Help with Joining lines in a file

[HHG]

Hi All,

Can anyone help I am pain stakingly trying to go through a file and in vi trying to join to line together. But the file is rather large there must be an easier way to do this.

e.g file :-

2226697 29/01/08
2226697 "630511" "0P0705011" 1.00
2226698 29/01/08
2226698 "630571" "0P0705071" 1.00
2226744 29/01/08
2226744 "396001" "0C2660001" 1.00
2226756 29/01/08
2226756 "630101" "0P0701001" 1.00
2226789 29/01/08
2226789 "620113" "0P0601013" 1.00
2226799 29/01/08
2226799 "815005" "0E0115005" 1.00
2226799 "140009" "0B0840009" 1.00
2226799 "721001" "0G0121001" 1.00
2226799 "730001" "0G0230001" 1.00
2226799 "855010" "0E0750010" 1.00
2226799 "815005" "0E0115005" 1.00
2226799 "140009" "0B0840009" 1.00
2226799 "721001" "0G0121001" 1.00
2226799 "730001" "0G0230001" 1.00


I need to put everything on one line where $1 matches like this.

e.g.2226697 29/01/08 2226697 "630511" "0P0705011" 1.00

any help would be grate. Thanks in advance

 

[feherke]

Hi

I would do it with [tt]awk[/tt]. One possible way :

awk -vORS= '{print l==$1?" ":"\n";l=$1}1;END{print""}' /input/file > /output/file

Tested with [tt]gawk[/tt].

Feherke.

http://rootshell.be/~feherke/

 

[HHG]

Hi thanks for this Feherke. I have put the following in a test script.

awk -vORS ='{print l==$1?" ":"\n";l=$1}1;END{print""}' inputpc.txt > pcompmw.txt

I am getting the following error.

# ./test.sh
awk: syntax error near line 1
awk: bailing out near line 1

Also, sorry didn't tell you I am running on solaris 9, unix, ksh.

 

[PHV]

solaris

nawk 'BEGIN{ORS=""}
{print (l==$1)?" ":"\n";l=$1;print}
END{print "\n"}
' inputpc.txt > pcompmw.txt

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

 

[feherke]

Hi

          [red]v-- the space should be here[/red]
awk -vORS=[highlight red] [/highlight]'{print l==$1?" ":"\n";l=$1}1;END{print""}' inputpc.txt > pcompmw.txt

Feherke.

http://rootshell.be/~feherke/

 

[HHG]

Thank you very much that worked a treat. I used PHV's concept and it worked a treat. Many thanks again to both Feherke and PHV for your kind help.

 

[Annihilannic]

FYI feherke, I don't think standard awks support expr ? trueval : falseval type logic.

Annihilannic.

 

[feherke]

Hi

As far as I know, the [tt]?:[/tt] conditional operator is a standard feature. There is a GNU extension indeed. But it only permits to wrap the expression for readability after the [tt]?[/tt] and [tt]:[/tt] characters :

[b]BEGIN[/b] {
  [b]print[/b] 1?
        2:
        3
}

While in standard [tt]awk[/tt] ( or [tt]gawk[/tt] with --posix switch ) you have to write it like :

[b]BEGIN[/b] {
  [b]print[/b] 1? [red]\[/red]
        2: [red]\[/red]
        3  
}

Please let me know if you ( or anyone else ) know about other difference in the [tt]?:[/tt] functionality. Thanks.

I am wondering why this extension... Maybe to permit easier commenting ?

[b]BEGIN[/b] {
  [b]print[/b] 1? [gray]# comment[/gray]
        2: [gray]# comment[/gray]
        3  [gray]# comment[/gray]
}

Feherke.

http://rootshell.be/~feherke/

 

[Annihilannic]

You're absolutely correct; I guess I just never tried it! Probably because I dislike the syntax; I don't find it very readable.

Thanks for the correction.

Annihilannic.