Shell Programming and Scripting

Code:

# Print list of word frequencies
{
$0 = tolower($0)
    for (i = 1; i <= NF; i++)
	   
		counter[$i]++	
}

END {
    
    for (word in counter)
        printf "%s\t%d\n",word, counter[word]		
}

I have this simple awk code from awk user guide to count the frequency of word. Now consider the following input to script:
hello HELLO Hello World

The output is:
hello 3
world 1

What if I want it to maintain the case of other words. i.e I want the output as
hello 3
World 1

(Note the case of World is retained). How do I do it. Please help

You could remove the tolower($0) but that would mess up your "hello"s...

Code:

{     for (i = 1; i <= NF; i++)                 
         counter[($i~"^[Ww]orld$"):$i:tolower($i)]++     
}

Hey Vgersh99,

Thank you for responding. How do I make it generic. I mean, the input is not always going to contain World.

The program will be run as follows:

gawk -f test.awk <input.txt

where, input.txt file can contain any text. Thanks

you'll need to know what words you'd like to keep in the original case. This can be made configurable and not be hard-wired to the code itself.

Thank you for your patience. I need to maintain the case all the words other than the repetitive ones (with different cases), in which case it would take it as lower case.

For eg:
make Make MAKE the World A better place

Output:
make 3
the 1
World 1
A 1
better 1
place 1

You see, it uses only lowercase for same word and different cases (make in this case) and maintains the case of other words. When I use tolower to ignore the case while keeping the count, it keeps the count properly but unfortunately also makes other words lower case

Code:

nawk 'BEGIN{ARGV[ARGC++]=ARGV[1]} FNR==NR{for(i=1;i<=NF;i++) a[tolower($i)]++;next} {for(i=1;i<=NF;i++) (a[tolower($i)]>1)?c[tolower($i)]++:c[$i]++}END{for(i in c) print i,c[i]}' myFile