I was looking for some help with output conversion codes, to see if there
was a simple way to output dates in YYYY-MM-DD format when using SORT or
LIST. CALLing a subroutine will probably be too slow for this particular
application.

Thanks,

Scott Deardorff

0008 A;23(D/)]A;LPV(G2/1):'-':LPV(G0/1):'-':LPV(G1/1)

where 23 = your date attr; ] is @VM

LIST TKT.TEST ID.SUP F23 DTE 05:16:35pm 16 Jan 2007 PAGE 1
F23............ DTE.........

9141 1993-01-09
9020 1992-19-01

Chandru

"Scott Deardorff" <smdamd (AT) bellsouth (DOT) net> wrote

Hi,

On Tue, 16 Jan 2007 17:05:11 -0600, "Scott Deardorff"
<smdamd (AT) bellsouth (DOT) net> wrote:

I don;t know much about "proper Pick" but doesn't something like
D4-YMD[,2,] work there, too? Does in Universe, etc

Hope I've been of assistance

Jaye

Since D3/NT 7.5 and D3/Linux 7.4.2 (starting with patch A206??)
the LPV doesn't work anymore as in previous releases.

I propose this two variants:

1) Direct reference to the attribute 23

0008 A;23(D/)(G2/1):"-"23(D/)(G1/1)):"-"23(D/)(G0/1))


2) Direct reference to attribute 23 and SET-DATE-xxx independent

0008 F;23;(D4Y);"-";:;23;(DM);(MR%2);"-";:;:;23;(DD);(MR%2);:


Thanks,
Stefano


Chandru Murthi ha scritto:

Stefano - major - major BUG!!!!
By any chance did you or yours report this to RD, and if so, what was their
response?
Curoius,
Tom

"stefano" <stefano (AT) friuldata (DOT) it> wrote

This works in Universe:

DYMD['-','-']

Ed

"Scott Deardorff" <smdamd (AT) bellsouth (DOT) net> wrote

Thanks for all the replies. I tried most of the solutions, and went with a
modified version of Stefano's first suggestion.

008 A;23(D4/)(G2/1):"-"23(D4/)(G0/1)):"-"23(D4/)(G1/1))

Just had to flip the month and the day to get YYYY-MM-DD. Even though the
D/ was defaulting to a 4 digit year, I also changed it to D4/ for
consistency with our other standards.

Found this worked great with D3NT and D3 Linux.

Thanks again for the help!

How about formatting the date with "-" instead of "/", then you can get
the mm-dd in one chunk with a single G0-2 instead of G0/1 + G1/1:

008 A;23(D4-)(G2-1):"-":23(D4-)(G0-2)

(Thats the big advantage of the G correlative vs the FIELD function.)

/Scott Ballinger
Pareto Corporation
Edmonds WA USA
206 713 6006

Scott Deardorff wrote:

"Scott Ballinger" <scott.ballinger (AT) gmail (DOT) com> wrote

Used to be. All FIELD functions have a chunk count as in x =
FIELD(string,'*',2,3)

C

Didn't know that! But too bad the syntax is not the same as the G
correlative:

x = "aaa.bbb.ccc"
oconv(x,"g1.2") = bbb.ccc
field(x,".",1,2) = aaa.bbb

Actually, the field syntax ("starting with" vs. "skipping") is probably
more intuitive, but how many ways do we really need to skin this cat?

/Scott

Chandru Murthi wrote: