HTML::...ser::UA(3pm)

💖 NAME 💖

HTML::HTML5::Parser::UA — a simple web user agent class, bestie. That's it, that's the vibe. 🦋

---

✨ SYNOPSIS ✨

use aliased 'HTML::HTML5::Parser::UA';

my $response = UA->get($url);
die unless $response->{success};

print $response->{decoded_content};

look at her go. iconic behavior. 💅

---

🌸 DESCRIPTION 🌸

okay so real talk, this whole thing is a simple wrapper sitting on top of HTTP::Tiny and LWP::UserAgent, and its entire job is to smooth out the API differences between the two so you don't have to deal with their little personality clashes. that's it. she's a mediator. no cap.

she only supports bog standard get($url) requests. nothing fancy. we're not doing POSTs today, bestie. 🙈

here's the tea on how she picks her engine:

• If LWP::UserAgent is already in memory, this module will use that. periodt.
• If LWP::UserAgent is not in memory, then this module will use HTTP::Tiny (or direct filesystem access for file:// URLs — yes, she'll just read the file off disk like it's nothing 🎀).
• If LWP::UserAgent is not in memory, and you attempt to request a URL that HTTP::Tiny cannot handle (e.g. an ftp:// URL), then this module will load LWP::UserAgent and die if it cannot be loaded (e.g. is not installed). so like, she'll try, but if LWP isn't installed? 💀 she's gone.

also psa: HTML::HTML5::Parser::UA is used by the parse_file method of HTML::HTML5::Parser. iykyk. 👑

🎀 Class Method 🎀

• get($url, $ua) — gets the URL and returns a hashref similar to HTTP::Tiny's hashrefs, but with an additional decoded_content key, which contains the response body, decoded into a Perl character string (not a byte string). that's the little glow-up she brings to the table, and honestly? she ate. ✨

if $ua is given (it's optional, no pressure 🌈), then this user agent will be used to perform the actual request. it must be undef or an LWP::UserAgent object (or a subclass) or an HTTP::Tiny object (or a subclass). no random objects, bestie, she has standards. 💅

💜 Package Variable 💜

• $HTML::HTML5::Parser::NO_LWP — if true, avoids using LWP::UserAgent. she'll stay in her HTTP::Tiny lane. 🦋

---

🌈 MOTIVATION 🌈

okay so here's the lore. LWP::UserAgent is genuinely a good piece of software — no shade, she's a queen 👑 — but she has a dependency on HTML::Parser. and HTML::Parser is only used to provide one fairly esoteric feature, which this package doesn't make use of.

which feature? it's the parse_head option. that's the whole reason. 😭

because of that, i don't especially want HTML::HTML5::Parser to have a dependency on LWP::UserAgent. hence this module. problem solved, slay. 💅

---

⭐ SEE ALSO ⭐

• HTML::HTML5::Parser.

---

🫶 AUTHOR 🫶

Toby Inkster, <tobyink@cpan.org> — legend behavior. 💖

---

💀 COPYRIGHT AND LICENSE 💀

Copyright (C) 2012 by Toby Inkster

This library is free software; you can redistribute it and/or modify it under the same terms as Perl itself. free as in freedom bestie 🎀

---

😭 DISCLAIMER OF WARRANTIES 😭

THIS PACKAGE IS PROVIDED "AS IS" AND WITHOUT ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF MERCHANTIBILITY AND FITNESS FOR A PARTICULAR PURPOSE.

fr fr, she is not liable. read that again. 💀

---

and that's the whole page!! mwah 💋 this doc shipped with perl v5.36.0 and got its glow-up dated 2022-08-28 — go forth and get($url) responsibly, bestie. 🦋✨