Shell Programming and Scripting

Dear Friends,

After a long night sleep and all your precious help I have found where was the problem.

I found blank space on the two_arch.log. That's why the command line return no line.

Then all works great. I wouldn't have succeed without you.

Dear vgersh99, you were right. The output screen is exactly what I wanted and it is exactly what I've got. Thanks for your patience and help. I will remember to use BB code tags in the future. BB code tag is the # on the top of the post message box ?

Code:

Like this ?

I hope I will be able to return all the help you gave me.

Just for info, do you mind tell me more things about : What all of this means ?

Code:

nawk -F/ 'FNR == NR { a[$0]++; next } a[$NF]'

PS : It is hard to be newbee, even if you like writing script.

So the idea is that we want to first build an (associative) array, in which each index is a line / filename in fileA. The line / filename is represented by $0, and the array is a[]; merely by setting a value to a[index] do we create a[index]; a[index]++ is an easy way of doing this. Then, for fileB, we want to see if the last part of the line, which represents the filename and is represented here by $NF, exists as an index in this associative array. If it does not, a[index] evaluates to 0; when this expression is 0 (does not exist), no further action is taken for that line; otherwise, when the expression is non-zero (does exist) the following implicit action is taken:

Code:

{ print $0; }

Earlier I said that $NF represents the last part of the path in fileB. This is because NF represents the number of fields in the current line, and $NF is the n-th field in the current line (which contains n fields).

Also, how do we distinguish between fileA and fileB? Easy: if the global record number (NR) is equal to the current file record number (FNR), then it must be the first file; the second file's FNR will start at 1, but by then, NR will be 1 plus the number of lines in the first file.