Hi,

I am new to using sed and I am trying to parse a file. I have a text file with customer information and I need to remove certain lines that do not have certain information. There are 23 lines for each customer record.

So I am trying to keep only the lines that has the customer account number and name (they are both on same line) and another line that has account type.

The lines are as follows:

1234 CustomerName
Account Type: Type1

The information starts at the beginning of each line with other lines between them.

When I try to parse each line one at a time my commands work but when I try to combine them I get a blank output file. I also have another issue where another line has a date (ie: 4/15/09) that will not be deleted because it starts with a number. I have tried to insert {3,4} after the square bracket to say that there should be 3-4 numbers but that did not help either. The file will be blank.

My commands look like this.
I tried to combine them like this but get a blank file.

Does anyone know what I might be doing wrong?

Thanks for any help.

The alternation operator (|) requires that you use extended regular expression. In SED, do this by adding the -r flag. (Or, escape the "|")

1 members found this post helpful.

Thank you very much. I put the escape character in there and it is working good.

As suggested, use "-r" - that will allow the {3,4} to work, and you won't need the escape.
Also put the carat (^) back in Account or you may get extraneous records.

1 members found this post helpful.

Thanks. That fixed the other issue. I will use "-r" instead of just the escape character.