I have an output file from another product which has a variable number
of columns, followed by the row-terminator. The first four columns are
always present, the remaining 6 may or may not be present in some number.

I would like to load that data into a Sybase ASE 11.5.1 table which has
10 columns, where the first four columns are the only NOT NULL columns
in the table.

I have tried using bcp -- both with tab-delimited input and with
fixed-length input -- but the load is garbage, since bcp seems to wrap
to the next row to get all 10 columns.

Is there any option I can use to achieve my end, or does anyone have an
alternative suggestion?

Thank you.

You could massage your data to produce 10, tab-delimited fields for each
line of input. If you have Perl, try the following stdin/stdout filter
for each input file:

#!/usr/bin/perl

while(<>) {
chomp;
print join("\t",(@x = split /\t/)), "\t" x (9 - $#x), "\n";
}


Anita Craig wrote:

--


-----------------------------------------------------------------------
pjhoust (AT) canada (DOT) com
-----------------------------------------------------------------------

pjhoust (AT) canada (DOT) com wrote:
Well, unfortunately that didn't quite work. I substituted the pipe
character "|" and used input and output files, and I really had 13
(rather than 10) columns so I ran:

open (INFILE, "<Myinput.txt"); # open input file
open (OUTFILE, ">Myoutput.txt"); # open output file

while(<INFILE>){
chomp;
print OUTFILE join ("|", (@x = split /|/)), "|" x (12 - $#x), "\n";
}

Instead of 13 columns delimited by the pipe character, with trailing
ones empty, I ended up with every character of the input line delimited
by the pipe character, including the pipe character delimited by the
pipe character, with no "extra" pipes at the end.

I've not learned Perl, so I don't know what I may have wrong
syntactically -- any other suggestions?

Sample input line:
2003|1|1|AA00

Desired output:
2003|1|1|AA00|||||||||

Output I got:
2|0|0|3|||1|||1|||A|A|0|0

On Mon, 15 Dec 2003 10:57:45 -0800, Anita Craig wrote:

Use split(/\|/) ('|' is a meta character in perl's regular expression
engine.

Michael
--
Michael Peppler Data Migrations, Inc.
mpeppler (AT) peppler (DOT) org http://www.mbay.net/~mpeppler
Sybase T-SQL/OpenClient/OpenServer/C/Perl developer available for short or
long term contract positions - http://www.mbay.net/~mpeppler/resume.html