Replace line X with String Y using sed/awk

[Database Administrator]

Hi

I have about 30 files, that are just a list of options to run a unix
program.

I need to find a way to replace a certain line number with a different
string I specify. The data in the files I am looking to replace are
dates.

Here is an example:

file1.txt:

Line 1 : Hello
Line 2 : Goodbye
Line 3 : 10102007
Line 4 : 10312007

For example, say line 3 I want to replace that date of 10102007 with
11012007.

But I do not want to have to reference the value of the file, for
example doing a : sed 's/10102007/11012007/' ... Rather specify the
whole line to replace.

How can I do that?


Thanks!

[Database Administrator]

thesoulcrusher wrote:
>
> I need to find a way to replace a certain line number with a different
> string I specify. The data in the files I am looking to replace are
> dates.
>
> Here is an example:
>
> file1.txt:
>
> Line 1 : Hello
> Line 2 : Goodbye
> Line 3 : 10102007
> Line 4 : 10312007
>
> For example, say line 3 I want to replace that date of 10102007 with
> 11012007.

sed "3s/10102007/11012007/" file1.txt

> But I do not want to have to reference the value of the file, for
> example doing a : sed 's/10102007/11012007/' ... Rather specify the
> whole line to replace.

--
Best regards | "The only way to really learn scripting is to write
Cyrus | scripts." -- Advanced Bash-Scripting Guide

[Database Administrator]

thesoulcrusher wrote:
> Hi
>
> I have about 30 files, that are just a list of options to run a unix
> program.
>
> I need to find a way to replace a certain line number with a different
> string I specify. The data in the files I am looking to replace are
> dates.
>
> Here is an example:
>
> file1.txt:
>
> Line 1 : Hello
> Line 2 : Goodbye
> Line 3 : 10102007
> Line 4 : 10312007
>
> For example, say line 3 I want to replace that date of 10102007 with
> 11012007.
>
> But I do not want to have to reference the value of the file, for
> example doing a : sed 's/10102007/11012007/' ... Rather specify the
> whole line to replace.
>
> How can I do that?
>
>
> Thanks!
>

awk 'NR==3{$0=11012007}1' file

Ed.

[Database Administrator]

thesoulcrusher wrote:
> Hi
>
> I have about 30 files, that are just a list of options to run a unix
> program.
>
> I need to find a way to replace a certain line number with a different
> string I specify. The data in the files I am looking to replace are
> dates.
>
> Here is an example:
>
> file1.txt:
>
> Line 1 : Hello
> Line 2 : Goodbye
> Line 3 : 10102007
> Line 4 : 10312007
>
> For example, say line 3 I want to replace that date of 10102007 with
> 11012007.
>
> But I do not want to have to reference the value of the file, for
> example doing a : sed 's/10102007/11012007/' ... Rather specify the
> whole line to replace.
>
> How can I do that?
>
>


sed '3s/ : .*/ : 11012007/'

awk -v v=11012007 'BEGIN{FS=OFS=" : "} NR==3{$2=v} 1'

--
Michael Tosch @ hp : com