Replace two characters in the middle of a string 1

[reclspeak]

Hi!

I am scripting a solution that requires two characters to be replaced in a string that appears in a file (with other unrelated strings) and I'm struggling to figure-out a routine that will satisfy the requirement.

The string is always consistent, and is in the form;

INV_20070430131255_H062849B.DAT

It appears in a file as the 8th record on the first line of numerous lines. Each record is delimited with a "|" character.

The string after "INV" is the date, month, day, hour and minute and then millisecond (in this case "55"). After this string, there is another underscore, followed by a string that always begins with "H" and ends with ".DAT".

What I am trying to do is replace just the two characters for the "55" with "zz" These are always the 13th & 14th characters in the long numeric string after "INV_"

I'm struggling to find a placeholder routine that will zero-in on just those two digits.

Any assistace with this would be gratefully received.


recl

 

[feherke]

Hi

sed 's/.._/zz_/2' /input/file

Tested with GNU [tt]sed[/tt].

Feherke.

http://rootshell.be/~feherke/

 

[feherke]

Hi

awk '{print substr($0,1,16)"zz"substr($0,19)}' /input/file

Tested with [tt]gawk[/tt] and [tt]awk95[/tt].

Feherke.

http://rootshell.be/~feherke/

 

[reclspeak]

Thanks Feherke,

I'll give them a try tomorrow morning (in the UK) and report how I got on.

recl.