Hi Friends,

I would like to make a session for generating flat files having fixed length columns/records for last three months data.

I could use ttaad4226m000 but you know I cant specify date there.

I have created a session which uses concat$ but the problem is it cant create fixed length columns per record.

Can you put some light?

regards,

-Suhas

Hi,

This link might help:
http://www.baanboard.com/baanboard/showthread.php?t=42287

You can generate the Excel file then save the file as you wish.

- George

Or use sprintf$. You might have to convert some strings (or declare them) as fixed.

Hi George,

Thanks for the link. But unfortunately, there is no functionality to extract fixed length columns. You can see my comments there.

Hi Mark,

I dont think sprintf will work correctly. Because if you see seq. dump with fixed length of tdsls045, you will see BaaN coverts 0 into big exponential format.

Is there any clue/logic, how ttaad4226m000 works with fixed length column?

regards,

-Suhas

Hi suhas,

first suggestion: use bdbpre and then pipe to a 3 line awk script to "swallow" records which are too old.
second thought: for some old fashioned data exchange formats we do produce fixed length records without difficulty. Here is a sniplet of code how I do this:
----snip-----
zeil=string.set$(" ",129)
zeil(1;4)="01D1"
zeil(5;3)=sprintf$("%03d",lfd)
zeil(8;10)=mycuno
zeil(18;35)=nama
zeil(53;35)=namb
if tcmcs080.cfrw="S25" then
zeil(100;3)="OHL"
endif
zeil(129;1)=chr$(13)
seq.puts(zeil,filpo)
----snap---------
Notice how displacement 88 to 99 are neglected without any effect .... the variable "namb" is declared as string(30), not (35) nor fixed.

hth

Norbert
Norbert

Hi Mark,

I dont think sprintf will work correctly. Because if you see seq. dump with fixed length of tdsls045, you will see BaaN coverts 0 into big exponential format.

Is there any clue/logic, how ttaad4226m000 works with fixed length column?

regards,

-Suhas

Sprintf will work - you are doing the conversion with the formating. Here is an example - you can remove the fixed from buff1 and see what I am talking about. With buff1 fixed you get 132 characters in the line, with it unfixed you get a total of 40 characters in the line.


function testfile()
{
domain tcmcs.str132 buff1 fixed
error.file.name = "testit"
errfile = seq.open(error.file.name,"w")
select tiitm001.*
from tiitm001
where tiitm001.item like "MS28775.*"
as set with 10 rows
selectdo
buff1 = sprintf$("%32s %08.4f", tiitm001.item, tiitm001.avpr)
rc = seq.puts(buff1,errfile)
endselect
seq.close(errfile)
}

Hi Norbert and Mark,

I already applied suggested solutions and seem its working perfectly.

Thanks.

regards,

-Suhas