pQuery - A port of jQuery.js to Perl

This document describes pQuery version 0.24.

use pQuery;

pQuery("https://google.com/search?q=pquery")
    ->find("h2")
    ->each(sub {
        my $i = shift;
        print $i + 1, ") ", pQuery($_)->text, "\n";
    });

pQuery is a pragmatic attempt to port the jQuery JavaScript framework to Perl. It is pragmatic in the sense that it switches certain JavaScript idioms for Perl ones, in order to make the use of it concise. A primary goal of jQuery is to "Find things and do things, concisely". pQuery has the same goal.

pQuery exports a single function called pQuery. (Actually, it also exports the special PQUERY function. Read below.) This function acts a constructor and does different things depending on the arguments you give it. This is discussed in the CONSTRUCTORS section below.

A pQuery object acts like an array reference (because, in fact, it is). Typically it is an array of pQuery::DOM elements, but it can be an array of anything.

pQuery::DOM is roughly an attempt to duplicate JavaScript's DOM in Perl. It subclasses HTML::TreeBuilder/HTML::Element so there are a few differences to be aware of. See the pQuery::DOM documentation for details.

Like jQuery, pQuery methods return a pQuery object; either the original object or a new derived object. All pQuery METHODS are described below.

The power of jQuery is that single method calls can apply to many DOM objects. pQuery does the exact same thing but can take this one step further. A single PQUERY object can contain several DOMs!

Consider this example:

> perl -MpQuery -le 'PQUERY(\
    map "https://search.cpan.org/~$_/", qw(ingy gugod miyagawa))\
    ->find("table")->eq(1)->find("tr")\
    ->EACH(sub{\
        printf("%40s - %s Perl distributions\n", $_->url, $_->length - 1)\
    })'
           https://search.cpan.org/~ingy/ - 88 Perl distributions
          https://search.cpan.org/~gugod/ - 86 Perl distributions
       https://search.cpan.org/~miyagawa/ - 138 Perl distributions

The power lies in PQUERY, a special constructor that creates a wrapper object for many pQuery objects, and applies all methods called on it to all the pQuery objects it contains.

The pQuery constructor is an exported function called pQuery. It does different things depending on the arguments you pass it.

If you pass pQuery a URL, it will attempt to get the page and use its HTML to create a pQuery::DOM object. The pQuery object will contain the top level pQuery::DOM object.

pQuery("https://google.com");

It will also set the global variable $pQuery::document to the resulting DOM object. Future calls to pQuery methods will use this document if none other is supplied.

If you already have an HTML string, pass it to pQuery and it will create a pQuery::DOM object. The pQuery object will contain the top level pQuery::DOM object.

pQuery("<p>Hello <b>world</b>.</p>");

If you pass pQuery a string that ends with .html and contains no whitespace, pQuery will assume it is the name of a file containing html and will read the contents and parse the HTML into a new DOM.

pQuery("my/webpage.html");

You can create a pQuery object with a selector string just like in jQuery. The problem is that Perl doesn't have a global document object lying around like JavaScript does.

One thing you can do is set the global variable, $pQuery::document, to a pQuery::DOM document. This will be used by future selectors.

Another thing you can do is pass the document to select on as the second parameter. (jQuery also has this second, context parameter).

pQuery("table.mygrid > td:eq(7)", $dom);

You can create a new pQuery object from another pQuery object. The new object will be a shallow copy.

my $pquery2 = pQuery($pquery1);

You can create a pQuery object as an array of anything you want; not just pQuery::DOM elements. This can be useful to use the each method to iterate over the array.

pQuery(\ @some_array);

Calling pQuery with no arguments will return a pQuery object that is just an empty array reference. This is useful for using it to call class methods that don't need a DOM object.

my $html = pQuery->get("https://google.com")->content;

The PQUERY constructor takes a list of any of the above pQuery forms and creates a PQUERY object with one pQuery object per argument.

This is a reference of all the methods you can call on a pQuery object. They are almost entirely ported from jQuery.

pquery()

Returns the version number of the pQuery module.

size()

Returns the number of elements in the pQuery object.

length()

Also returns the number of elements in the pQuery object.

each($sub)

This method takes a subroutine reference and calls the subroutine once for each member of the pQuery object that called each. When the subroutine is called it is passed an integer count starting at 0 at incremented once for each call. It is also passed the current member of the pQuery object in $_.

pQuery("td", $dom)->each(sub {
    my $i = shift;
    print $i, " => ", pQuery($_)->text(), "\n";
});

The each method returns the pQuery object that called it.

EACH($sub)

This method can only be called on PQUERY objects. The sub is called once for every pQuery object within the PQUERY object. If you call each() on a PQUERY object, it iterates on all the DOM objects of each pQuery object (as you would expect).

find($selector)

This method will search all the pQuery::DOM elements of the its caller for all sub elements that match the selector string. It will return a new pQuery object containing all the elements found.

my $pquery2 = $pquery1->find("h1,h2,h3");

html() html($html)

This method is akin to the famous JavaScript/DOM function innerHTML.

If called with no arguments, this will return the the inner HTML string of the first DOM element in the pQuery object.

If called with an HTML string argument, this will set the inner HTML of all the DOM elements in the pQuery object.

toHtml()

This extremely handy method is not ported from jQuery. Maybe jQuery will port it back some day. :)

This function takes no arguments, and returns the outer HTML of the first DOM object in the pQuery object. Outer HTML means the HTML of the current object and its inner HTML.

For example:

pQuery('<p>I <b>like</b> pie</p>')->toHtml;

returns:

<p>I <b>like</b> pie</p>

while:

pQuery('<p>I <b>like</b> pie</p>')->html();

returns:

I <b>like</b> pie

end()

Revert the most recent 'destructive' operation, changing the set of matched elements to its previous state (right before the destructive operation). This method is useful for getting back to a prior context when chaining pQuery methods.

pQuery("table", $dom)     # Select all the tables
    ->find("td")          # Select all the tds
    ->each(sub { ... })   # Do something with the tds
    ->end()               # Go back to the tables selection
    ->each(sub { ... });  # Do something with the tables

get($index) get($url)

If this method is passed an integer, it will return that specific element from the array of elements in the pQuery object.

Given a URL, this method will fetch the HTML content of the URL and return a HTML::Response object.

my $html = pQuery->get("https://google.com")->content;

index($elem)

This method returns the index number of its argument if the elem is in the current pQuery object. Otherwise it returns -1.

reset()

This method releases resources associated with pQuery and prevents memory leaks.

This module is still being written. The documented methods all work as documented (but may not be completed ports of their jQuery counterparts yet).

The selector syntax is still very limited. (Single tags, IDs and classes only).

Version 0.02 added the pQuery::DOM class which is a huge improvement, and should facilitate making the rest of the porting easy.

But there is still much more code to port. Stay tuned...

Ingy döt Net <[email protected]>

Copyright 2008-2016. Ingy döt Net.

This program is free software; you can redistribute it and/or modify it under the same terms as Perl itself.

See https://www.perl.com/perl/misc/Artistic.html