Perl Cheat Sheet

Perl is a highly capable, feature-rich programming language with over 25 years of development. These notes follow the path of the Learning Perl book.

• First line should be #!/usr/bin/perl or #!/usr/bin/env perl
• Perl compiles to byte-code before executing
• Do not use an extension for scripts.
• To specify a minimum version use “use 5.010;”
• Semicolons (;) required to separate statements
• “perl -w” turns on warnings for entire program. For just the current file use “use warnings;”
• “use diagnostics;” adds explanations to warnings
• “use strict;” requires variable declaration
• Comments start with # (no block comments)
• Parenthesis are optional unless part of syntax.
• @ARGV contains arguments, $0 contains program name.
• die/warn can be used to exit/warn, $! will contain any system error message. Without \n at the end, perl will append line number to error message.

• All numbers are stored in the same format e.g.-6.5e24, 037 (octal), 0x2a13b (hex), 0b11010 (binary)
• Underscores for readability are ignored e.g. 0x1377_0B77
• Arithmetic operators are * / + - % (modulus) ** (exponentiation)
• Comparison operators are < ⇐ == >= > !=

• To use unicode in source code - “use utf8;”
• Single-quoted literals '⅚∞☃☠' can contain any character except single-quote (use \') and backslash (use \\).
• Double-quoted strings can contain control-characters starting with a backslash and are variable interpolated e/g/ “${var}s $var\n”.
• . is for concatenation, x for string repetition.
• To insert a character by code - chr(0x05d0). To convert a character to its code - ord('א')
• String comparison operators are lt, le, eq, ge, gt, and ne

• Scalars are numbers or strings (or undefined or references), used interchangeably.
• $calar variables start with $ and a letter or underscore, but can contain letters (including unicode), numbers or underscores.
• Scalar assignment looks like $a = 3; or $a .= “suffix”; or $a **= 3;
• Recommended to use all lowercase variable names with underscores for word separation e.g. $var_name
• print takes a single scalar or a list of scalars (separated by commas)
• When a scalar is interpreted as a boolean, the following are false : 0, '0', '', undefined .
• !! is a handy shortcut to convert any scalar to 0 or 1 (to represent a boolean)
• Variables have value undef before being assigned. undef acts as a 0 or “” as needed, but will throw a warning if printed.
• defined() checks if a variable has been defined. Can also set a variable to undef.

• A list is an ordered set of elements (each of which is a scalar value), starting with position 0.
• @rrays are variables that store a list (separate namespace from scalar variables). @x refers to entire array x.
• Undefined arrays start out as (), the empty list and not undef.
• To access an element of an array - $myarr[0] - undef if never set.
• Index of last element is in $#arr
• Negative array indices wrap around (only once) so -1 refers to last element.
• Literal lists - ('abc','def') or (1..6) (integers 1 to 6) or qw(abc def) or qw#abc def# (quoted by whitespace).
• Can assign list values to variables e.g.

  ($fred, $barney, $dino) = ("flintstone", "rubble", undef);
  ($fred, $barney) = ($barney, $fred); # swap those values

• push adds element(s) to end of array. pop removes a single element and returns it.
• unshift adds element(s) to start of array. shift removes a single element and returns it.
• splice removes elements from middle of array, returns them and optionally replaces them

  splice @arr, start, len, @newelems

• sort and reverse functions return the modified list (can be saved to original array variable).
• Expressions parsed in either a scalar context or a list context. Scalars are promoted to single-element lists in list context.
• List functions may return different scalars - array variables return number of elements. The scalar function forces a scalar context e.g. for print function.

• A hash is a list indexed by a string (key) - a collection of key-value pairs. %hash has its own namespace.
• Uses scalable, efficient algorithms. Used to be called associative arrays.
• $hash{$some_key} to access a value (curly-braces instead of square braces).
• Assigning a hash to an array unwinds (flattens) it.
• To initialize a hash :

  %some_hash = ('a', 1, 'b', 2, 'c', 3);
  %some_hash = ( 'a' => 1, 'b'=>2, 'c'=>3,);

• When using a big arrow (a fat-comma) or when accessing a value, simple keys don't have to be quoted (barewords) e.g.

  %some_hash = ( a => 1}; $some_hash{a};

• %revhash = reverse %hash to reverse a hash (for non-unique values, last one wins).
• keys %hash; and values %hash; return a list of keys or value in same order (or # of keys/values in scalar context).
• %hash is true only if hash has at least one key-value pair.
• To iterate over hash :

  while ( ($key, $value) = each %hash ) {
    print "$key => $value\n"; }

  or in order of keys

  foreach $key (sort keys %hash) {
    print "$key => $hash{$key}\n"; } 

• %ENV hash holds environment variables.

• if-elsif-else

  if ($a == $b) {
    a = 0;
  } elsif ($a > $b) {
    a = 1;
  } else {
    a = -1;
  }

• while loop

  $count=0;
  while ($count < 10) {
      $count += 2;
  }

• foreach loop

  foreach my $rock (@rocks) {
    # modifications to $rock modify the list element
    # $_ is used if loop variable is omitted
  }

• Read a line of input - $line = <STDIN>;
• chomp removes a newline e.g. chomp($line=<STDIN>);
• Assigning <STDIN> to a list reads all input up till EOF e.g.

   chomp(@lines = <STDIN>);

• Loop over input ($_ can only be used in this specific case)

  while (<STDIN>) {
      print "I saw $_";
  }

• foreach can also be used but uses more memory since it reads all input into memory first.
• <> iterates over all files in @ARGV (or STDIN if no args) like e.g. cat/sed/awk.

  while (<>) {
      chomp;
      print LOGFILE "It was $_ that I saw!\n";
  }

• print takes list of items and sends all to STDOUT (unseparated). print @array; vs print “@array”;

  print <>;          # source code for 'cat'
  print sort <>;     # source code for 'sort'

• C-like printf function %g for number auto-format,%10s, %-10d etc.

  my @items = qw( wilma dino pebbles );
  my $format = "The items are:\n" . ("%10s\n" x @items);
  printf $format, @items;
  printf "The items are:\n".("%10s\n" x @items), @items;

• Filehandles can be barewords (upper-cased) or variables. Special filehandles are : STDIN, STDOUT, STDERR, DATA, ARGV, and ARGVOUT .

  open CONFIG, '<dino';  # < is optional
  open BEDROCK, '>fred' || die "Cannot open fred: $!";
  open LOG, '>>:encoding(UTF-8)','logfile'; # for perl >= 5.6
  open my $bedrock, '>:crlf', $file_name; # DOS-formatted output
  binmode STDOUT, ':encoding(UTF-8)';

• select can be used to change default output filehandle. $| = 1; unbuffers currently selected output.

• Subroutines are in their own namespace and are typically called using &mysub (sometimes the ampersand can be omitted).
• Subroutines can be defined anywhere e.g.

   sub mysub {
    my($m, $n) = @_
    $m + $n;
  } 

• The value of the last expression evaluated is returned. But return $a; can be used to return immediately.
• A list can be returned if the subroutine is called in a list context (wantarray can be used to detect list or scalar context).
• Arguments are in the list @_ . If &mysub; is used, the parent's argument list is inherited.
• Lexical scoped variables can be used in any block by prefixing with my.
• In Perl 5.10 and up state $x; can be used to declare a private variable that keep state between calls.
• Ampersand is optional if subroutine has previously been declared or if parenthesis are used. Ampersand is not optional if built-in subroutine is being overridden.