Mdthbs

I have taken 2 variables in the grep command for search in the file. User is only putting one and pressing enter on the second variable, its throwing the error.

I want to set empty value on the variable second if user doesn't enter anything for the second one.

#!/bin/bash 

find /var/log/nginx/ -type f ( -iname "access.log*" ! -iname "access.log*.gz" ) -exec cat {} + >> output.txt 

echo enter the file name read file1 

echo "enter varible 1" 

read var1 

echo "enter varible 2" 

read var2 

grep -w $var1 output.txt | grep $var2 | cut -d' ' -f1,4,12,25 > $file1 

grep -w $var1 output.txt | grep $var2 |sed 's#.*(aff_id=[[:digit:]]+).*(off_id=[[:digit:]]‌​+).*(trans_‌​id=[[:alnum:]]+)‌​;.*#1 2 3#' > out2.txt 

paste -d' ' $file1 out2.txt > out3.txt 

rm -rf output.txt 

rm -rf $file1

It seems like what you might be looking for is the ability to test for whether or not a variable was assigned a value. This can be done using Bash string-comparison operators. These operators are documented in several different common references, e.g.:

• Bash Guide for Beginners: Chapter 7 - Conditional Statements




• Advanced Bash-Scripting Guide: Section 7.3 - Other Comparison Operators

In particular, you can use the if [ -n "${string}" ] to test for non-null strings. Adapting your script to use this test might result in something like the following:

#!/bin/bash

find /var/log/nginx/ -type f ( -iname "access.log*" ! -iname "access.log*.gz" ) -exec cat {} + >> output.txt

echo enter the file name read file1

echo "enter varible 1"

read var1

echo "enter varible 2"

read var2



if [[ -n "${var2}" ]]; then var2=""; fi



grep -w $var1 output.txt | grep $var2 | cut -d' ' -f1,4,12,25 > $file1

grep -w $var1 output.txt | grep $var2 | sed 's#.*(aff_id=[[:digit:]]+).*(off_id=[[:digit:]]‌​+).*(trans_‌​id=[[:alnum:]]+)‌​;.*#1 2 3#' > out2.txt



paste -d' ' $file1 out2.txt > out3.txt

rm -rf output.txt

rm -rf $file1

That said, I don't think that will help solve your problem. To avoid your grep errors it should suffice to quote the variables, e.g. grep -w "${var1}" output.txt. Here is what your script looks like with quoted variables:

#!/bin/bash

find /var/log/nginx/ -type f ( -iname "access.log*" ! -iname "access.log*.gz" ) -exec cat {} + >> output.txt

echo enter the file name read file1

echo "enter varible 1"

read var1

echo "enter varible 2"

read var2



if [[ -n "${var2}" ]];

    var2=""

fi



grep -w "${var1}" output.txt | grep "${var2}" | cut -d' ' -f1,4,12,25 > $file1

grep -w "${var1}" output.txt | grep "${var2}" | sed 's#.*(aff_id=[[:digit:]]+).*(off_id=[[:digit:]]‌​+).*(trans_‌​id=[[:alnum:]]+)‌​;.*#1 2 3#' > out2.txt



paste -d' ' $file1 out2.txt > out3.txt

rm -rf output.txt

rm -rf $file1

An explanation for why you need to quote the variables in your grep expressions can be found in the Bash Guide for Beginners: Section 3.4.8 - Word Splitting:

The shell scans the results of parameter expansion, command substitution, and arithmetic expansion that did not occur within double quotes for word splitting. [...] Explicit null arguments ("""" or "''") are retained. Unquoted implicit null arguments, resulting from the expansion of parameters that have no values, are removed. If a parameter with no value is expanded within double quotes, a null argument results and is retained.