Searching in fixed locations with AWK...

[bondtrails]

Hi all,
how can I match using fixed locations. It seems like AWK allows pattern matching, but I need to match in a fixed position. Ex:
FileA.txt contains:
aaaaJobCode bbbbb
aaaaJobControlbbbbb
aaaaJobStatus ccccc
eeeeJob dddddd

If I want only those records that match "Job " in locations 5-13, using awk '/Job /' FileA.txt won't work.

How to do?? I need to search by fixed location!!

--Bondster!!

 

[CaKiwi]

Maybe

'/^....Jobs /'


CaKiwi

"I love mankind, it's people I can't stand" - Linus Van Pelt

 

[Baraka69]

I believe CaKiwi got an extra "s" in, so it should actually read:
[tt]
'/^....Job /'
[/tt]

I really wonder, why your example does not work for you, because it does for me ... your pattern matches exactly the desired line in your example ... that is really odd.

have a look at relative expressions in general.

 

[bondtrails]

Okay, thanks CaKiwi and Baraka, I got it to work
However, in the general sense, my question still lingers. If I wanted to search records of a file, but limiting my search within specific column locations, how would I do it. I am thinking perhaps the function "substr" might do it. Would the following work?
awk 'substr($0,start, length) ~ /REGEXP/ {print $0}'

Or is there a more efficient way to go about it?

--Bondster!!
P.S. THis is an awesome forum!

 

[Salem]

> awk 'substr($0,start, length) ~ /REGEXP/ {print $0}'
That's the normal way

But if you have GNU AWK instead of the more regular versions, then you could try this

http://www.gnu.org/manual/gawk-3.0.3/html_mono/gawk.html#SEC44