Hi,

I'm not familiar with awk, but if that's better then fine!

Basically I need to delete any lines which contains records from a SQL output.

Here is what I started coding (I'm posting only the part I need help with):


...
sqlplus -s $APPS_USR/$APPS_PWD <<ENDOFSQL
SET HEADING OFF;
SET FEEDBACK OFF;
SET LINESIZE 300;
SELECT banque, guichet, compte
FROM (select abb.bank_number banque,
abb.bank_num guichet,
abaa.bank_account_num compte,
rbi.import_date date_dernier_import,
row_number() over (partition by rbi.bank_account_id order by rbi.import_date desc) rn
from ap.ap_bank_accounts_all abaa,
ap.ap_bank_branches abb,
rb.rb_bank_accounts_all rbaa,
rb.rb_batch_import rbi
where rbaa.bank_account_id = abaa.bank_account_id
and abb.bank_branch_id = abaa.bank_branch_id
and rbi.bank_account_id = rbaa.bank_account_id
and rbaa.status = 'Y'
)
WHERE rn = 1
AND date_dernier_import > add_months(sysdate, -2);
ENDOFSQL
} | while read ligne
do
if [ "$ligne" ] # Ligne non NULL
then
set $ligne
banque="$1"
guichet="$2"
compte="$3"

sed '/'"$banque"'.*'"$guichet"'.*'"compte"'/ !d' < $fich_src > $fich_dest
cp $fich_dst $fich_src
fi
done
...

So I'd like to log which line numbers have been deleted.

Thanks