Re: IDL or PV~Wave? [message #1028 is a reply to message #1025] |
Mon, 24 May 1993 14:41   |
thompson
Messages: 584 Registered: August 1991
|
Senior Member |
|
|
alan@elroy.Jpl.Nasa.Gov (Alan S. Mazer) writes:
> In article <1993May21.164403.1583@ll.mit.edu> knight@ll.mit.edu (Fred Knight) writes:
>> Here are some features added to IDL since v2.1 that I particularly like:
>>
>> ---integrated support for HDF, CDF, and netCDF.
>> ---on-line help in a separate window
>> ---IOCTL to access low-level file operations
>> ---logarithm Z axis for surface plots
>> ---anonymous structures
>> ---create_struct (v3.1)
>> ---reads: like readf only from an IDL variable instead of from a file
>> ---device independent screen preview in EPSI format to see IDL plot in desktop publishing progams
>> ---integrated TIFF and GIF support
>> ---format capability for tick labels on plots
>>
>> Are there PV~WAVE equivalents?
> Not that I know of. It's hard to know what some of these things are (e.g.,
> anonymous structures). The question really is, what do you need? Wave has
> Motif lookalike functions that allow one to develop Motif GUIs very rapidly.
> In addition to their regular widgets. And I find the Wave widgets easier to
> use. Wave doesn't crash on me, locking up my workstation. Perhaps the
> stuff you've cited is critical to other people, but for me Wave offers the
> right combination of reliability and flexibility (and polish) that I need.
> I recommend it enthusiastically.
> --
>
> -- Alan # Mountain Dew and doughnuts...
> ..!ames!elroy!alan # because breakfast is the most important meal
> alan@elroy.jpl.nasa.gov # of the day.
Sounds like you must have had a bad experience with IDL somewhere along the
line. Just for the record, IDL has never locked up my workstation, and I don't
know anybody for whom it has. I certainly don't think of IDL as an unreliable
product, and certainly have no reason to believe that PV-Wave is going to be
any more (or less) reliable. :^)
Anonymous structures are unnamed structures that the user creates on the fly.
It's easier to explain by describing named structures first. For instance,
suppose I define a named structure through the command
IMAGE = {SOHO_IMAGE, IMAGE: INTARR(1024,1024,/NOZERO), $
PMT: 0.0, $
X: 0L, $
Y: 0L, $
Z: 0L}
then the name of this structure is SOHO_IMAGE. I can then define variables as
being of this datatype, e.g.
A = {SOHO_IMAGE}
However, once a structure is defined, then it is static. For instance, A.IMAGE
will always be a 1024x1024 integer array.
Anonymous structures are defined on the fly, and are thus more dynamic. For
example, if I write
A = {IMAGE: myimage,
PMT: pmt-reading,
X: x-value,
Y: y-value,
Z: z-value}
then A doesn't have a structure name associated with it. I can thus create
structures with just the structure that I need when I need it, rather than
defining it for once and always at the start of the program as in C or FORTRAN.
In IDL anonymous structure variables can be redefined dynamically.
Bill Thompson
|
|
|