Hi,

I have several files with data that have to be imported to a database. These files contain records with separator characters. Some records are corrupt (2 separators are missing) and I need to correct them prior to importing them into the db.
Example:

|field1|field2|field3|field4|field5|field6|field7|field8|field9|field10|field11|field12|||
|field1|field2|field3|field4|field5|field6|field7|field8|field9|field10|field11|field12|field13|fiel d14|field15
|field1|field2|field3|field4|field5|field6|field7|field8|field9|field10|field11|field12|field13

The records 1 and 2 are correct. They consist of 15 optional fields separated by 15 | characters.
Record 3 is corrupt since it's missing 2 separator characters ahead field13.
It should look like:
|field1|field2|field3|field4|field5|field6|field7|field8|field9|field10|field11|field12|||field13

How can I achieve this by using sed or awk or something else?
Any help is greatly appreciated!

TIA,
Stephan.

ok - not going to win a prize for pretty code, but,

works. With the input you provided, the following is the o/p:

HTH,

Hey, thanks for your fast reply!

I don't care about pretty solutions - it works and that's important to me! I just tested on my real file and that's what I looked for!

Maybe one day I have the time to pimp the statement - but I won't win a prize as well...

Thanks again!

If that's the requirement:

Use nawk or /usr/xpg4/bin/awk on Solaris.

now there you go - that's what a real guru comes up with, and I'm impressed - never even thought of that sneaky solution.

nice piece of code radoulov - I keep learning something new from your posts - you da man.

Wow, that's a really nice solution. I never would have found that on my own - of course, that's the reason why I post to the Dummies forum... Thank you!