Perl Oneliners

These are Perl implementations of the examples used in Awesome awk oneliners. See perl1line.txt for a very educational set of Perl code in this vein.

# Replace multiple spaces with single space

command | awk '{ gsub(/[ ]+/," "); print }'  # awk
command | perl -pE 's/\s+/ /g'               # perl

# Explanation

• -p tells perl to process & print each line
• -E tells perl to run a line of code
• s/PATTERN/REPLACEMENT/options searches and replaces text.
  • /\s+/ matches one or more whitespace characters
  • / / is a single space to replace the match
  • g is an option to find all occurences

# Trim leading and trailing spaces

command | awk '{ gsub(/^[ \t]+|[ \t]+$/, ""); print }'  # awk
command | perl -pE 's/^\s+|\s+$//g'                     # perl

# Explanation

• -p tells perl to process & print each line
• -E tells perl to run a line of code
• s/PATTERN/REPLACEMENT/options searches and replaces text.
  • /^\s+|\s+$/ includes two patterns separated by a | and matches the first OR second pattern:
    • ^\s+ matches one or more whitespace characters at the beginning of a line
    • \s+$ matches one or more whitespace characters at the end of a line
  • // is an empty replacement, i.e. remove any matches from the line
  • g is an option to find all occurences

# Replace string with another one only in lines containing specific string

command | awk '/,/{gsub(/ /,""); print}'   # awk
command | perl -pE 'if (m/,/) { s/ //g }'  # perl
command | perl -pE 'if (/,/) { s/ //g }'   # perl - remove optional m
command | perl -pE 's/ //g if (/,/)'       # perl - even more concise

# Explanation

• -p tells perl to process & print each line
• -E tells perl to run a line of code
• s/PATTERN/REPLACEMENT/options searches and replaces text.
  • / / matches a single space
  • // is an empty replacement, i.e. remove any matches from the line
• m/PATTERN/options searchs a string for a pattern, returning true it succeeds
  • /,/ matches a comma
  • m is optional if the delimiter is /

# Print lines where column equals values based on conditions

command | awk -F, '(tolower($6)~"rhel" || tolower($6)~"linux"){print $0}'  # awk
command | awk -F, '(tolower($6)~"rhel|linux"){print $0}'  # awk
command | perl -F, -naE 'print if ($F[5] =~ /rhel|linux/i)'  # perl

# Explanation

• -F specifies "," as the column separator to split the line
• -n tells perl to process each line
• -a turn on autosplit mode, splitting fields to @F
• -E tells perl to run a line of code
• ($F[5] =~ m//) compares the sixth (6th) field to the regexp
• /rhel|/linux/i matches "rhel" or "linux"
  • i option specifies a case-insenstive patten match

# Example

$ { cat <<EOF
1,2,3,4,5,6,7
1,2,3,4,5,rhel,6,7
1,2,3,4,5,Linux,6,7
EOF
} | perl -F, -naE 'print if ($F[5] =~ /rhel|linux/i)'
1,2,3,4,5,rhel,6,7
1,2,3,4,5,Linux,6,7

# Print specific column numbers

command | awk -F, '{print $1,$2,$7}'               # awk
command | perl -F, -naE 'print "$F[0],$F[1],$F[6]\;n' # perl

# Explanation

$F[0], $F[1], $F[6] specifies the 1st, 2nd, and 7th column of the line

# Replace first occurrence of string

command | awk '{sub(/,/,"1,")}1'  # awk
command | perl -pE 's/,/1,/'      # perl

# Example

1$ echo "foo,bar,baz" | perl -pE 's/,/1,/'
2foo1,bar,baz

# Replace last occurrence of string

command | rev | awk '{sub(/,/,"1,")}1' | rev # awk
command | rev | perl -pE 's/,/1,/' | rev     # perl
command | perl -pE 's/(.*),/${1},new-text/'         # perl - no other commands required

# Explanation

The final example does more with regular expressions.

• (.*) captures everything matched by the expression in the parenthesis (.* is any string of characters) into a variable that can be used in the replacement operation.
• /(.*),/ matches the longest possible string followed by a , capturing the string as $1.
• /$1,1 replaces the match with the string, a ,, and new-text.

# Example

1$ echo "foo,bar,baz" | rev | awk '{sub(/,/,"1,")}1' | rev 
2foo,bar,1baz
3$ echo "foo,bar,baz" | perl -pE 's/(.*),/$1,new-text/'
4foo,bar,new-textbaz