awk script replace positions if certain positions equal prescribed value

halplessProblem · September 14, 2010, 5:01pm

I am attempting to replace positions 44-46 with YYY if positions 48-50 = XXX.

awk -F "" '{if (substr($0,48,3)=="XXX") $44="YYY"}1' OFS="" $filename > $tempfile

But this is not working, 44-46 is still spaces in my tempfile instead of YYY. Any suggestions would be greatly appreciated.

anbu23 · September 14, 2010, 5:49pm

Can you show sample input?

halplessProblem · September 14, 2010, 6:20pm

sure can

before:

RR  20100828 00599 0001230111 +00000000123     XXX 100111
SS  20100828 00599 0001230111 +00000000123     XXX 100222
TT  20100828 00599 0001230111 +00000000123     ABC 100333

hoping for after:

RR  20100828 00599 0001230111 +00000000123 YYY XXX 100111
SS  20100828 00599 0001230111 +00000000123 YYY XXX 100222
TT  20100828 00599 0001230111 +00000000123     ABC 100333

*used the code tags to keep the spacing.

danmero · September 14, 2010, 6:33pm

# cat file
RR  20100828 00599 0001230111 +00000000123     XXX 100111
SS  20100828 00599 0001230111 +00000000123     XXX 100222
TT  20100828 00599 0001230111 +00000000123     ABC 100333
# awk '{$0=(substr($0,48,3)=="XXX")?substr($0,0,43)"YYY"substr($0,47):$0}1'  file
RR  20100828 00599 0001230111 +00000000123 YYY XXX 100111
SS  20100828 00599 0001230111 +00000000123 YYY XXX 100222
TT  20100828 00599 0001230111 +00000000123     ABC 100333

sol_nov · September 14, 2010, 8:22pm

cat file.txt

RR  20100828 00599 0001230111 +00000000123     XXX 100111
SS  20100828 00599 0001230111 +00000000123     XXX 100222
TT  20100828 00599 0001230111 +00000000123     ABC 100333
RR  20100828 00599 0001230111 +00000000123     XXX 100111
SS  20100828 00599 0001230111 +00000000123     XXX 100222
TT  20100828 00599 0001230111 +00000000123     DEF 100333
RR  20100828 00599 0001230111 +00000000123     XXX 100111
SS  20100828 00599 0001230111 +00000000123     XXX 100222
TT  20100828 00599 0001230111 +00000000123     DDD 100333

awk '{ if ( substr($0,48,3) == "XXX" ) { print substr($0,0,43) "YYY" substr($0,47)} else { print substr($0,0,43) "   " substr($0,47) } }' file.txt

RR  20100828 00599 0001230111 +00000000123 YYY XXX 100111
SS  20100828 00599 0001230111 +00000000123 YYY XXX 100222
TT  20100828 00599 0001230111 +00000000123     ABC 100333
RR  20100828 00599 0001230111 +00000000123 YYY XXX 100111
SS  20100828 00599 0001230111 +00000000123 YYY XXX 100222
TT  20100828 00599 0001230111 +00000000123     DEF 100333
RR  20100828 00599 0001230111 +00000000123 YYY XXX 100111
SS  20100828 00599 0001230111 +00000000123 YYY XXX 100222
TT  20100828 00599 0001230111 +00000000123     DDD 100333

rdcwayx · September 14, 2010, 10:10pm

if no care of the output format.

awk '$6=="XXX" {$6="YYY" FS $6}1' infile

danmero · September 14, 2010, 10:21pm

Let's remove useless code

awk '{ if ( substr($0,48,3) == "XXX" ) { print substr($0,0,43) "YYY" substr($0,47)} else {print} }' file.txt

I hope you'll catch the mistake.

---------- Post updated at 10:21 PM ---------- Previous update was at 10:17 PM ----------

And if we care about output format

awk '$6=="XXX" {$1=$1 FS;$6="YYY" FS $6}1' file

rdcwayx · September 14, 2010, 10:21pm

More shorter.

awk ' substr($0,48,3) == "XXX"  { $0=substr($0,0,43) "YYY" substr($0,47)}1' infile

danmero · September 14, 2010, 10:28pm

You win 5 point ... but not shorter that my last one

halplessProblem · September 15, 2010, 10:35am

Thanks guys those work great!