another issue:

[Jim Locker]

a statement like this will always fail, even when the referenced array contents are indeed identical: if array1[“VAL1”] == array2[1,ii] and array1[“VAL2”] == array2[2,ii] and array1[“VAL3”] == array2[3,ii] else end if The above statement will ALWAYS take the else branch. If rewritten like this: testval1 = array1[“VAL1”] : testval2 = array2[1,ii] if testval1 == testval2 testval1 = array1[“VAL2”] : testval2 = array2[2,ii] if testval1 == testval2 testval1 = array1[“VAL3”] : testval2 = array2[3,ii] if testval1 == testval2 end if end if end if it will evaluate correctly.

[Michael]

Jim wrote:
> a statement like this will always fail, even when the referenced
> array contents are indeed identical:
>
> if array1["VAL1"] == array2[1,ii] and array1["VAL2"] == array2[2,ii]
> and array1["VAL3"] == array2[3,ii]
>
> else
>
> end if
>
> The above statement will ALWAYS take the else branch.
>
> If rewritten like this:
>
> testval1 = array1["VAL1"] : testval2 = array2[1,ii] if testval1 ==
> testval2 testval1 = array1["VAL2"] : testval2 = array2[2,ii] if
> testval1 == testval2 testval1 = array1["VAL3"] : testval2 =
> array2[3,ii] if testval1 == testval2
>
> end if end if end if
>
> it will evaluate correctly.
>

Again, a small reproducible would be good here. I have occasionally seen
this sort of behavior, but I invariably discover something about my data
upon closer inspection.

Ciao, Neil

[Jim Locker]

Neil Robinson wrote:

> Jim wrote:
>> a statement like this will always fail, even when the referenced
>> array contents are indeed identical:
>>
>> if array1["VAL1"] == array2[1,ii] and array1["VAL2"] == array2[2,ii]
>> and array1["VAL3"] == array2[3,ii]
>>
>> else
>>
>> end if
>>
>> The above statement will ALWAYS take the else branch.
>>
>> If rewritten like this:
>>
>> testval1 = array1["VAL1"] : testval2 = array2[1,ii] if testval1 ==
>> testval2 testval1 = array1["VAL2"] : testval2 = array2[2,ii] if
>> testval1 == testval2 testval1 = array1["VAL3"] : testval2 =
>> array2[3,ii] if testval1 == testval2
>>
>> end if end if end if
>>
>> it will evaluate correctly.
>>

> Again, a small reproducible would be good here. I have occasionally seen
> this sort of behavior, but I invariably discover something about my data
> upon closer inspection.

> Ciao, Neil

Finding that one took me awhile, and I analyzed that data to a fare thee
well before rewriting the code; there are NO data issues.

However on this one I'll pass; setting up a demo example would take a fair
amount of time, and the issue really isn't *that* important; I'll just
know to avoid the construction that fails, and I'll also know to look for
that problem.

There are a number of unexpected subtleties in how arrays work, and I am
inclined to call these subtleties bugs because they lead to
inconsistencies in how things work. I am using them for indirection and
for hashing, and for many of those things that I would use data structures
and pointers for in C. However, the actual passing of the contents or a
proper reference to the contents is not totally reliable, as this case
indicates. I am learning that often I am safest passing the array member
contents to a temporary variable and using the temporary variable.

I would prefer to not have to do this, but needs must.

This is the same basic issue that came up with my "object not found"
post.; contents of an array (or the array reference) were not passed in
the expected fashion.

[Michael]

Jim wrote:
> Finding that one took me awhile, and I analyzed that data to a fare
> thee well before rewriting the code; there are NO data issues.
>
> However on this one I'll pass; setting up a demo example would take a
> fair amount of time, and the issue really isn't *that* important;
> I'll just know to avoid the construction that fails, and I'll also
> know to look for that problem.
>
> There are a number of unexpected subtleties in how arrays work, and I
> am inclined to call these subtleties bugs because they lead to
> inconsistencies in how things work. I am using them for indirection
> and for hashing, and for many of those things that I would use data
> structures and pointers for in C. However, the actual passing of the
> contents or a proper reference to the contents is not totally
> reliable, as this case indicates. I am learning that often I am
> safest passing the array member contents to a temporary variable and
> using the temporary variable.

I would suggest, if you have any sort of complex data, createa type to
match it, and store the type in the array. Then everything should work
as expected.

> I would prefer to not have to do this, but needs must.
>
>
> This is the same basic issue that came up with my "object not found"
> post.; contents of an array (or the array reference) were not passed
> in the expected fashion.

Arrays in SIMPOL are not arrays of a given type, and as such, they work
in slightly different ways. Array elements that contain a value are the
only case that I am aware of where the element cannot be treated as an
object. That is because the array is the object, and this is just a
value in the object. If you want the content of the array to act like
objects, assign them as object references, using =@, not =.

Ciao, Neil

[Author]

Posts