Re: Constant arrays

Now this I like... and not only for arrays, I'd like to see it for data structures, too :)

Though, I wouldn't use square brackets... I'd do something like this instead:

arr(*) = ('*PGM' : '*SRVPGM' : '*CMD': '*FILE');

Though, IBM will probably want to make it into a BIF:

arr(*) = %val( '*PGM': '*SRVPGM': '*CMD': '*FILE );


On 4/4/2011 9:27 AM, hockchai Lim wrote:

Not a big fan of using D spec to create compile time array. Mainly because
of fix lenght on D spec. CTDATA seems much cleaner.

Unless D spec goes free form, instead of creating a new InzArray D spec
keyword, I would rather see RPG allowing us assigning multiple values to
multiple elements in one eval statement:

arr[] = ('*PGM':'*SRVPGM':'*CMD':'*FILE');


"Barbara Morris"<bmorris@xxxxxxxxxx> wrote in message
news:mailman.37138.1301747060.2702.rpg400-l@xxxxxxxxxxxx...

On 4/1/2011 5:19 PM, Alan Campin wrote:

Would be nice if IBM would just give use an INZARRAY keyword.

InzArray('*PGM':'*SRVPGM':'*CMD':'*FILE')

On 4/1/2011 6:29 PM, Scott Klement wrote:

Ick.

Why ick? I don't think a new keyword is needed; INZ could do fine, just
with a new rule that allows multiple values for arrays.

It wouldn't be great for longer values, but that's true of all INZ values.

D arr s 100a dim(2)
D inz(
D 'I must go down to the +
D sea again, +
D to the lonely sea +
D and the sky,' :
D 'And all I ask is a tall +
D ship and a star +
D to steer her by,')