UNIX for Dummies Questions & Answers

Hi,

I have several files that look like this:
File1.txt
File2.txt
File3.txt
File4.txt
What I need is a piece of code (bash/awk) that will check and compare the data of each file and will return each file only with the UNIQ data. For example, for the above data the expected output will be:
File1.txt
File2.txt
File3.txt
File4.txt
If there are multiple UNIQ data in a file, it will only give the first UNIQ data. As 'File4.txt' has two UNIQ data: 'Data12 and Data30', but it'll give the first one as :
File4.txt
Thanks in advance for your help.

Code:

cat File*.txt | sort | uniq -u | xargs -i grep {} File*.txt

Thanks for your reply.
However, the code is generating the following error message:
cat File*.txt | sort | uniq -u | xargs -i grep {} File*.txt
xargs: illegal option -- i

Check the correct syntax

Code:

man xargs

maybe you have the -I alternate option available to achieve the same kind of work.
maybe something like

Code:

cat File*.txt | sort | uniq -u | xargs -I{} grep {} File*.txt

or

Code:

cat File*.txt | sort | uniq -u | xargs -I {} grep {} File*.txt

or

Code:

cat File*.txt | sort | uniq -u | xargs -I "{}" grep {} File*.txt

Thanks. It's nice and simple. I get the following output:
File2.txtata10
File4.txtata12
File3.txtata17
File1.txtata20
File4.txtata30

Just two small issues:
1) I need the output in files rather than in the terminal. In this case there will be five output files with their corresponding data.
2) File4.txt has two uniq data and hence it is present two times in the terminal. As mentioned in the second part of my query, If one file has more than one uniq data, only the first uniq data will be written. For example:
File4.txt
Data12

Cheers.

not sur about the grep -f <( ) syntax (may need some syntax fix)
but maybe something like

Code:

for i in File*.txt; do grep -f <(cat File*.txt | sort | uniq -u | sed 's/.*/\^&\$/') $i | head -1 >$i.out; done

or maybe something like (i didn't test it so it may require some syntax adjustment)

Code:

cat File*.txt | sort | uniq -u | sed 's/.*/\^&\$/' | xargs -i grep {} File*.txt | while IFS=: read a b; do [[ ! -f $a.out ]] && echo "$b" >"$a".out; done

the sed part has been added to ensure that grep Data2 would grep ^Data2$ so if there are also occurrence of Data22 or Data23 they won't be displayed

Thanks a lot.