Shell Programming and Scripting

Hi All,

I tried this but I am having trouble formulating this:
I have a file that looks like this (this is a sample file words can be different):

Code:

network
router
frame
network
router
computer
card
host
computer
card

One can see that in this file "network" and "router" occur together two times and also "computer" and "card" two times. I want to find the count of those words in the file that occur together in the same file taking two words at a time. I expect my output to be like this:

Code:

1
1
1
2
1
1
1
1
2

The output above means that "network" and "router" have occurred 1 time in the first occurrence. Then "router" and "frame" occur 1 time. Then "frame" and
"network" occurs 1 time. Then again we encounter "network", "router" and now this count becomes 2. And we keep on doing this for the rest of the file.

This is what I tried, but the problem is I am supplying the words manually. Moreover, I have more than one file all with .dat extension. One can see that I am reading .dat files and storing the result in .txt files.:

Code:

ls -1 *.dat | while read page
do
cat $page | grep "$network[[:blank:]]*$router" $page>$page.txt
done

I am using BASH in Linux.

Something like this?

Code:

ls *.dat | while read page
do
  awk '{print ++a[$0]}' $page > $page.txt
done

Thanks. But what I am taking into account are two words at a time rather than one. That's why in the original file you can see 10 words and in the result file 9 numbers. So, this is how things should work actually:
1. Take "network" and "router" together. But now count of "network" and "router" is 1.
2. Go to "router" and "frame" again count is one as previous two words were "network" and "router". Hence, don't match with this current two words.
3. Take "frame" and "network", again count is 1
4. Take "network" and "router" but now count of "network" and "router" is 2 as I have already seen "network" and "router" before.
5. In this way till the file ends.

Code:

ls *.dat | while read page
do
  awk '!p{p=$0; next} {if(a[$0]==p) {s="2"} else {a[$0]=p; s="1"} p=$0; print s}' $page > $page.txt
done