| Re: Oh No... [message #53500] |
Wed, 18 April 2007 06:45  |
Paolo Grigis
Messages: 171
Registered: December 2003
|
Senior Member |
|
|
F�LDY Lajos wrote:
>
> On Wed, 18 Apr 2007, Paolo Grigis wrote:
>
>>
>> Speaking of which, what about
>>
>> IDL> delvar,a ;undefines a
>> IDL> a[(a=2*findgen(10))]=1 ;agreed, this is a bit crazy, but hold on
>> IDL> print,a
>> 1.00000 1.00000 1.00000 6.00000 8.00000
>> 10.0000 1.00000
>> 14.0000 1.00000 1.00000
>>
>> Now, this must be a bug, surely... (or a very strange feature indeed).
>>
>
> No, invalid code, with undefined result :-)
>
> We have no formal definition of the IDL syntax, but there is a rule in
> Fortran/C/C++/... which should be true for IDL, too: a memory cell can
> be modified at most once in an expression. You are modifying 'a' twice
> in a single expression.
Yes, this seems a sensible precaution, but then I think that in the interest
of safety it may be better if such an expression would throw a compiler
or at least a runtime error in IDL... I don't think there is much in the
way of a sensible usage for such kind of expressions, so not much would
be lost.
On the other hand, a similar example which does not access memory cells
out of the array boundaries seems to function more or less as one would
expect:
IDL> a[(a=0.5*findgen(10))]=7*findgen(10)
IDL> print,a
7.00000 21.0000 35.0000 49.0000 63.0000 2.50000 3.00000
3.50000 4.00000 4.50000
so in IDL such expression seems to be valid, and the innermost array
is generated first, and its values are then used for indexing itself...
Ciao,
Paolo
>
> (Section 6.5#2 of the C99 specification: "Between the previous and next
> sequence point an object shall have its stored value modified at most
> once by the evaluation of an expression. Furthermore, the prior value
> shall be accessed only to determine the value to be stored.")
>
> regards,
> lajos
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
