What is wrong with this PHP code used to convert css to inline?

IanArcher · October 31, 2011, 7:41pm

Here is the code

It’s an Open type free use code for convert css to inline style codes by capturing

from your own html code or a url and it outputs an html code when finished.

Because i need to send webpages in an e-mail.

I am brand new to PHP so i wouldn’t know where to start to use this code on my

own separate from the page url below.

It’s the source of this page

http://premailer.dialect.ca

Can someone shine some light on what do with the the below code to get the

result i am seeking? Thanks.

<?php
/**
 * Premailer API PHP class
 * Premailer is a library/service for making HTML more palatable for various 

inept email clients, in particular GMail
 * Primary function is to convert style tags into equivalent inline styles so styling 

can survive <head> tag removal
 * Premailer is owned by Dialect Communications group
 * @link http://premailer.dialect.ca/api
 * @author Marcus Bointon <marcus@synchromedia.co.uk>
 */
 
class Premailer {
	/**
	 * The Premailer API URL
	 */
	const ENDPOINT = 'http://premailer.dialect.ca/api/0.1/documents';

	/**
	 * Central static method for submitting either an HTML string or a 

URL, optionally retrieving converted versions
	 * @static
	 * @throws Exception
	 * @param string $html Raw HTML source
	 * @param string $url URL of the source file
	 * @param bool $fetchresult Whether to also fetch the converted 

output
	 * @param string $adaptor Which document handler to use (hpricot 

(default) or nokigiri)
	 * @param string $base_url Base URL for converting relative links
	 * @param int $line_length Length of lines in the plain text version 

(default 65)
	 * @param string $link_query_string Query string appended to links
	 * @param bool $preserve_styles Whether to preserve any link 

rel=stylesheet and style elements
	 * @param bool $remove_ids Remove IDs from the HTML document?
	 * @param bool $remove_classes Remove classes from the HTML 

document?
	 * @param bool $remove_comments Remove comments from the 

HTML document?
	 * @return array Either a single strclass object containing the decoded 

JSON response, or a 3-element array containing result, html and plain parts if 

$fetchresult is set
	 */
	protected static function convert($html = '', $url = 

'http://test2.chamberhome.com/viewPage.php?ID=Newsletter_1', $fetchresult = 

true, $adaptor = 'hpricot', $base_url = '', $line_length = 65, $link_query_string 

= '', $preserve_styles = true, $remove_ids = false, $remove_classes = false, 

$remove_comments = false) {
		$params = array();
		if (!empty($html)) {
			$params['html'] = $html;
		} elseif (!empty($url)) {
			$params['url'] = $url;
		} else {
			throw new Exception('Must supply an html or url 

value');
		}
		if ($adaptor == 'hpricot' or $adaptor == 'nokigiri') {
			$params['adaptor'] = $adaptor;
		}
		if (!empty($base_url)) {
			$params['base_url'] = $base_url;
		}
		$params['line_length'] = (integer)$line_length;
		if (!empty($link_query_string)) {
			$params['link_query_string'] = 

$link_query_string;
		}
		$params['preserve_styles'] = ($preserve_styles?'true':'false');
		$params['remove_ids'] = ($remove_ids?'true':'false');
		$params['$remove_classes'] = 

($remove_classes?'true':'false');
		$params['$remove_comments'] = 

($remove_comments?'true':'false');
		$options = array(
			'timeout' => 15,
			'connecttimeout' => 15,
			'useragent' => 'PHP Premailer',
			'ssl' => array('verifypeer' => false, 'verifyhost' => 

false)
		);
		$h = new HttpRequest(self::ENDPOINT, 

HttpRequest::METH_POST, $options);
		$h->addPostFields($params);
		try {
			$response = $h->send();
			$result = json_decode($response->getBody());
			$code = $response->getResponseCode();
			if ($code != 201) {
				switch ($code) {
					case 400:
						throw new 

Exception('Content missing', 400);
						break;
					case 403:
						throw new 

Exception('Access forbidden', 403);
						break;
					case 500:
					default:
						throw new 

Exception('Error', $code);
				}
			}
			$return = array('result' => $result);
			if ($fetchresult)() {
				//Get HTML and plain versions in 

parallel
				http_persistent_handles_clean();
				$pool = new HttpRequestPool;
				$pool->attach(new HttpRequest

($result->documents->html, HttpRequest::METH_GET, $options));
				$pool->attach(new HttpRequest

($result->documents->txt, HttpRequest::METH_GET, $options));
				$pool->send();
				foreach($pool as $request) {
					if ($request->getUrl() == 

$result->documents->html) {
						$return['html'] = 

$request->getResponseBody();
					} elseif ($request->getUrl() 

== $result->documents->txt) {
						$return['plain'] = 

$request->getResponseBody();
					}
				}
				return $return;
			}
			return $result;
		} catch (HttpException $e) {
			var_dump($h, $e->getMessage());
			return false;
		}
	}

	/**
	 * Central static method for submitting either an HTML string or a 

URL, optionally retrieving converted versions
	 * @static
	 * @throws Exception
	 * @param string $html Raw HTML source
	 * @param bool $fetchresult Whether to also fetch the converted 

output
	 * @param string $adaptor Which document handler to use (hpricot 

(default) or nokigiri)
	 * @param string $base_url Base URL for converting relative links
	 * @param int $line_length Length of lines in the plain text version 

(default 65)
	 * @param string $link_query_string Query string appended to links
	 * @param bool $preserve_styles Whether to preserve any link 

rel=stylesheet and style elements
	 * @param bool $remove_ids Remove IDs from the HTML document?
	 * @param bool $remove_classes Remove classes from the HTML 

document?
	 * @param bool $remove_comments Remove comments from the 

HTML document?
	 * @return array Either a single element array containing the 'result' 

object, or three elements containing result, html and plain if $fetchresult is set
	 */
	public static function html($html, $fetchresult = true, $adaptor = 

'hpricot', $base_url = '', $line_length = 65, $link_query_string = '', 

$preserve_styles = true, $remove_ids = false, $remove_classes = false, 

$remove_comments = false) {
		return self::convert($html, '', $fetchresult, $adaptor, 

$base_url, $line_length, $link_query_string, $preserve_styles, $remove_ids, 

$remove_classes, $remove_comments);
	}

	/**
	 * Central static method for submitting either an HTML string or a 

URL, optionally retrieving converted versions
	 * @static
	 * @throws Exception
	 * @param string $url URL of the source file
	 * @param bool $fetchresult Whether to also fetch the converted 

output
	 * @param string $adaptor Which document handler to use (hpricot 

(default) or nokigiri)
	 * @param string $base_url Base URL for converting relative links
	 * @param int $line_length Length of lines in the plain text version 

(default 65)
	 * @param string $link_query_string Query string appended to links
	 * @param bool $preserve_styles Whether to preserve any link 

rel=stylesheet and style elements
	 * @param bool $remove_ids Remove IDs from the HTML document?
	 * @param bool $remove_classes Remove classes from the HTML 

document?
	 * @param bool $remove_comments Remove comments from the 

HTML document?
	 * @return array Either a single element array containing the 'result' 

object, or three elements containing result, html and plain if $fetchresult is set
	 */
	public static function url($url, $fetchresult = true, $adaptor = 'hpricot', 

$base_url = '', $line_length = 65, $link_query_string = '', $preserve_styles = 

true, $remove_ids = false, $remove_classes = false, $remove_comments = 

false) {
		return self::convert('', $url, $fetchresult, $adaptor, 

$base_url, $line_length, $link_query_string, $preserve_styles, $remove_ids, 

$remove_classes, $remove_comments);
	}
}

/*
Simplest usage:
$pre = Premailer::html($var_with_some_html_in);
$html = $pre['html'];
$plain = $pre['plain'];
//Similarly for URLs:
$pre = Premailer::url($url);
*/
?>

ParkinT · November 2, 2011, 12:28pm

You need to call this file on a PHP server (almost any hosting service available on the planet).
Sample usage is at the bottom of the file. So, the simplest way to use this is:

remove the comments from that block of code at the bottom (delete the /* at the beginning and the */ at the end)
move the last line [$pre = Premailer::url($url)] up to the top of the block of code from which you removed the comments (but just below the other similar line)
replace ‘$url’ with the URL of a page you want to modify
Call the page (http://yourhosteddomain/whateverthefilenameis.php

The variables ($html and $plain) will contain the modified text for you

If you add this code at the end of the same file, the output will be displayed


echo $html;

IanArcher · November 2, 2011, 5:29pm

Thanks for the help!

Here’s my code now

<?php
/**
* Premailer API PHP class
* Premailer is a library/service for making HTML more palatable for various inept email clients, in particular GMail
* Primary function is to convert style tags into equivalent inline styles so styling can survive <head> tag removal
* Premailer is owned by Dialect Communications group
* @link http://premailer.dialect.ca/api
* @author Marcus Bointon <marcus@synchromedia.co.uk>
*/
 
class Premailer {
/**
* The Premailer API URL
*/
const ENDPOINT = 'http://premailer.dialect.ca/api/0.1/documents';

/**
* Central static method for submitting either an HTML string or a URL, optionally retrieving converted versions
* @static
* @throws Exception
* @param string $html Raw HTML source
* @param string $url URL of the source file
* @param bool $fetchresult Whether to also fetch the converted output
* @param string $adaptor Which document handler to use (hpricot (default) or nokigiri)
* @param string $base_url Base URL for converting relative links
* @param int $line_length Length of lines in the plain text version (default 65)
* @param string $link_query_string Query string appended to links
* @param bool $preserve_styles Whether to preserve any link rel=stylesheet and style elements
* @param bool $remove_ids Remove IDs from the HTML document?
* @param bool $remove_classes Remove classes from the HTML document?
* @param bool $remove_comments Remove comments from the HTML document?
* @return array Either a single strclass object containing the decoded JSON response, or a 3-element array containing result, html and plain parts if $fetchresult is set
*/


protected static function convert($html = '', $url = "test2.chamberhome.com/viewPage.php?ID=Newsletter_1", $fetchresult = true, $adaptor = 'hpricot', $base_url = '', $line_length = 65, $link_query_string = '', $preserve_styles = true, $remove_ids = false, $remove_classes = false, $remove_comments = false) {
$params = array();
if (!empty($html)) {
$params['html'] = $html;
} elseif (!empty($url)) {
$params['url'] = $url;
} else {
throw new Exception('Must supply an html or url value');
}
if ($adaptor == 'hpricot' or $adaptor == 'nokigiri') {
$params['adaptor'] = $adaptor;
}
if (!empty($base_url)) {
$params['base_url'] = $base_url;
}
$params['line_length'] = (integer)$line_length;
if (!empty($link_query_string)) {
$params['link_query_string'] = $link_query_string;
}
$params['preserve_styles'] = ($preserve_styles?'true':'false');
$params['remove_ids'] = ($remove_ids?'true':'false');
$params['$remove_classes'] = ($remove_classes?'true':'false');
$params['$remove_comments'] = ($remove_comments?'true':'false');
$options = array(
'timeout' => 15,
'connecttimeout' => 15,
'useragent' => 'PHP Premailer',
'ssl' => array('verifypeer' => false, 'verifyhost' => false)
);
$h = new HttpRequest(self::ENDPOINT, HttpRequest::METH_POST, $options);
$h->addPostFields($params);
try {
$response = $h->send();
$result = json_decode($response->getBody());
$code = $response->getResponseCode();
if ($code != 201) {
switch ($code) {
case 400:
throw new Exception('Content missing', 400);
break;
case 403:
throw new Exception('Access forbidden', 403);
break;
case 500:
default:
throw new Exception('Error', $code);
}
}
$return = array('result' => $result);
if ($fetchresult) {
//Get HTML and plain versions in parallel
http_persistent_handles_clean();
$pool = new HttpRequestPool;
$pool->attach(new HttpRequest($result->documents->html, HttpRequest::METH_GET, $options));
$pool->attach(new HttpRequest($result->documents->txt, HttpRequest::METH_GET, $options));
$pool->send();
foreach($pool as $request) {
if ($request->getUrl() == $result->documents->html) {
$return['html'] = $request->getResponseBody();
} elseif ($request->getUrl() == $result->documents->txt) {
$return['plain'] = $request->getResponseBody();
}
}
return $return;
}
return $result;
} catch (HttpException $e) {
var_dump($h, $e->getMessage());
return false;
}
}


//Simplest usage:
$pre = file_get_contents(http://test2.chamberhome.com/viewPage.php?ID=Newsletter_1);
$html = $pre['html'];
$plain = $pre['plain'];

echo $html;




?>

I constantly get errors about parsing for example, with that current code it get the error:
“Parse error: syntax error, unexpected T_VARIABLE, expecting T_FUNCTION in C:\xampp\htdocs\mydiv\Premailer.php on line 110”

Line 110 is

$pre = file_get_contents(http://test2.chamberhome.com/viewPage.php?ID=Newsletter_1);

SpikeZ · November 2, 2011, 9:01pm

You need some quotes around the url

$pre = file_get_contents(‘http://test2.chamberhome.com/viewPage.php?ID=Newsletter_1’);

IanArcher · November 3, 2011, 2:01am

Thanks you, but i am still receiving the same error.

SpikeZ · November 3, 2011, 7:32am

also missing a } just above that line (to match the opening { in the class declaration line)

IanArcher · November 3, 2011, 3:36pm

Thanks for the help. I don’t get errors anymore, but when i load the page now i just get a:

<

so i added a code to write to a file called echo.htm, and when i open the echo.htm it looks fine but i look at the source code of it and it’s still depending on the CSS files from the server instead of inline. i even went offline refreshed the page and it looked completely different, that’s how i knew.

Sorry to bother so much, but please tell me what must i do next or what i’m doing wrong? Thanks

<?php
/**
* Premailer API PHP class
* Premailer is a library/service for making HTML more palatable for various inept email clients, in particular GMail
* Primary function is to convert style tags into equivalent inline styles so styling can survive <head> tag removal
* Premailer is owned by Dialect Communications group
* @link http://premailer.dialect.ca/api
* @author Marcus Bointon <marcus@synchromedia.co.uk>
*/
 
class Premailer {
/**
* The Premailer API URL
*/
const ENDPOINT = 'http://premailer.dialect.ca/api/0.1/documents';

/**
* Central static method for submitting either an HTML string or a URL, optionally retrieving converted versions
* @static
* @throws Exception
* @param string $html Raw HTML source
* @param string $url URL of the source file
* @param bool $fetchresult Whether to also fetch the converted output
* @param string $adaptor Which document handler to use (hpricot (default) or nokigiri)
* @param string $base_url Base URL for converting relative links
* @param int $line_length Length of lines in the plain text version (default 65)
* @param string $link_query_string Query string appended to links
* @param bool $preserve_styles Whether to preserve any link rel=stylesheet and style elements
* @param bool $remove_ids Remove IDs from the HTML document?
* @param bool $remove_classes Remove classes from the HTML document?
* @param bool $remove_comments Remove comments from the HTML document?
* @return array Either a single strclass object containing the decoded JSON response, or a 3-element array containing result, html and plain parts if $fetchresult is set
*/


protected static function convert($html = '', $url = "test2.chamberhome.com/viewPage.php?ID=Newsletter_1", $fetchresult = true, $adaptor = 'hpricot', $base_url = '', $line_length = 65, $link_query_string = '', $preserve_styles = true, $remove_ids = false, $remove_classes = false, $remove_comments = false) {
$params = array();
if (!empty($html)) {
$params['html'] = $html;
} elseif (!empty($url)) {
$params['url'] = $url;
} else {
throw new Exception('Must supply an html or url value');
}
if ($adaptor == 'hpricot' or $adaptor == 'nokigiri') {
$params['adaptor'] = $adaptor;
}
if (!empty($base_url)) {
$params['base_url'] = $base_url;
}
$params['line_length'] = (integer)$line_length;
if (!empty($link_query_string)) {
$params['link_query_string'] = $link_query_string;
}
$params['preserve_styles'] = ($preserve_styles?'true':'false');
$params['remove_ids'] = ($remove_ids?'true':'false');
$params['$remove_classes'] = ($remove_classes?'true':'false');
$params['$remove_comments'] = ($remove_comments?'true':'false');
$options = array(
'timeout' => 15,
'connecttimeout' => 15,
'useragent' => 'PHP Premailer',
'ssl' => array('verifypeer' => false, 'verifyhost' => false)
);
$h = new HttpRequest(self::ENDPOINT, HttpRequest::METH_POST, $options);
$h->addPostFields($params);
try {
$response = $h->send();
$result = json_decode($response->getBody());
$code = $response->getResponseCode();
if ($code != 201) {
switch ($code) {
case 400:
throw new Exception('Content missing', 400);
break;
case 403:
throw new Exception('Access forbidden', 403);
break;
case 500:
default:
throw new Exception('Error', $code);
}
}
$return = array('result' => $result);
if ($fetchresult) {
//Get HTML and plain versions in parallel
http_persistent_handles_clean();
$pool = new HttpRequestPool;
$pool->attach(new HttpRequest($result->documents->html, HttpRequest::METH_GET, $options));
$pool->attach(new HttpRequest($result->documents->txt, HttpRequest::METH_GET, $options));
$pool->send();
foreach($pool as $request) {
if ($request->getUrl() == $result->documents->html) {
$return['html'] = $request->getResponseBody();
} elseif ($request->getUrl() == $result->documents->txt) {
$return['plain'] = $request->getResponseBody();
}
}
return $return;
}
return $result;
} catch (HttpException $e) {
var_dump($h, $e->getMessage());
return false;
}
}
}

//Simplest usage:
$pre = file_get_contents('http://test2.chamberhome.com/viewPage.php?ID=Newsletter_1');
$html = $pre['html'];
$plain = $pre['plain'];

// Write to file
$f = fopen('echo.htm', 'w');
fwrite($f, $pre);
fclose($f);

echo $html;




?>

SpikeZ · November 3, 2011, 3:52pm

I cant see where you are actually calling the class…

IanArcher · November 3, 2011, 4:04pm

Do you think i would need this code?

Or in the simplest usage section do something like $params because that looks like what activates the conversion. That’s really what i’m getting after, activating the conversion.

	/**
	 * Central static method for submitting either an HTML string or a URL, optionally retrieving converted versions
	 * @static
	 * @throws Exception
	 * @param string $html Raw HTML source
	 * @param bool $fetchresult Whether to also fetch the converted output
	 * @param string $adaptor Which document handler to use (hpricot (default) or nokigiri)
	 * @param string $base_url Base URL for converting relative links
	 * @param int $line_length Length of lines in the plain text version (default 65)
	 * @param string $link_query_string Query string appended to links
	 * @param bool $preserve_styles Whether to preserve any link rel=stylesheet and style elements
	 * @param bool $remove_ids Remove IDs from the HTML document?
	 * @param bool $remove_classes Remove classes from the HTML document?
	 * @param bool $remove_comments Remove comments from the HTML document?
	 * @return array Either a single element array containing the 'result' object, or three elements containing result, html and plain if $fetchresult is set
	 */
	public static function html($html, $fetchresult = true, $adaptor = 'hpricot', $base_url = '', $line_length = 65, $link_query_string = '', $preserve_styles = true, $remove_ids = false, $remove_classes = false, $remove_comments = false) {
		return self::convert($html, '', $fetchresult, $adaptor, $base_url, $line_length, $link_query_string, $preserve_styles, $remove_ids, $remove_classes, $remove_comments);
	}

	/**
	 * Central static method for submitting either an HTML string or a URL, optionally retrieving converted versions
	 * @static
	 * @throws Exception
	 * @param string $url URL of the source file
	 * @param bool $fetchresult Whether to also fetch the converted output
	 * @param string $adaptor Which document handler to use (hpricot (default) or nokigiri)
	 * @param string $base_url Base URL for converting relative links
	 * @param int $line_length Length of lines in the plain text version (default 65)
	 * @param string $link_query_string Query string appended to links
	 * @param bool $preserve_styles Whether to preserve any link rel=stylesheet and style elements
	 * @param bool $remove_ids Remove IDs from the HTML document?
	 * @param bool $remove_classes Remove classes from the HTML document?
	 * @param bool $remove_comments Remove comments from the HTML document?
	 * @return array Either a single element array containing the 'result' object, or three elements containing result, html and plain if $fetchresult is set
	 */
	public static function url($url, $fetchresult = true, $adaptor = 'hpricot', $base_url = '', $line_length = 65, $link_query_string = '', $preserve_styles = true, $remove_ids = false, $remove_classes = false, $remove_comments = false) {
		return self::convert('', $url, $fetchresult, $adaptor, $base_url, $line_length, $link_query_string, $preserve_styles, $remove_ids, $remove_classes, $remove_comments);
	}
}

IanArcher · November 9, 2011, 1:08am

you’re right i wasn’t calling the class at all, that’s why it didn’t work. Can you write me something for that please?