Tips for Shell

There will be some case-based tips for Shell.

1. Replace the sequence space with newline

Input example

a b  c   d

Output example

a
b
c
d

Summary

sed is the best solution.

Solution

sed -r -e ‘s/[[:space:]]/\n/g‘

2. Sort the unordered word list (with duplication) by frequency

Summary

It is very useful when counting word frequency.

Solution

sort | uniq -c | sort -rn

3. Awk print line without first column

Summary

This command actually clear the first column from the file.

Solution

awk ‘{$1=""; print $0}‘

4. Show the longest line in the file

Summary

Taking advantage of the const length in awk

Solution

awk ‘{print $0" "length}‘ | sort -k2 -nr | head -1

5. Ignore line that match some pattern

Summary and Solution

Using

grep -v

6. Batchly replace namespace

Summary

Using for to loop over the file and sed to perform the replacement.

Solution

for f in `find ./src -type f`; do sed -i ‘s/oldnamespace/newnamespace‘ $f; done

7. Generate 02, …, 24 sequence

Summary

Perl can be a better solution.

Solution

perl -e ‘for{$i=2;$i<=24;$i++}{printf("%02d\n", $i)}‘

8. Command line calculator

Awk Solution

awk ‘{print 1/10}‘

and additional enter is needed

Perl Solution

More flexiable (with formatting).

perl -e ‘printf("%f\n", 1./10);‘