UNIX for Dummies Questions & Answers

Hi everyone,

i have files containing lines of number:

109
107
67
62
..
..

i want to calculate the difference between numbers in lines 1 and 2, 3 and 4, 5 and 6 and so on. would someone help me please?. Thanks

Try this
awk '{s=$0;getline;print s-$0;next}' file

thanks rakeshawasthi,

it works, however i need the result to be in absolute value (always positive value). would you help me? thanks

awk '{s=$0;getline;r=s-$0;if (r<0)r=-r;print r;next}' file

thanks, it solves my problem perfectly.
i have one more question, how can i display only lines which the third field is different from the third field of previous line. for ex:

a/fruit/mango
a/fruit/mango
a/fruit/apple
a/fruit/apple
a/fruit/jackfruit

i need the result to be:
a/fruit/mango
a/fruit/apple
a/fruit/jackfruit

thanks, i really appreciate your help.

Well thats easy.. if everything is going to be "a/fruit/fruitname" and you only want to do consecutive lines, then use the command "uniq". If you want to remove duplicate lines that are not consecutive then you can use "sort -u". Example(s):

Starting data: (added a few rows to help show how it works).

Code:

-bash-3.00$ cat test.txt
a/fruit/mango
a/fruit/mango
a/fruit/apple
a/fruit/apple
a/fruit/jackfruit
a/fruit/mango
a/fruit/apple
a/fruit/mango
a/fruit/mango

Using "uniq":

Code:

-bash-3.00$ uniq test.txt
a/fruit/mango
a/fruit/apple
a/fruit/jackfruit
a/fruit/mango
a/fruit/apple
a/fruit/mango

Using "sort -u"

Code:

-bash-3.00$ sort -u test.txt
a/fruit/apple
a/fruit/jackfruit
a/fruit/mango

Thanks,
the complete example above is like this:
a/fruit/mango/1
a/fruit/mango/6
a/fruit/apple/3
a/fruit/apple/7
a/fruit/jackfruit/4
a/fruit/jackfruit/8

what i want is remove the previous line when its third field contains the same value as the third field of current line. so the result will be like this:
a/fruit/mango/6
a/fruit/apple/7
a/fruit/jackfruit/8

would you or someone help me please?