Shell Programming and Scripting

Hi,

I am currently using grep -c to scan lines for certain data. I am just wondering if you can search a specific column of a file using grep -c.

Thanks

Can you post some sample data from yourfile?

Using combination of awk and grep i think you should be able do achieve it.

Thanks

Files go on this structure

83.67.99.63 Wed Apr 12 01:50:11 2006


Basically if the user enters the data "Wed" "Apr" and "12" i want it to count all the files with those lines in them. Normally i would use the "grep -c". Though if i dont select what columns, the day they enter, for example "12" would also match up to a line which has a time of 12:12:10, butt doesnt have the day of "12". So when they enter the day number I need to know how to search that single column, as its repeating data.

Also, would this be possible without AWK?

Thanks for any help

Are you saying

Code:

grep -c "Wed Apr 12"

does not give you what you need ?

Well im putting them into variables, such as when they enter day, then month followed by day number i do a grep of;

grep -c "$day $month $daynum" ....

I thought that as the daynum contains a figure such as 12, it may match up with a line which doesnt have a daynumber of that value, but a time involving that number.

Also when i put the $year variable in the grep code after $daynum, e.g. "2006" it breaks the count and none of the hits appear anymore...

yes,

grep -c "$day $month $daynum" would work absolutly fine as....

"$day $month $daynum" will be a complete pattern to match

if you want to include year as well the you can try...

cat inputfile | grep "$day $month $daynum" | grep -c "$year"

cheers,

Hi,

I tried putting:

Code:

cat ./logs/*.logs | grep "$month $day" | grep -c "$year"

But it just gives me a single numerical output, not a list of the files with a number of matches beside them.

Any ideas whats going wrong?

Thanks for the help btw