The London Perl and Raku Workshop takes place on 26th Oct 2024. If your company depends on Perl, please consider sponsoring and/or attending.

LOGO

                       ⢀⣀⡤ ⢀⣤⣿⡗ ⣀⣀⣀
           ⢀⣤⣤⣤⣄⡀    ⣠⡶⠿⠛⣹⡾⠛⢁⡼⠟⢛⠉⠉⠉⣉⣣⣀⣀⣀⣀⣀⣀⣀⣀⣀⣀⡄
       ⢉⠻⣯⣉⡛⠒⠻⡷⢮⡙⠳⣤⡐⣾⠟⣀⣴⠋⠁⣀⡴⠋ ⣠⡟ ⠐⠚⠉⠉⠉⠉⠉⠉⠉⠉⠉⠉⢩⠛
       ⠘⣧ ⠹⣿⡳⡀⠙⢦⡈⠳⠈⢱⡟ ⠋⣼⣿⣿⢿⠁⠰⣶⠏⢐⡆⢠  ⣠⣖⣢⠤⠶⠶⠂ ⡽⢃  ⣀
        ⠈⢗⣲⠞⠓⠛⢦⡌⡿  ⡾⠃  ⣿⣿⡾   ⣿ ⣼⣠⠏⢀⡾⣿⠟⣂⣠⡤⠤⠴⠶⠛⠛⠛⢋⡿
    ⢀⡴⡲⠹⠍⠁ ⠐⢶⡂⠈⣓⠱⣆⡼⠃  ⢰⣿⡟⢳ ⢀⣾⢇⡜⠋⠁⣰⣯⠾⠷⠚⠉ ⢀⣴⠎    ⢸⡇
    ⠘⠙⠳⠤⣕ ⠳⣄ ⠉⠓⢴⣱⣿⡅⣀⣤⠾⣟⣯⣤⣶⡶⢿⣿⣯⠆ ⢈⣽⠃⣀⣀⣠⣴⣾⣯⠄     ⣴⠇
       ⢀⣹⣶⡀⢈⣷⣶⣤⣼⣿⡿⢗⡋⣩⣶⡟⣛⣿⣿⣷⣾⣛⣉⣀⡤⠾⠛⠒⠋⠉⠛⣿⡿⠋     ⢠⡏
      ⠙⠛⣲⡶⣤⣤⣿⡿⠋⠁⠻⠿⠛⠛⠙⠛⠛⠋⠉⠹⠿⠿⢿⣿⣏⣠⡖⣀⢀⣠⠤⢀⣈⣳⣄     ⢨⣶⣦⡤⣄⣀
       ⠉⢁⣴⣋⣸⠟       ⣰⣶⠴⠒      ⠈⠛⠻⢿⣿⣿⡛⠋⠉⠙⣿   ⣠⡶⣫⣭⠶⣭⡀
      ⢀⣴⠟⠉⢡⡏⡼   ⢠⡞  ⠉            ⢸⣿⡿⢿⡒⠒⠲⠿⠶⠶⠶⠟⠋⠁⣀⣀⣀⠉⠳⣄
     ⠲⣿⠷⠃⢀⣾⠷⠿⠦⢤⣤⡟   ⢀⣀⣤⣶⣯⣥⣤⣤⡞⠁   ⠈⣼⣿⣷⣝⡳⠤⣤⣀⣀   ⠉  ⠙⠻⢦⣈⢳⡄
    ⢀⡼⢋⣤⠴⠋⠁   ⣴⠿⠿⢶⣶⣿⣿⠟⠛⢻⣿⣿⠟⠁      ⠈⠻⣿⡍⠛⠷⣦⣄⡀⠳⢤⡀      ⠙⠧⣄
   ⣠⣿⠟⠉    ⣀⣀⡀ ⣤⣤⣼⣿⣿⣷⣂⣴⣿⡿⠋      ⠰⡆ ⢻⣿⣿⣶⣄⡈⠻⣝ ⠈⠙⠲⣤⣀⡀  ⠑⢦⣌⡙⠒
  ⢰⡟⠁     ⠛⢩⠶⠖⠛⣀⡏⠉⠙⠿⣿⣿⡟⠉         ⣷  ⣿⣿⣧⡙⢷⣄⡈⠂     ⠉⠉⠙⢷⡄⠈⠛⢦
 ⣠⡿⠛⢶⣦⣤⣤⣴⣶ ⠈⡿⠟⠛⠉⠁⢀⣀⣀ ⠉⠙⠛⠒⠂       ⡿ ⣽⣿⠘⢻⣷⡀⠈⠉⠉         ⠹⣆  ⠁
 ⡏  ⢸⣿⡿⠉⠙⠋ ⠈      ⠈⠉⣉⠅ ⠓⠲⢤⣄⡀    ⣼⠃ ⢿⣿  ⣿⠇⢠⡀       ⠠⣄⣄ ⢹⡆
 ⣷⡀  ⡿       ⣀⠔   ⣠⣞⣁⣀⣠⣤⣤⣷⣌⠙⢦⡀⢀⡾⠃  ⢸⣿⡆⣻⠇  ⢹⣄       ⢹⡌⢳⣜⡟
 ⢻⣧⣠⣸⡇          ⣠⡾⠟⠛⠉⣥⡾⢿⣿⣿⣿⣆ ⠙⠃    ⣿⢏⣿⡿⡀   ⠻⣷⢤⡀    ⢸⡇ ⢿⡇
  ⠉⢻⢿⣿⣶⣤⣤⣀⣀⣀⣀⣤⣴⡿⠋⠁⣠⡴⠟⢁⣴⣿⣿⣿⣿⣿⡆     ⣼⡟⣼⣿⣷⢻⡜⣆  ⠘⢷⡙  ⣠⣤⡿ ⠈⠛⠁
   ⠘⠦⢿⣍⠉⠉⠉⠙⢿⠩⢻⣿⣾⠞⠛⠁  ⣾⠏⠈⢻⣿⣿⣿⣿⡀⡀   ⢻⣰⠟⠁⠘⢦⡻⣿⡆  ⢸⣷  ⣿⡟⠁
      ⠙⠋⠛⠳⣶⣶⠷⢾⣿⣿    ⢀⣿   ⢻⣿⣿⣿⡧   ⢀⣴⠋    ⠁⠈⢳  ⣸⠙⣦⢰⡟
          ⠘⣿⣄⢼⣿⣿⣇⠒⢢⣿⣼⣧⡀ ⢤⡀⣿⣿⣿⡧  ⢀⣾⠃  ⢀⢠⡆  ⡞⢀⡴⣃⣸⡟⠳⣇
           ⠹⡽⣾⣿⠹⣿⣆⣾⢯⣿⣿ ⡞ ⠻⣿⣿⣿⠁ ⢠⣿⢏  ⡀ ⡟  ⢀⣴⣿⠃⢁⡼⠁ ⠈
             ⠈⠛ ⢻⣿⣧⢸⢟⠶⢾⡇  ⣸⡿⠁ ⢠⣾⡟⢼  ⣷ ⡇ ⣰⠋⠙⠁
                ⠈⣿⣻⣾⣦⣇⢸⣇⣀⣶⡿⠁⣀⣀⣾⢿⡇⢸  ⣟⡦⣧⣶⠏ unleashed
                 ⠸⢿⡍⠛⠻⠿⠿⠿⠋⣠⡾⢋⣾⣏⣸⣷⡸⣇⢰⠟⠛⠻⡄  v1.29
                   ⢻⡄   ⠐⠚⠋⣠⡾⣧⣿⠁⠙⢳⣽⡟
                   ⠈⠳⢦⣤⣤⣀⣤⡶⠛ ⠈⢿⡆  ⢿⡇
                         ⠈    ⠈⠓  ⠈

NAME

e - beast mode unleashed

SYNOPSIS

Add a trace marker:

    $ perl -Me -e 'sub f1 { trace } sub f2 { f1 } f2'

Watch a reference for changes:

    $ perl -Me -e 'my $v = {}; sub f1 { watch( $v ) } sub f2 { f1; $v->{a} = 1 } f2'

    $ perl -Me -e '
        package A {
            use e;
            my %h = ( aaa => 111 );

            watch(\%h);

            sub f1 {
                $h{b} = 1;
            }

            sub f2 {
                f1();
                delete $h{aaa};
            }
        }

        A::f2();
    '

Benchmark two snippets of code:

    $ perl -Me -e 'n { slow => sub{ ... }, fast => sub{ ... }}, 10000'

Launch the Runtime::Debugger:

    $ perl -Me -e 'repl'

Invoke the Tiny::Prof:

    $ perl -Me -e 'prof'

Convert a data structure to json:

    $ perl -Me -e 'say j { a => [ 1..3] }'

Convert a data structure to yaml:

    $ perl -Me -e 'say yml { a => [ 1..3] }'

Pretty print a data structure:

    $ perl -Me -e 'p { a => [ 1..3] }'

Data dump a data structure:

    $ perl -Me -e 'd { a => [ 1..3] }'

Devel::Peek dump a data structure:

    $ perl -Me -e 'dd { a => [ 1..3] }'

Print data as a table:

    $ perl -Me -e 'table( [qw(key value)], [qw(red 111)], [qw(blue 222)] )'
    +------+-------+
    | key  | value |
    +------+-------+
    | red  | 111   |
    | blue | 222   |
    +------+-------+

Encode/decode UTF-8:

    $ perl -Me -e 'printf "%#X\n", ord for split //, enc "\x{5D0}"'
    0XD7
    0X90

    $ perl -C -Me -e 'say dec "\xD7\x90"'
    $ perl -Me -e 'utf8; say dec "\xD7\x90"'
    א

DESCRIPTION

This module imports many features that make one-liners and script debugging much faster.

It has been optimized for performance to not import all features right away: thereby making its startup cost quite low.

SUBROUTINES

Investigation

repl

Add a breakpoint to code.

Basically inserts a Read Evaluate Print Loop.

Enable to analyze code in the process.

    CODE ...

    # Breakpoint
    repl

    CODE ...

Simple debugger on the command line:

    $ perl -Me -e 'repl'

trace

Show a stack trace.

    trace( OPTIONS )

OPTIONS:

    -levels  => NUM,           # How many scope levels to show.
    NUM,                       # Same.

    -raw => 1,                 # Include internal calls.
    -NUM,                      # Same.

    -message => STR,           # Message to display.
    STR,                       # Same.

watch

Watch a reference for changes.

    watch( $ref, OPTIONS )

OPTIONS:

    -clone => 0,               # Will not watch cloned objects.

    -methods => "fetch",       # Monitor just this method.
    -methods => [ "fetch" ],   # Same.

    -levels  => NUM,           # How many scope levels to show.
    NUM,                       # Same.

    -raw => 1,                 # Include internal calls.
    -NUM,                      # Same.

    -message => STR,           # Message to display.
    STR,                       # Same.

prof

Profile the code from this point on.

    my $obj = prof;
    ...
    # $obj goes out of scope and builds results.

n

Benchmark and compare different pieces of code.

    Time single block of code.
    n sub{ ... };
    n sub{ ... }, 100000;

    # Compare blocks of code.
    n {
        slow => sub{ ... },
        fast => sub{ ... },
    };
    n {
        slow => sub{ ... },
        fast => sub{ ... },
    }, 10000;

    $ perl -Me -e '$v = 333; n { concat => sub { 111 . $v }, interp => sub { "111$v" }, list => sub { 111,$v } }, 100000000'

              Rate interp concat   list
    interp  55248619/s     --    -6%   -62%
    concat  58479532/s     6%     --   -60%
    list   144927536/s   162%   148%     --

Format Conversions

j

JSON Parser.

    my $bytes = j([1, 2, 3]);
    my $bytes = j({foo => 'bar'});
    my $value = j($bytes);

Encode Perl data structure or decode JSON with "j" in Mojo::JSON.

Convert Perl object to JSON string:

    $ perl -Me -e 'say j { a => [1..3]}'

Convert JSON string to Perl object:

    $ perl -Me -e 'p j q({"a":[1,2,3]})'

x

XML parser.

    my $dom = x('<div>Hello!</div>');

Turn HTML/XML input into Mojo::DOM object.

    $ perl -Me -e 'say x("<div>hey</dev>")->at("div")->text'

Force HTML semantics:

    $ perl -Me -e 'say x->xml(0)->parse("<Tag>Name</Tag>")'
    <tag>Name</tag>

Force XML semantics (case sensitive tags and more):

    $ perl -Me -e 'say x->xml(1)->parse("<Tag>Name</Tag>")'
    <Tag>Name</Tag>

yml

YAML parser.

Convert Perl object to YAML string:

    $ perl -Me -e 'say yml { a => [1..3]}'

Convert YAML string to Perl object:

    $ perl -Me -e 'p yml "---\na:\n- 1\n- 2\n- 3"'

clone

Storable's deep clone.

    $ perl -Me -e '
        my $arr1   = [ 1..3 ];
        my $arr2   = clone $arr1;
        $arr2->[0] = 111;

        say $arr1;
        p $arr1;

        say "";
        say $arr2;
        p $arr2;
    '

    # Output:
    ARRAY(0x5d0b8a408518)
    [
        [0] 1,
        [1] 2,
        [2] 3,
    ]

    ARRAY(0x5d0b8a42d9e0)
    [
        [0] 111,
        [1] 2,
        [2] 3,
    ]

enc

Encode UTF-8 code point to a byte stream:

    $ perl -Me -e 'printf "%#X\n", ord for split //, enc "\x{5D0}"'
    0XD7
    0X90

dec

Decode a byte steam to UTF-8 code point:

    $ perl -C -Me -e 'say dec "\xD7\x90"'
    א

utf8

Set STDOUT and STDERR as UTF-8 encoded.

If given a filehandle, will set the encoding for it to UTF-8.

    utf8($fh);

Enhanced Types

b

Work with strings.

    my $stream = b('lalala');

Turn string into a Mojo::ByteStream object.

    $ perl -Me -e 'b(g("mojolicious.org")->body)->html_unescape->say'

c

Work with arrays.

    my $collection = c(1, 2, 3);

Turn list into a Mojo::Collection object.

set

Work with sets.

    my $set = set(2,4,6,4);

Turn list into a Set::Scalar object.

    $ perl -Me -e 'say set(2,4,6,2)'
    (2 4 6)

Get elements:

    $ perl -Me -e 'say for sort(set(2,4,6,2)->elements)'
    $ perl -Me -e 'say for sort(set(2,4,6,2)->@*)'
    2
    4
    6

Check for existence of an element:

    $ perl -Me -e 'say set(2,4,6,2)->has(7)'
    $ perl -Me -e 'say set(2,4,6,2)->has(4)'
    1

Intersection:

    $ perl -Me -e 'say set(2,4,6,2) * set(3,4,5,6)'
    (4 6)

Create a new universe:

    # Universe 1:
    # ...
    Set::Scalar::Universe->new->enter;
    # Universe 2:
    # ...

Operations:

    set                         value

    $a                          (a b c d e _ _ _ _)
    $b                          (_ _ c d e f g _ _)
    $c                          (_ _ _ _ e f g h i)

    union:        $a + $b       (a b c d e f g _ _)
    union:        $a + $b + $c  (a b c d e f g h i)
    intersection: $a * $b       (_ _ c d e _ _ _ _)
    intersection: $a * $b * $c  (_ _ _ _ e _ _ _ _)
    difference:   $a - $b       (a b _ _ _ _ _ _ _)
    difference:   $a - $b - $c  (a b _ _ _ _ _ _ _)
    unique:       $a % $b       (a b _ _ _ f g _ _)
    symm_diff:    $a / $b       (a b _ _ _ f g _ _)
    complement:   -$a           (_ _ c d e f g h i)

Files Convenience

f

Work with files.

    my $path = f('/home/sri/foo.txt');

Turn string into a Mojo::File object.

    $ perl -Me -e 'say r j f("hello.json")->slurp'

Math Help

max

Get the biggest number in a list.

    $ perl -Me -e 'say max 2,4,1,3'
    4

min

Get the smallest number in a list.

    $ perl -Me -e 'say max 2,4,1,3'
    1

Output

say

Obnoxious print with a newline.

    $ perl -Me -e 'say 123'
    $ perl -Me -e 'say for 1..3'

Always sends output to the terminal even when STDOUT and/or STDERR are redirected:

    $ perl -Me -e '
        say "Shown before";
        close *STDOUT;
        close *STDERR;
        say "Shown with no stdout/err";
        print "Print not seen\n";
    '
    111
    222

p

Pretty data printer.

    $ perl -Me -e 'p [1..3]'

np

Return pretty printer data.

    $ perl -Me -e 'my $v = np [1..3]; say "got: $v"'

Can be used with say to output to the terminal (incase STDOUT/STDERR are redirected):

    $ perl -Me -e '
        close *STDOUT;
        close *STDERR;
        say np [ 1.. 3 ];
    '

d

Data dumper.

    $ perl -Me -e 'd [1..3]'

dd

Internal data dumper.

    $ perl -Me -e 'dd [1..3]'

dye

Color a string.

    $ perl -Me -e 'say dye 123, "RED"'

table

Print data as a table:

    $ perl -Me -e 'table( [qw(key value)], [qw(red 111)], [qw(blue 222)] )'
    +------+-------+
    | key  | value |
    +------+-------+
    | red  | 111   |
    | blue | 222   |
    +------+-------+

Context sensitive!

    - Void   - output table.
    - List   - return individual lines.
    - Scalar - return entire table as a string.

g

    my $res = g('example.com');
    my $res = g('http://example.com' => {Accept => '*/*'} => 'Hi!');
    my $res = g('http://example.com' => {Accept => '*/*'} => form => {a => 'b'});
    my $res = g('http://example.com' => {Accept => '*/*'} => json => {a => 'b'});

Perform GET request with "get" in Mojo::UserAgent and return resulting Mojo::Message::Response object.

    $ perl -Me -e 'say g("mojolicious.org")->dom("h1")->map("text")->join("\n")'

post

    my $res = post('example.com');
    my $res = post('http://example.com' => {Accept => '*/*'} => 'Hi!');
    my $res = post('http://example.com' => {Accept => '*/*'} => form => {a => 'b'});
    my $res = post('http://example.com' => {Accept => '*/*'} => json => {a => 'b'});

Perform POST request with "get" in Mojo::UserAgent and return resulting Mojo::Message::Response object.

    $ perl -Me -e 'say post("mojolicious.org")->dom("h1")->map("text")->join("\n")'

l

Work with URLs.

    my $url = l('https://mojolicious.org');

Turn a string into a Mojo::URL object.

    $ perl -Me -e 'say l("/perldoc")->to_abs(l("https://mojolicious.org"))'

Asynchronous

This sector includes commands to run asynchronous (or pseudo-async) operations.

It is not entirely clear which method to always use.

runf limits to number of action or 20 (whichever is smaller).

runt and runio have no such limits.

Typically using threads (with runt) seems to be fastest.

Some statistics using different run commands:

    $ gitb status -d
           s/iter   runt  runio   runf series
    runt     1.74     --   -35%   -59%   -74%
    runio    1.12    55%     --   -36%   -59%
    runf    0.716   142%    56%     --   -36%
    series  0.456   281%   146%    57%     --

    $ gitb branch -d
              Rate   runt   runf series  runio
    runt   0.592/s     --   -71%   -81%   -83%
    runf    2.02/s   240%     --   -34%   -42%
    series  3.05/s   415%    51%     --   -12%
    runio   3.47/s   486%    72%    14%     --

    $ gitb pull -d
           s/iter  runio series   runt   runf
    runio    4.27     --    -7%   -21%   -33%
    series   3.97     8%     --   -15%   -28%
    runt     3.38    26%    17%     --   -15%
    runf     2.87    49%    38%    18%     --

runf

Run tasks in parallel using Parallel::ForkManager.

Returns the results.

    $ perl -Me -e '
        p {
            runf
            map {
                my $n = $_;
                sub{ $n => $n**2 };
            } 1..5
        }
    '
    {
        1 => 1,
        2 => 4,
        3 => 9,
        4 => 16,
        5 => 25,
    }

Takes much overhead to start up!

Will use up to 20 processes.

runio

Run tasks in parallel using Mojo::IOLoop.

Returns the results.

    $ perl -Me -e '
        p {
            runio
            map {
                my $n = $_;
                sub{ $n => $n**2 };
            } 1..5
        }
    '
    {
        1 => 1,
        2 => 4,
        3 => 9,
        4 => 16,
        5 => 25,
    }

This is apparently better to use for IO related tasks.

runt

Run tasks in parallel using threads.

Returns the results.

    $ perl -Me -e '
        p {
            runt
            map {
                my $n = $_;
                sub{ $n => $n**2 };
            } 1..5
        }
    '
    {
        1 => 1,
        2 => 4,
        3 => 9,
        4 => 16,
        5 => 25,
    }

This is the fastest run* command usually.

Package Tools

monkey_patch

Insert subroutines into the symbol table.

Extracted from Mojo::Util for performance.

Import methods into another package (as done in this module):

    $ perl -e '
        package A;
        use e;
        sub import {
            my $c = caller();
            monkey_patch
                $c,
                new => sub { say "Im new" };
        }
        package main;
        A->import;
        new();
    '
    Im new

Import methods into the same package (probably not so useful):

    $ perl -e '
        package A;
        use e;
        sub import {
            my $c = caller();
            monkey_patch
                $c,
                new => sub { say "Im new" };
        }
        A->import;
        A->new();
    '
    Im new

Perhaps can be updated based on the outcome of this issue: https://github.com/mojolicious/mojo/pull/2173

pod

Work with perl pod.

import

[Internal] Imports the DSL into another package.

Can be used in a sub class to import this class plus its own commands like this:

    package e2;
    use parent qw( e );

    sub import {
        my ( $class ) = @_;
        my $class = caller;
        $class->SUPER::import( $caller );
        $class->can("monkey_patch")->(
            $caller,
            my_command_1 => sub {},
            my_command_2 => sub {},
            my_command_3 => sub {},
        );
    }

AUTHOR

Tim Potapov, <tim.potapov[AT]gmail.com>

BUGS

Please report any bugs or feature requests to https://github.com/poti1/e/issues.

SUPPORT

You can find documentation for this module with the perldoc command.

    perldoc e

You can also look for information at:

https://metacpan.org/pod/e

https://github.com/poti1/e

Logo was generated using: https://emojicombos.com/dot-art-editor

LICENSE AND COPYRIGHT

This software is Copyright (c) 2024 by Tim Potapov.

This is free software, licensed under:

    The Artistic License 2.0 (GPL Compatible)