Shell Programming and Scripting

Hi All,

i have to check a log file for the patterns like error, msg, fail or warn...
for this i can do, like

here log file is the input file to the script

egrep -i "error | fail | warn | unable" logfile.log
if [ $? -eq 0 ]
then
echo " patteren found "
else
echo " pattern not found "

but, what i want here is i want to know which are all the patterns found and which are all the patterns not found.

so that i can print the patterns which are all found an not found seperately for a precise manner.

my expected output is:

error, warn found in the log file
fail, unable not found in the log file

Do each of your four greps separately.
Use similar logic on return values, just instead of saying fount/not, echo what you just searched for to a variable/file for found/not.
After doing all four, that variable/file will hold the paramaters that met the condition.

Thanks Joeyg,
i want to accomplish with less LOC.

could you please guide me, to make this effective.
Thanks.

A solution with awk:

Code:

#!bin/sh

awk '
{$0=tolower($0)}
/error/{m[1]++}
/fail/{m[2]++}
/warn/{m[3]++}
/unable/{m[4]++}
END{
printf("%s\n",m[1]?"Error found":"Error not found")
printf("%s\n",m[2]?"Fail found":"Fail not found")
printf("%s\n",m[3]?"Warn found":"Warn not found")
printf("%s\n",m[4]?"Unable found":"Unable  not found")
}' logfile

Thanks Franklin,

could you please tell me how my input file is used in this code and why we
are using m[1]++,m2,etc..

i am novice user.
your help helps me a lot.

Thanks.

An explanation of the code:

Code:

{$0=tolower($0)}

To ignore case distinctions, we convert the line to lowercase to match the patterns in lowercase.

Code:

/error/{m[1]++}
/fail/{m[2]++}
/warn/{m[3]++}
/unable/{m[4]++}

For every pattern we use an array m, m[1] for errors, m[2] for fails and so forth.
If a pattern matches we increase the array with 1.

Code:

END{
printf("%s\n",m[1]?"Error found":"Error not found")
printf("%s\n",m[2]?"Fail found":"Fail not found")
printf("%s\n",m[3]?"Warn found":"Warn not found")
printf("%s\n",m[4]?"Unable found":"Unable not found")

After we read all the records we use a printf command with a conditional expression to print the results. If the value of one of the element of the array is true ( > 0 ) the first string ("Error found") is printed otherwise we print the second string ("Error not found").


Regards

function check_log
{
awk '{$0=tolower($0)}
/error/{err++}
/fail/{fail++}
/msg/{msg++}
/unable/{unable++}
/unsuccess/{uns++}
END{ print("%s\n",err?"Error found":"Error not found")
print("%s\n",fail?"Fail found":"Fail not found")
print("%s\n",msg?"Msg found":"Msg not found")
print("%s\n",unable?"Unable found":"Unable not found")
print("%s\n",uns?"Unsuccess found":"Unsuccess not found")
}' logfile

i created like this but i got the error for the lines in bold