First character of variable

Shell Programming and Scripting
1

I am trying to detect 'nan' (not a number) string in my large txt file filled with lots of numbers. A file may have no or many 'nan'. It should be simple but getting errors. Any help appreciated. Here is what I have:

#!/bin/tcsh

set case1 = test0.txt
set chk = `grep 'nan' $case1`
set op = `$?`
if ($op == 0) then
     echo "no nan found"
else
    echo "nan found"
endif

Whatever is the file containing 'nan' or not, it always outputs "nan found"

Thank you.

2

grep command returns zero when it match the pattern and non zero when it does not match the pattern:

EXIT STATUS

The following exit values are returned:
0         One or more lines were selected.
1         No lines were selected.
>1        An error occurred.
3

try something like:

#!/bin/tcsh
 set case1 = test0.txt
set nan=1
grep -q "nan" $case1 || set nan=0
if ( $nan == 0 ) then
     echo "no nan found"
else
    echo "nan found"
endif
4

Why not use 'regular' return codes?

#!/bin/tcsh
set case1 = test0.txt
set search = nan

# Search value & set retval
grep -q "$search" "$case1"
set RET = $?

# Print result
if ( 0 == $RET ) then	
	echo "$RET - $search found"
else
	echo "$RET - $search not found"
endif

hth

EDIT:
By regular return value i mean:

EXPRESION || set nan=0

This will change the variable to zero (success) on fail, which is quite irritating when someone later has/will edit the script.

grep -q "$search" "$case1"
set RET = $?

This will set the return code always to what it was.

Also this way you could catch the error of a non existing file..
Example:

 ~ $ tcsh
grep: test0.txt: No such file or directory
2 - nan not found
2 ~ $ touch test0.txt
:) ~ $ tcsh
1 - nan not found
 ~ $ echo nan > test0.txt 
+ ~ $ tcsh
0 - nan found
 ~ $
5

Thank you all of the replies. Very helpful. Ended up using rdrtx1's solution.

Thank you.