Shell Programming and Scripting

I get a CSV file every day with 2 columns and multiple rows
ex:
date1,date2 ( both the fields are varchar fields)
This data has to be updated in a table which is being done manually and i want to automate that.
1. I have to select all the data from the prod table( 2 columns { date1,date2}) into a staging table.
2. after that i have to insert the values from the CSV file into the table
3. run some validation procedures and then copy this data into prod table.
environment : sybase
OS: AIX(ksh)
I am having trouble getting data from CSV into to sybase table.
I tried to format the CSV file using SED and AWK with each column seperated by tab space and line seperated by newline.
"sed '1d' dates.txt | awk -F "," '{printf "%s\t%s\n",$1,$2}' > trial.txt"
but that gives me an when I bcp data into table. something to do with format.
any one has any idea what the right format to insert into 2 varchar columns?
or is there any other better way of doing this?

---------- Post updated at 12:02 PM ---------- Previous update was at 11:16 AM ----------

Both the fields are datetime fields.

I was thinking about storing the fields into variables and inserting these into the table using a loop.

any ideas about that ?

Firstly you need to put any code between code tags as this makes your post more readable.

Secondly you need to be specific if you want help: -

"but that gives me an when I bcp data into table. something to do with format."

I assume the missing word here is error, in which case the error message / code (between code tags) would be more helpful than "something to do with format"

Good luck

I was in a hurry and didnt pay much attention to formatting.

I found a temporary work around (I think it might be permanent solution).
earlier I converted the excel file to csv manually by copying them to a text file and seperating them by ','

This time i converted the excel to CSV by saving it as CSV and my script worked fine.

--------------------------------------
but I tried to build a script which inserts the values into the table reading the text file line by line.( I have pasted it below)

---------------------------------------------------
text.txt
----------------------------------------------------
1/1/2009,12/30/2008
2/1/2009,1/31/2009
3/1/2009,2/28/2009

TABLE :
----------------------------------------------------
has 2 columns { A (datetime),B(datetime)}

My Script
---------------------------------------------------
while read line
echo ${line}

a= `awk -F "," '{print$1} ${line}`( i kow i cant specify ${line} here- just wanted to tell you that i want to read line by line)
b= `awk -F "," '{print$2} ${line}`

isql -U${USER_NAME} -D${DB} -b -w400 -A8000<<!
${PASSWORD}
insert into TABLE values(${a},${b})
go
!
done < text.txt


i know that i have done a basic mistake specifying ${line} where AWK tries to find a file named by ${line} and scolds me.

I want your help to build something that reads the text file line by line and inserts those values into table after every line until end of file


Thanks in advance.

Code:

a= `echo ${line}|awk -F "," '{print $1}`
b= `echo ${line}|awk -F "," '{print $2}`

Whoof..I missed such a small link..Hopefully i wont ask such silly questions going forward..

Cheers rdcwayx