[SciPy-User] ide for numpy?
Wed Oct 6 08:36:07 CDT 2010
On Wed, Oct 6, 2010 at 9:24 AM, Neal Becker <firstname.lastname@example.org> wrote:
> email@example.com wrote:
>> On Wed, Oct 6, 2010 at 8:50 AM, Dharhas Pothina
>> <Dharhas.Pothina@twdb.state.tx.us> wrote:
>>>>>> "Ioan-Alexandru Lazar" <firstname.lastname@example.org> 10/6/2010 6:58 AM >>>
>>>> Have a look at Spyder. It's Matlab-like and comes with a number of
>>>> goodies like Pylint integration.
>>> Spyder is definitely worth looking at. I prefer using emacs and ipython
>>> for most of what I do but my less technical colleagues have found spyder
>>> easier to use. The interface is more matlab like and also it doesn't feel
>>> like they are having to muck around in the command line for stuff like
>>> changing directories and running scripts etc.
>> I have been using Spyder for more than a year as my main interactive
>> shell. The new 2.0 is currently in beta, and has a few problems left,
>> but looks very good, especially coming from a matlab ide experience.
>> Now, it also has a ipython shell integrated (with matlab like name
>> space pollution, "from mlab import *", by default).
>> I wouldn't recommend eclipse for non-software-developers. I like it to
>> get around in a package, but I found the views confusing when I
>> started with it and, the pydev version that I use doesn't come with an
>> interactive shell.
>> Otherwise my favorite is IDLE, plain and simple and (almost) no magic.
> I'm trying spyder 2.0beta4. Looks interesting, but I'm not quite sure how
> to use it.
> I was expecting to load a program, then be able to set breakpoints, and
> examine variables.
> I see I can use the interactive console to manually enter:
> u = np.arange (10)
> and then see it in variable window and right click to plot.
> I can set a breakpoint in my source code (displayed on the left in editor),
> but only does something if I run using debug. But running from debug puts
> me in pdb. Not sure if trying to do everything under pdb is newb friendly.
> Am I missing something?
No, debugging is still one of the weak points. And I still don't know
how to use pdb.
I haven't tried out how the pdb integration works.
(I'm debugging with print statements, and stack inspection after an exception.)
I played a bit with winpdb in the past, and it looks more beginner friendly.
> SciPy-User mailing list
More information about the SciPy-User