在shell脚本中打破循环

Question

The user inputs sentences for which the number of words has to be counted. If the number is greater than three the sentence should be considered Long . If three or fewer then Short .

I know how to do that, but the rest of the task says make a loop so the user can input sentences until he inputs "n". The problem is I when I input "n" it counts it, but that "n" shouldn't be counted. (It should be a break statement.) How can I break that loop before counting the letter "n"?

#!/bin/bash
recenica=0
while [ $recenica != "n" ]
do
    echo Unesite recenicu:
    read recenica
    if [ $recenica = "n" ]; then # ->>>>> this doesn't work
        break
    fi  
    echo $recenica > datoteka
    br=$(wc -l datoteka | cut -c1-2)
    echo recenica ima $br rijeci
    if [ $br -gt 3 ] 
    then
        echo $recenica > Duge.Recenice.IB
    elif [ $br -le 3 ] 
    then
        echo $recenica > Kratke.Recenice.IB
    else
        echo Molimo unijeti nešto
    fi  
done

Answer 1

You need to use quotes around symbols that expand to more than one word (contain spaces)... this works:

#!/bin/bash
recenica=0
while [ "$recenica" != "n" ] # koristiti dvostruke navodnike ovdje
do
    echo Unesite recenicu:
    read recenica
    if [ "$recenica" = "n" ]; then # koristiti dvostruke navodnike ovdje
        break
    fi  
    echo $recenica > datoteka
    br=$(wc -l datoteka | cut -c1-2)
    echo recenica ima $br rijeci
    if [ $br -gt 3 ] 
    then
        echo $recenica > Duge.Recenice.IB
    elif [ $br -le 3 ] 
    then
        echo $recenica > Kratke.Recenice.IB
    else
        echo Molimo unijeti nešto
    fi  
done

Answer 2

The problem with read is that it will read the entire line and place the contents in recenica . With one word this is fine for the test:

while [ $recenica != "n" ]

However, if you have read more than one word, then the test will fail due to [: too many arguments (as noted by amdn in his answer regarding quoting)

When you enter n you care only about the first word, so limit your test accordingly using parameter expansion/substring extraction :

while [ ${recenica%% *} != "n" ]; do

That allows you to let the while loop control the break logic reducing the entire structure of your code to:

#!/bin/bash
recenica=0
while [ ${recenica%% *} != "n" ]
do
    echo Unesite recenicu:
    read recenica

    echo $recenica > datoteka
    br=$(wc -l datoteka | cut -c1-2)
    echo recenica ima $br rijeci

    if [ $br -gt 3 ] 
    then
        echo $recenica > Duge.Recenice.IB
    elif [ $br -le 3 ] 
    then
        echo $recenica > Kratke.Recenice.IB
    else
        echo Molimo unijeti nešto
    fi  
done

There are a number of ways to do this, but this approach is fine.