| Motley library on Sourceforge [message #90191] |
Mon, 09 February 2015 17:58  |
![Mark[1] is currently offline](theme/default/images/xoffline.png.pagespeed.ic.XRkd1fkXye.png "Mark[1] is currently offline")
Mark[1]
Messages: 66
Registered: February 2008
|
Member |
|
|
The old-timers on the group may remember Motley, a library of IDL routines that I published from time to time, notably on David Fanning's Coyote site
http://www.idlcoyote.com/hadfield/idl/README.html
I see this file was published in 2009. That was a long time ago. My wife & I have acquired 7 grandchildren since then.
I have continued to use IDL and to maintain and develop Motley in minor ways. I have now published it on SourceForge, here
https://sourceforge.net/projects/mghmotley/
I am new to managing Sourceforge projects, but i think the best way to acquire it is to go to the project's Mercurial page
https://sourceforge.net/p/mghmotley/mercurial/ci/default/tre e/
From here you can clone the Mercurial repository (which I recommend, as Mercurial is a very good revision control tool) or download a snapshot as a ZIP archive.
Incidentally, what I do in my day job is ocean modelling using the ROMS model from Rutgers University. I have developed quite a collection of tools for preprocessing, postprocessing and analysing ROMS files. However I feel a bit like Don Quixote tilting at IDL windmills while my colleagues are getting the job done (less elegantly in my opinion) with Matlab and Python. So I think I will be tapering off my IDL activity and I may try to port some of my code to Python.
|
|
|
|
| Re: Motley library on Sourceforge [message #90194 is a reply to message #90191] |
Mon, 09 February 2015 19:57  |
chris_torrence@NOSPAM
Messages: 528
Registered: March 2007
|
Senior Member |
|
|
On Monday, February 9, 2015 at 6:58:23 PM UTC-7, Mark Hadfield wrote:
>
> Incidentally, what I do in my day job is ocean modelling using the ROMS model from Rutgers University. I have developed quite a collection of tools for preprocessing, postprocessing and analysing ROMS files. However I feel a bit like Don Quixote tilting at IDL windmills while my colleagues are getting the job done (less elegantly in my opinion) with Matlab and Python.
Mark, don't give up hope! I would agree with you 100% on the "tilting at windmills", and on your analysis of Matlab and Python. It's funny, there are all these comparisons of IDL and Python, with the conclusion that Python is "so much better". In my opinion, Python is *just as wacky* of a language as IDL. They both have their strange quirks, inconsistencies, and 50 different ways to accomplish the same task. But... it's hard to argue with free software.
Anyway, I would recommend that you avoid porting your code. We're actually working on a Python<->IDL bridge that should make your life a lot easier. You'll be able to call Python code from within IDL, and IDL code from within Python, pass variables and objects back & forth, etc. So if you have a library of IDL code that people find useful, they will be able to seamlessly call that from within Python (assuming you have an IDL license).
Finally, thanks so much for publishing your code on SourceForge! It's great to have these libraries up to date.
Cheers,
Chris
p.s. I'm hoping I'm retired by the time I have 7 grandchildren...
|
|
|
|
|
|
| Re: Motley library on Sourceforge [message #90210 is a reply to message #90194] |
Wed, 11 February 2015 01:51 |
Fabzi
Messages: 305
Registered: July 2010
|
Senior Member |
|
|
On 10.02.2015 04:57, Chris Torrence wrote:
> In my opinion, Python is*just as wacky* of a language as IDL.
haha, true! I dislike numpy's syntax (who wants to write
np.array([1,2,3]) for getting a [1,2,3] array???), and I dislike
matplotlib even more.
However, after trying stuffs like Pandas, Scikit-Learn and Statsmodels,
i've found the limitations of IDL. So a bridge will be welcome! Have you
thought about making an official Jupyter (ex-iPython notebook)-Jupyter
bridge? This would make IDL faster than Matlab in adopting the fancy
"cool kids" stuffs.
Fabien
|
|
|
|
| Re: Motley library on Sourceforge [message #90217 is a reply to message #90210] |
Wed, 11 February 2015 14:33 |
Michael Galloy
Messages: 1114
Registered: April 2006
|
Senior Member |
|
|
On 2/11/15 2:51 AM, Fabien wrote:
> On 10.02.2015 04:57, Chris Torrence wrote:
>> In my opinion, Python is*just as wacky* of a language as IDL.
>
> haha, true! I dislike numpy's syntax (who wants to write
> np.array([1,2,3]) for getting a [1,2,3] array???), and I dislike
> matplotlib even more.
>
> However, after trying stuffs like Pandas, Scikit-Learn and Statsmodels,
> i've found the limitations of IDL. So a bridge will be welcome! Have you
> thought about making an official Jupyter (ex-iPython notebook)-Jupyter
> bridge? This would make IDL faster than Matlab in adopting the fancy
> "cool kids" stuffs.
>
> Fabien
Yes, this one is good, but does not work completely:
https://github.com/lstagner/idl_kernel
An official one would be great!
Mike
--
Michael Galloy
www.michaelgalloy.com
Modern IDL: A Guide to IDL Programming (http://modernidl.idldev.com)
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
