comp.lang.idl-pvwave archive: archive » Re: Functions and arrays

Home » Public Forums » archive » Re: Functions and arrays

Show: Today's Messages :: Show Polls :: Message Navigator
E-mail to friend

Re: Functions and arrays [message #7601 is a reply to message #7598]

Wed, 04 December 1996 00:00

steinhh
Messages: 260
Registered: June 1994

Senior Member

> Please complain directly to RSI about this! We ran into this about a year
> ago, and were unable to get past the first level support folks telling us that
> this was an inevitable consequence of IDL's weak typing.

People at RSI (and not just support people) do read this newsgroup (and
act on it as well: XPAD/YPAD/SPACE=0 will once again rule, in IDL v 5.0!),
so this place may very well be the best place for a campaign to remove
this quite serious (and dangerous) "feature". I do think that we need to
show some enthusiasm to make it happen, though (i.e., the more people
agreeing about this being a rather nasty thing that they'd like to see
removed, the better).

I just cannot see that this is actually an *inevitable* consequence of IDL's
typing... it could be an inevitable consequence of a single-pass/single-
statement compiler, though. That, however, has nothing to do with
the language as such.

Now, assume I want to refer to an element of an array called "test1"
somewhere inside a program, e.g.,:

value = test1(0)

Assuming, of course, that I do *not* write a program that references
an undefined variable on purpose, is there *any* way of making this line
a non-crashing statement *without* referring to the name "test1" (note: no
parentheses) *anywhere* inside the same procedure?

Unless someone can actually show a compelling example of this, I'd say
that there is no problem in altering the compiler to behave reasonably,
e.g., to automatically detect which names are actually variables, and
act accordingly.

Due to the fact that one may write non-crashing programs referencing
array elements in statements placed *before* any assignment statement,
it's necessary to have either a two-pass compiler or to have some
back-tracking for this to work (this must already be done for e.g., the
labels in GOTO statements).

If this is too much of a problem, could we at least have a
"forward_array" or "forward_variable" statement ? :-)

Or at least get a compile time error about the possible mixup,
something like "Error: test1 interpreted as a function in line 5,
but as a variable in line 10".

Stein Vidar H. Haugan

Report message to a moderator

[Message index]

		Re: Functions and arrays By: David R. Klassen on Sat, 07 December 1996 00:00
		Re: Functions and arrays By: Tim Patterson on Fri, 06 December 1996 00:00
		Re: Functions and arrays By: haimov on Fri, 06 December 1996 00:00
		Re: Functions and arrays By: steinhh on Fri, 06 December 1996 00:00
		Re: Functions and arrays By: greenwoodde on Thu, 05 December 1996 00:00
		Re: Functions and arrays By: William Clodius on Thu, 05 December 1996 00:00
		Re: Functions and arrays By: thompson on Thu, 05 December 1996 00:00
		Re: Functions and arrays By: Mirko Vukovic on Thu, 05 December 1996 00:00
		Re: Functions and arrays By: brian.jackel on Thu, 05 December 1996 00:00
		Re: Functions and arrays By: Anonymous on Thu, 05 December 1996 00:00
		Re: Functions and arrays By: llobet on Thu, 05 December 1996 00:00
		Re: Functions and arrays By: Peter Mason on Thu, 05 December 1996 00:00
		Re: Functions and arrays By: thompson on Wed, 04 December 1996 00:00
		Re: Functions and arrays By: steinhh on Wed, 04 December 1996 00:00
		Re: Functions and arrays By: gunter on Wed, 04 December 1996 00:00
		Re: Functions and arrays By: rigby on Wed, 04 December 1996 00:00

Previous Topic:	Re: exp function bug
Next Topic:	Re: Q regarding spawning in windows95

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Sat Oct 11 05:52:50 PDT 2025

Total time taken to generate the page: 0.88236 seconds