| Re: IDL2MATLAB [message #29478 is a reply to message #19024] |
Tue, 26 February 2002 08:35   |
notspecified
Messages: 14
Registered: February 2002
|
Junior Member |
|
|
On Tue, 26 Feb 2002 10:41:07 -0500, James Kuyper
<kuyper@gscmail.gsfc.nasa.gov> wrote:
> Matt Feinstein wrote:
>
>> On 26 Feb 2002 05:02:19 -0800, the_cacc@hotmail.com (trouble) wrote:
>>
>>
>>> A long shot: is there an IDL to MATLAB source code translator out there ?
>>
>>
>> A long shot indeed. In fact, the basic syntactical aspect of such a
>> translation would be pretty trivial (except, I guess, for
>> vectorization). Unfortunately for would-be translators, most of the
>> real work in IDL and MATLAB is done by built-in functions and
>> procedures & there's no general way of doing that kind of translation.
>
> Knowing nothing about MATLAB, I'd naively expect that there should still
> be a general approach for dealing with that problem: emulation. Create a
> library of MATLAB functions and procedures (or whatever MATLAB feature
> is a good substitute for a function or procedure) that emulate the
> capabilities of IDL's built-ins.
>
Well, maybe so-- bearing in mind that both MATLAB and IDL probably
have around a thousand documented functions, completely different
graphics/GUI models, different memory models (MATLAB has no pointers
or heap variables), etc.-- I suppose you could emulate both IDL and
MATLAB with Turing Machines... Hmm.
In any event, my opinion is that the practical answer to the great
majority of IDL<->MATLAB programming questions is that the code needs
to be rewritten-- and that the sooner one realizes that rewriting the
code is actually the easy way of making a translation, the better.
Matt Feinstein does not include his email address
in the text of usenet postings.
--------
Harvard Law of Automotive Repair: Anything that goes away
by itself will come back by itself.
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
