you are only interested in using IP libraries, I don't see the point
> of using IP.
I think I can't explain myself correctly today :(
I would reply however that if the point of IP was solely to let .NET
developers play with a Python syntax and type less code then effectively
I have nothing to do with IP. I assume of course I'm wrong here in the
fact that the IP teams aims at providing an implementation that both
communities feel safe to engage for their own project and slowly open
their door to the other side.
> But having all the .NET libraries available to you from IP, plus all the
> other libraries that overtime will be ported to .Net, that's just great.
Yes. That was exactly my point, wasn't it? Overtime it will get just
great. The question was as IP stands today and I answered as it is today
stating exactly what you just said.
> I would be happy if the IP team concentrates all their efforts in just
> making IP rock solid, stable and (whenever possible), speedy.
So would I. I guess that's what they do everyday.
Now before I get flamed again, I repeat that I very much love the work
done so far by the IP team and I am sure I will enjoy IP as it grows but
currently for my purpose (although I assume I'm not the only one) the
current status of the project does not make me feel confident enough to
support it (well I do support it in some limited areas).
More information about the Ironpython-users