计算文件中的指定单词

Question

I want to make script, that finds how many words, that I chose, are in my chosen file.

Here is my script:

#!/bin/bash
echo "Iveskite reiksme kurios kurios kurios ieskosime faile"
read reiksme
echo "Nurodykite faila kuriame ieskosime reiksmes"
read failas
kiek=$(awk '/$reiksme/' $failas | wc -l)
echo $kiek > kiek.txt

if $failas contains lines like:

$reiksme
$reiksme
$reiksme

then it's ok, but if $failas contains lines like:

$reiksme $reiksme
random word $reiksme $reiksme
random random $reiksme

Then my script fails to count how many words.

Answer 1

awk is only splitting on lines. You could do it in a few steps, though there may well be easier ways than this:

kiek=$(sed -e 's/\s/\n/g' $failas | grep "\b${reiksme}\b" | wc -l)

to replace all whitespace with newline, then use grep (or your awk if you like) to find what you're looking for then count them

Answer 2

I found the answer on my own. I just changed

kiek=$(awk '/$reiksme/' $failas | wc -l)

to

kiek=$(tr -cs 'A-Za-z' '\n' < $failas | grep -c "$reiksme")

Answer 3

The best way to do this with grep is to use the -o option as described in this post . Putting it together for you:

kiek=$(grep -o "${reiksme}" | wc -l)