compare and select unmatched lines

[rkdash]

Hi all,

I have two files for which the first column of file1 contains all the first column entries of file2 and some more. I want to write another file from file1 whose first column entries are not present in the first column of file2.

For example:

file1:
--------
9375 63.511 76.900 0.004 3.960 0.080 1
9376 63.522 76.906 0.004 3.962 0.080 1
9377 63.514 76.936 0.004 3.966 0.080 1
9378 63.537 76.924 0.004 3.967 0.080 1
9379 61.967 78.543 0.004 4.109 0.080 1
9380 61.951 78.542 0.004 4.108 0.080 1


file2:
-------
9379 61.967 78.543 0.004 -0.217 0.080 1
9380 61.951 78.542 0.004 -0.218 0.080 1


Desired file3:
--------------
9375 63.511 76.900 0.004 3.960 0.080 1
9376 63.522 76.906 0.004 3.962 0.080 1
9377 63.514 76.936 0.004 3.966 0.080 1
9378 63.537 76.924 0.004 3.967 0.080 1


Any help is appreciated.

rk

 

[PHV]

A starting point (typed, untested):
awk '{NR==FNR{++a[$1];next}!a[$1]}' file2 file1 > file3

Hope This Helps, PH.
Want to get great answers to your Tek-Tips questions? Have a look at FAQ219-2884 or FAQ181-2886

 

[rkdash]

Sorry PHV, it doesn't work. Could you please check the syntax again ?

Thanks

rk

 

[feherke]

Hi

Just remove the outer braces ( {} ) and works fine :

awk 'NR==FNR{++a[$1];next}!a[$1]' file2 file1 > file3

Feherke.

http://rootshell.be/~feherke/

 

[taupirho]

You might also consider using the join command if you have it - something like:

join file1 file2 -v2

 

[rkdash]

Thanks feherke, it works.

taupirho, i had tried with join commnand (join -a1 file1 file2) before. It works for small files, but fails with the number of lines is large.

Is there any reason for that ?

rk

 

[Annihilannic]

join requires the files to be sorted on the joined fields, that's most likely the cause.

Annihilannic.