remove lines from second matched line (including) the the end of file 2

[w5000]

input:
one
two
two
one
one
there
one

so lest say I am looking for "one" and starting from second occurence I want to remove to the end of file.

expected output to above example input:

one
two

 

[feherke]

Hi

From your description I understand this :
[pre]
one [gray]<--- 1^st occurrence[/gray]
two
two
[red][black]one[/black][/red] [gray]<--- 2^nd occurrence[/gray]
[red][black]one[/black][/red] [gray]|[/gray]
[red][black]there[/black][/red] [gray]|- remove these lines[/gray]
[red][black]one[/black][/red] [gray]<-'[/gray]
[/pre]
But then why the output contains a single "two" ?

Feherke.
[link feherke.github.com/]

http://feherke.github.com/

[/url]

 

[PHV]

Something like this ?
awk '/one/{++n}n{if(n==1)print;else exit}' /path/to/input

Hope This Helps, PH.
FAQ219-2884
FAQ181-2886

 

[w5000]

yes, you are right.

on expected output from the example should have been:

one
two
two

 

[feherke]

Hi

Ok. Then here is how I would do it :

sed '/one/{x;/^$/!Q;g}' /input/file

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

Note that my solution is not extensible. If there are chances to ever change your mind and what to remove starting from the 3^rd occurrence, then use PHV's Awk code.

Feherke.
[link feherke.github.com/]

http://feherke.github.com/

[/url]

 

[PHV]

Tested with GNU sed
With legacy sed, the result is:
sed: Unrecognized command: /one/{x;/^$/!Q;g}

With GNU sed, I get:
[!]input:[/!]
one
two
two

 

[feherke]

Hi

Thanks you for the note on legacy [tt]sed[/tt] behavior.

But I definitely not understand what you mean in your second paragraph, especially the red line :

[blue]master #[/blue] awk '/one/{++n}n{if(n==1)print;else exit}' w5000.txt 
one
two
two

[blue]master #[/blue] sed '/one/{x;/^$/!Q;g}' w5000.txt
one
two
two

[blue]master #[/blue] diff <( awk '/one/{++n}n{if(n==1)print;else exit}' w5000.txt ) <( sed '/one/{x;/^$/!Q;g}' w5000.txt )

[blue]master #[/blue] diff -y <( awk '/one/{++n}n{if(n==1)print;else exit}' w5000.txt ) <( sed '/one/{x;/^$/!Q;g}' w5000.txt )
one                                                             one
two                                                             two
two                                                             two

Teste with [tt]gawk[/tt] 4.0.1 and [tt]sed[/tt] 4.2.1.

Feherke.
[link feherke.github.com/]

http://feherke.github.com/

[/url]

 

[PHV]

This is the output given with GNU sed version 4.1.5
The red colour was to outline the garbage output.
Furthermore, I get the same result with the cygwin sed 4.2.1.

BTW, this is the 1st line of my test input file !
I did that to ensure to start at 1rst occurence of /one/ and stop at the 2nd

 

[PHV]

w5000, just to know: is your problem solved ?

 

[w5000]

yes, thank you both PHV and Feherke for your valuable posts.

chapeau bas!
kind regards,