| Re: "private" and "public" attributes in IDL. [message #26329] |
Thu, 23 August 2001 07:21  |
Paul van Delst
Messages: 364
Registered: March 1997
|
Senior Member |
|
|
Martin Schultz wrote:
>
> JD Smith <jdsmith@astro.cornell.edu> writes:
>
>>>
>>> JD Smith wrote:
>>>>
>> ... and I for one have pined for a native method
>> within IDL to allow this. This does not mean, however, that allowing
>> such carte blanche access is always good idea. Typically, a *small*
>> subset of a class' data fields are useful and stable for public
>> consumption.
>>
>> JD
>
> Wouldn't it be lovely, had the folks at RSI thought about a "public" and "private"
> attribute for object fields?
I don't use objects alot (but I'm working on it on the weekends) but your above statement
is totally absolutely true. It would've be lovely - even for object-challenged folk like
myself. The public and private attribute in fortran 90 (for entire modules, individual
variables, structure components and/or internal subprograms) is one of the best additions
to that language I reckon.
paulv
p.s. My weekend IDL object tinkerings have certainly improved my Fortran 90 coding
techniques - or at the very least how I approach problems in that domain - that's fer
sure. And I can write my "get_properties" methods to do whatever I like. :o)
--
Paul van Delst A little learning is a dangerous thing;
CIMSS @ NOAA/NCEP Drink deep, or taste not the Pierian spring;
Ph: (301)763-8000 x7274 There shallow draughts intoxicate the brain,
Fax:(301)763-8545 And drinking largely sobers us again.
Alexander Pope.
|
|
|
|
| Re: "private" and "public" attributes in IDL. [message #26469 is a reply to message #26329] |
Thu, 23 August 2001 11:00  |
John-David T. Smith
Messages: 384
Registered: January 2000
|
Senior Member |
|
|
Paul van Delst wrote:
>
> Martin Schultz wrote:
>>
>> JD Smith <jdsmith@astro.cornell.edu> writes:
>>
>>>>
>>>> JD Smith wrote:
>>>> >
>>> ... and I for one have pined for a native method
>>> within IDL to allow this. This does not mean, however, that allowing
>>> such carte blanche access is always good idea. Typically, a *small*
>>> subset of a class' data fields are useful and stable for public
>>> consumption.
>>>
>>> JD
>>
>> Wouldn't it be lovely, had the folks at RSI thought about a "public" and "private"
>> attribute for object fields?
>
> I don't use objects alot (but I'm working on it on the weekends) but your above statement
> is totally absolutely true. It would've be lovely - even for object-challenged folk like
> myself. The public and private attribute in fortran 90 (for entire modules, individual
> variables, structure components and/or internal subprograms) is one of the best additions
> to that language I reckon.
>
> paulv
>
> p.s. My weekend IDL object tinkerings have certainly improved my Fortran 90 coding
> techniques - or at the very least how I approach problems in that domain - that's fer
> sure. And I can write my "get_properties" methods to do whatever I like. :o)
We should keep in mind, however, that most "experts" in OOP claim that
providing this type of open access is *always* a bad idea, i.e. they
lambaste the idea of having anything other than methods public; the RSI
engineers may have taken a class or two back in college from a few such
experts. Yet, IDL is not a clean and prissy langauge for academics to
build gedanken-algorithms with. It's an warty, meat and potatoes tool
for getting work done. Would you be wise to concentrate on flexible
accessibility to your object's internals? Absolutely. Should you be
forced to, even when a deadline looms? Probably not.
JD
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
