I have a report that runs against a flatfile and only generates data if it finds an error. This is a report we actually want to come back empty.

Right now it just checks the field F11 to make sure it is not equal to 0.

WHERE (ELAPSED_TIME_CSV.ELAPSED_TIME_CSV.F2 IS-NOT MISSING )  
OR  (ELAPSED_TIME_CSV.ELAPSED_TIME_CSV.F4 NE '0')  
OR  (ELAPSED_TIME_CSV.ELAPSED_TIME_CSV.F7 IS-NOT MISSING )  
OR  (NOT ELAPSED_TIME_CSV.ELAPSED_TIME_CSV.F8 IN ('A'))  
OR  (ELAPSED_TIME_CSV.ELAPSED_TIME_CSV.F11 EQ 0) 
OR  (ELAPSED_TIME_CSV.ELAPSED_TIME_CSV.F12 IS-NOT MISSING )

Now I need to check it to make sure the last 5 digits are all zero's. This is payroll data and the system the flatfile is imported to only does time in 1/10th of an hour (6, 12, 18, 24, 30, etc) shown in the flat file in decimal 10ths as .1 .2 .3 .4 etc.

In this example:

"REG","10.250000","","0"""
 "REG","10.500000","","0","""

Line one would be in the report as it is not a tenth, but line 2 would pass.

So I need another OR statement for F11 that will make sure
XX.X99999 that the 9 decimal places are equal to Zero.

Am I even barking up the right tree?

This message has been edited. Last edited by: Greg,

Whats the format of the field ?

You could right justify the field, then pull the last 5 characters to check that they are all zeros.

They are all Alpha since the reports runs against a flat file.

Or EDIT([thefield]) and check for the specific characters in the resultant string...

IF F11 EQ '00.000000' THEN 'This' ELSE 'That' ;

Perhaps there is a way to SUM the value of that field to make sure it is divisible by 10? and therefore correct? .1 .2 .3 .4 etc.

Change it to numeric and do the math?

Right Justify and pull the last 5 chars.

EX -LINES 7 EDAPUT MASTER,TST_DATA,CV,FILE
FILENAME=TST_DATA, SUFFIX=COM,$
SEGNAME=TST_DATA, $
  FIELD=FIELD1 ,ALIAS=  ,A20 ,A20 ,$
  FIELD=FIELD2 ,ALIAS=  ,A20 ,A20 ,$
  FIELD=FIELD3 ,ALIAS=  ,A20 ,A20 ,$
  FIELD=FIELD4 ,ALIAS=  ,A20 ,A20 ,$

FILEDEF TST_DATA DISK tst_data.ftm (RECFM V
SET PCOMMA=ON

EX -LINES 3 EDAPUT FOCTEMP,TST_DATA,CV,FILE
"REG","10.250000","","0"
"REG","10.500000","","0"

-RUN

TABLE FILE TST_DATA
PRINT *
WHERE EDIT(RJUST(20,FIELD2,'A20'),'$$$$$$$$$$$$$$$99999') NE '00000'
END

Since my F11 field is A255V, will I need 250 $ dollar signs for the RJUST statement.

Like this:

 WHERE EDIT(RJUST(255,FIELD11,'A255V'),'$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$99999') NE '00000' 

You could reverse the string and then it would be the first 5 chars you need

EDIT(REVERSE(255,RJUST(255,TXT,'A255'),'A255'),'99999');

Or you could use SUBSTR.

quote:

Originally posted by Waz:
Or you could use SUBSTR.

This turned out to be the solution.

DEFINE DECIMAL_STRING_6/A6= GETTOK(F11X, 10, 2, '.', 6, DECIMAL_STRING);

THis count the 6 spaces to the right of the decimal and made sue the last 5 were all Zero's.

Function CHKFMT might be another option:

F11_CHECK/I1 = CHKFMT(10,F11,'99.900000','I1');

F11_CHECK will be zero if the value in F11 fits the mask, otherwise it contains the position of the first character failing the mask.