Need to find and replace in a file

UNIX for Beginners Q & A
1

Hi All,

I am having below sample data in a file.
I need to find all the line form this file with word ABC and i need to replace the characters at position 120 which is "CO:BOGFDUI"(30chars) in the lines with blank space.
I have tried using grep to find the word with ABC (grep ABC filename), but i am unable to substite the word at this position.
Could you please help me in this.

file data:

20840GFC00524096100000LAU1LDNUSD CR       6.322017041120170411              00016574200165737CO:BASAGIO
20840GFC00524096100000LAU1LDNUSD CR       6.322017041120170411              00016574200165737CO:BRGRUUR
20840ABC00524096100000LAU1LDNUSD CR       6.322017041120170411              00016574200165737CO:BTGFGFG
20840ABC00524096100000LAU1LDNUSD CR       6.322017041120170411              00016574200165737CO:BGGFEWE
20840ABC00524096100000LAU1LDNUSD CR       6.322017041120170411              00016574200165737CO:BIGFGFG
20840ABC00524096100000LAU1LDNUSD CR       6.322017041120170411              00016574200165737CO:BOGFDUI
20840ABC00524096100000LAU1LDNUSD CR       6.322017041120170411              00016574200165737CO:BPGFDSY
20840XYZ00524096100000LAU1LDNUSD CR       6.322017041120170411              00016574200165737CO:BUGFGFG
20840XYZ00524096100000LAU1LDNUSD CR       6.322017041120170411              00016574200165737CO:BTGFFSF

output file data:

20840GFC00524096100000LAU1LDNUSD CR       6.322017041120170411              00016574200165737CO:BASAGIO
20840GFC00524096100000LAU1LDNUSD CR       6.322017041120170411              00016574200165737CO:BRGRUUR
20840ABC00524096100000LAU1LDNUSD CR       6.322017041120170411              00016574200165737
20840ABC00524096100000LAU1LDNUSD CR       6.322017041120170411              00016574200165737
20840ABC00524096100000LAU1LDNUSD CR       6.322017041120170411              00016574200165737
20840ABC00524096100000LAU1LDNUSD CR       6.322017041120170411              00016574200165737
20840ABC00524096100000LAU1LDNUSD CR       6.322017041120170411              00016574200165737
20840XYZ00524096100000LAU1LDNUSD CR       6.322017041120170411              00016574200165737CO:BUGFGFG
20840XYZ00524096100000LAU1LDNUSD CR       6.322017041120170411              00016574200165737CO:BTGFFSF
2

That specification is not too consistent: It's not clear if any ABC pattern should be matched, or just the one at positions 6 - 8. There's no 120th char position in the input sample, and esp. not 30 onwards. The replacement as well is not 30 char but 10 chars long. Will it always be the rest of the line to be replaced?
To simulate your sample output, try (GNU sed)

sed '/ABC/ s/./ /94g' file
20840GFC00524096100000LAU1LDNUSD CR       6.322017041120170411              00016574200165737CO:BASAGIO
20840GFC00524096100000LAU1LDNUSD CR       6.322017041120170411              00016574200165737CO:BRGRUUR
20840ABC00524096100000LAU1LDNUSD CR       6.322017041120170411              00016574200165737          
20840ABC00524096100000LAU1LDNUSD CR       6.322017041120170411              00016574200165737          
20840ABC00524096100000LAU1LDNUSD CR       6.322017041120170411              00016574200165737          
20840ABC00524096100000LAU1LDNUSD CR       6.322017041120170411              00016574200165737          
20840ABC00524096100000LAU1LDNUSD CR       6.322017041120170411              00016574200165737          
20840XYZ00524096100000LAU1LDNUSD CR       6.322017041120170411              00016574200165737CO:BUGFGFG
20840XYZ00524096100000LAU1LDNUSD CR       6.322017041120170411              00016574200165737CO:BTGFFSF

or

awk '/ABC/ {print substr ($0, 1, 93); next} 1' file
3

Thank you RudiC.
The ABC pattern can be any where in the file. Please find he corrected data:

20840GFC00524096100000LAU1LDNUSD CR       6.322017041120170411                       						  	       00016574200165737CO:BASAGIOSDSDSDSDSDSDSDSDSDSDSDSDSDSDSD
20840GFC00524096100000LAU1LDNUSD CR       6.322017041120170411             								       00016574200165737CO:BRGRUURSDSDSDSDSDSDSDSDSDSDSDSDSDSDSD
20840ABC00524096100000LAU1LDNUSD CR       6.322017041120170411              								       00016574200165737CO:BTGFGFGSDSDSDSDSDSDSDSDSDSDSDSDSDSDSD
20840ABC00524096100000LAU1LDNUSD CR       6.322017041120170411           								       00016574200165737CO:BGGFEWESDSDSDSDSDSDSDSDSDSDSDSDSDSDSD
20840ABC00524096100000LAU1LDNUSD CR       6.322017041120170411                       						  	       00016574200165737CO:BOGFDUISDSDSDSDSDSDSDSDSDSDSDSDSDSDSD
20840ABC00524096100000LAU1LDNUSD CR       6.322017041120170411                       						  	       00016574200165737CO:BPGFDSYSDSDSDSDSDSDSDSDSDSDSDSDSDSDSD
20840XYZ00524096100000LAU1LDNUSD CR       6.322017041120170411                       						  	       00016574200165737CO:BUGFGFGSDSDSDSDSDSDSDSDSDSDSDSDSDSDSD
20840XYZ00524096100000LAU1LDNUSD CR       6.322017041120170411                       						  	       00016574200165737CO:BTGFFSFSDSDSDSDSDSDSDSDSDSDSDSDSDSDSD
20840ABC00524096100000LAU1LDNUSD CR       6.322017041120170411            								       00016574200165737CO:BIGFGFGSDSDSDSDSDSDSDSDSDSDSDSDSDSDSD

I have tried using the awk, it replaces the pattern, but also i need other lines from the file which is not getting displayed.
I have tried using below. Could you please let me know what is next and 1 in the command are. I have also tried using sed which is not working due to syntax error. i have tried correcting it but having the same issue.
awk '/ABC/ {print substr ($0, 1, 119);}' file

4

Hi abhi_123,
Your data is not consistent. It has spaces and tabs mixed in. ^I is how I told the cat utility to show tabs.

cat -t abhi_123.example

20840GFC00524096100000LAU1LDNUSD CR       6.322017041120170411                       ^I^I^I^I^I^I  ^I       00016574200165737CO:BASAGIOSDSDSDSDSDSDSDSDSDSDSDSDSDSDSD
20840GFC00524096100000LAU1LDNUSD CR       6.322017041120170411             ^I^I^I^I^I^I^I^I       00016574200165737CO:BRGRUURSDSDSDSDSDSDSDSDSDSDSDSDSDSDSD
20840ABC00524096100000LAU1LDNUSD CR       6.322017041120170411              ^I^I^I^I^I^I^I^I       00016574200165737CO:BTGFGFGSDSDSDSDSDSDSDSDSDSDSDSDSDSDSD
20840ABC00524096100000LAU1LDNUSD CR       6.322017041120170411           ^I^I^I^I^I^I^I^I       00016574200165737CO:BGGFEWESDSDSDSDSDSDSDSDSDSDSDSDSDSDSD
20840ABC00524096100000LAU1LDNUSD CR       6.322017041120170411                       ^I^I^I^I^I^I  ^I       00016574200165737CO:BOGFDUISDSDSDSDSDSDSDSDSDSDSDSDSDSDSD
20840ABC00524096100000LAU1LDNUSD CR       6.322017041120170411                       ^I^I^I^I^I^I  ^I       00016574200165737CO:BPGFDSYSDSDSDSDSDSDSDSDSDSDSDSDSDSDSD
20840XYZ00524096100000LAU1LDNUSD CR       6.322017041120170411                       ^I^I^I^I^I^I  ^I       00016574200165737CO:BUGFGFGSDSDSDSDSDSDSDSDSDSDSDSDSDSDSD
20840XYZ00524096100000LAU1LDNUSD CR       6.322017041120170411                       ^I^I^I^I^I^I  ^I       00016574200165737CO:BTGFFSFSDSDSDSDSDSDSDSDSDSDSDSDSDSDSD
20840ABC00524096100000LAU1LDNUSD CR       6.322017041120170411            ^I^I^I^I^I^I^I^I       00016574200165737CO:BIGFGFGSDSDSDSDSDSDSDSDSDSDSDSDSDSDSD

Therefore you can not rely on positions how you have be proposing from the beginning of the thread.

Counting on this uncertainty, the following two options might or might not do what you want.

perl -pe '/ABC/ and s/CO:B\w+$//' abhi_123.example

or

perl -pe 's/(ABC.*?)CO:B\w+$/$1/' abhi_123.example