8

I am following a series of tutorials to learn Bash shell script. One of the exercises is to loop through files in the current directory and search for a pattern in those files. If the pattern is found then the script should sum the file size of those files.

#!/bin/sh
patern=echo
totalSize=0

for file in * 
do
    [ ! -f $file ] && continue  

    if grep $patern $file > /dev/null
    then    
        echo "pattern matched in $file" 
        echo "file size is `stat -c%s $file`"
        fileSize=`stat -c%s $file`

        totalSize=`expr $totalSize + $fileSize`
        echo "size so far is $totalSize bytes"
        echo                                                    
    fi
done

I have one other folder in the directory from which I am running the script. The folder is called "somedir". It is empty. I have took a snippet of output text pasted below. The middle 3 lines show when the loop iterates over the directory "somedir" which it prints as "sum_files"- my script name along with a file size.

I don't understand this behaviour. How can an empty directory have a file size?
But my main concern is as to why the continue keyword is not stopping the loop iteration. The output is showing that the script is running the below if statement containing the grep command even though it should stop if a directory is found.

I have verified that the test command[ ! -f $file ] does in fact return 0 when the loop gets to the directory and thus the && continue should be invoked. So why does the program continue with the rest of the loop code and try to grep the directory rather than just skipping the loop iteration at continue as expected? I know this is rather trivial but would like to know what's going on here.

Output text

pattern matched in retirement
file size is 396
size so far is 6385 bytes

pattern matched in sum_files
file size is 398
size so far is 6783 bytes

pattern matched in tp0
file size is 164
size so far is 6947 bytes

Improve this question
7
  • Use shellcheck.net to identify a few other errors in your script. Commented Sep 20, 2015 at 14:11
  • Your question doesn't make sense… this doesn't make sense: The middle 3 lines show when the loop iterates over the directory "somedir" which it prints as "sum_files". Try set -x to see what happens. Commented Sep 20, 2015 at 14:29
  • @gniourf_gniourf "pattern matched in sum_files---file size is 398---size so far is 6783 bytes". From the terminal output. Each file outputs 3 lines of text showing the file name if a pattern match was found, its file size and the summed total. Here I am displaying the anomaly where the for loop iteration runs over the folder "somedir", which shouldn't happen. Where should I "set -x" in the code? Commented Sep 20, 2015 at 14:33
  • The loop runs over the file sum_file. Put set -x just after #!/bin/sh. Commented Sep 20, 2015 at 14:41
  • Actually @gniourf_gniourf you are right because I am running the script in my bin file where the script file is so of course this doesn't make any sense because the script is finding itself. LOL. Commented Sep 20, 2015 at 14:50

2 Answers 2

Reset to default
10

continue continues the loop, as if it reached its bottom.

To break out of the loop use break.

More on bash scripting here: http://tldp.org/LDP/abs/html/index.html

Improve this answer
Sign up to request clarification or add additional context in comments.

9 Comments

I understand from the tutorial I am following and tldp.org/LDP/abs/html/loopcontrol.html#EX28 from this link that: Continue will stop the current iteration of the loop, where break will stop the whole for loop as if it reached the end of all files. When I subsitute break for continue the script will stop searching any other files once it finds a directory which is not the behaviour I want
FYI, a return code of 0 means success, so the rhs of the && will be executed. Shell is not C.
@gniourf_gniourf: This is not about return codes but about the boolean expression to which [ ! -f $file ] evaluates, that is true of false. (tldp.org/LDP/abs/html/fto.html)
return code of 0 means true (or, rather, success). Try it: true; echo $? and false; echo $?. Once again, shell is not C.
Yes this is what I understand. The output is as expected now. I verified the [ ! -f $file ] by running echo $? after it and got 0 when the directory iterated in the loop. Sorry for the confusion. I could not see the wood for the trees!
|
2

As to your question about how a empty directory can have a size: directories technically just files themselves, and in order to establish themselves as directories they need some data to let the file system know that.

enter image description here

Improve this answer

1 Comment

You shouldn't rely on this though. Some filesystems return 0. Ceph-fuse, for example, reports the total size of all containing files as the folder's size.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.