/**
* testHTML2BBcode, parse html to BBC and checks that the results are what we expect
*/
public function testHTML2BBcode()
{
foreach ($this->bbcTestCases as $testcase) {
$name = $testcase[0];
$test = $testcase[1];
$expected = $testcase[2];
$parser = new Html_2_BBC($test);
// Convert the html to bbc
$result = $parser->get_bbc();
// Remove pretty print newlines
$result = str_replace("\n", '', $result);
$this->assertEqual($expected, $result);
}
}
/**
* Converts text / HTML to BBC
*
* What it does:
* - protects certain tags from conversion
* - strips original message from the reply if possible
* - If the email is html based, this will convert basic html tags to bbc tags
* - If the email is plain text it will convert it to html based on markdown text
* conventions and then that will be converted to bbc.
*
* @uses Html2BBC.class.php for the html to bbc conversion
* @uses markdown.php for text to html conversions
* @package Maillist
* @param string $text
* @param boolean $html
*/
function pbe_email_to_bbc($text, $html)
{
// Define some things that need to be converted/modified, outside normal html or markup
$tags = array('~\\*\\*(.*)\\*\\*~isUe' => '\'**\'.ltrim(\'$1\').\'**\'', '~<\\*>~i' => '<*>', '~-{20,}~' => '<hr>', '~#([0-9a-fA-F]{4,6}\\b)~' => '#$1');
// We are starting with HTML, our goal is to convert the best parts of it to BBC,
if ($html) {
// Convert the email-HTML to BBC
$text = preg_replace(array_keys($tags), array_values($tags), $text);
require_once SUBSDIR . '/Html2BBC.class.php';
$bbc_converter = new Html_2_BBC($text);
$text = $bbc_converter->get_bbc();
// Run our parsers, as defined in the ACP, to remove the original "replied to" message
// before we do any more work.
$text_save = $text;
$result = pbe_parse_email_message($text);
// If we have no message left after running the parser, then they may have replied
// below and/or inside the original message. People like this should not be allowed
// to use the net, or be forced to read their own messed up emails
if (empty($result) || trim(strip_tags(pbe_filter_email_message($text))) === '') {
$text = $text_save;
}
} else {
// Run the parser to try and remove common mail clients "reply to" stuff
$text_save = $text;
$result = pbe_parse_email_message($text);
// Bottom feeder? If we have no message they could have replied below the original message
if (empty($result) || trim(strip_tags(pbe_filter_email_message($text))) === '') {
$text = $text_save;
}
// Set a gmail flag for special quote processing since its quotes are strange
$gmail = (bool) preg_match('~<div class="gmail_quote">~i', $text);
// Attempt to fix textual ('>') quotes so we also fix wrapping issues first!
$text = pbe_fix_email_quotes($text, $html && !$gmail);
// Convert this (markup) text to html
$text = preg_replace(array_keys($tags), array_values($tags), $text);
require_once EXTDIR . '/markdown/markdown.php';
$text = Markdown($text);
// Convert any resulting HTML created by markup style text in the email to BBC
require_once SUBSDIR . '/Html2BBC.class.php';
$bbc_converter = new Html_2_BBC($text, false);
$text = $bbc_converter->get_bbc();
}
// Some tags often end up as just empty tags - remove those.
$emptytags = array('~\\[[bisu]\\]\\s*\\[/[bisu]\\]~' => '', '~\\[[bisu]\\]\\s*\\[/[bisu]\\]~' => '', '~(\\n){3,}~si' => "\n\n");
$text = preg_replace(array_keys($emptytags), array_values($emptytags), $text);
return $text;
}
/**
* Converts text / HTML to BBC
*
* What it does:
* - protects certain tags from conversion
* - strips original message from the reply if possible
* - If the email is html based, this will convert basic html tags to bbc tags
* - If the email is plain text it will convert it to html based on markdown text
* conventions and then that will be converted to bbc.
*
* @uses Html2BBC.class.php for the html to bbc conversion
* @uses markdown.php for text to html conversions
* @package Maillist
* @param string $text
* @param boolean $html
*/
function pbe_email_to_bbc($text, $html)
{
// Define some things that need to be converted/modified, outside normal html or markup
$tags = array('~\\*\\*\\s?(.*?)\\*\\*~is' => '**$1**', '~<\\*>~i' => '<*>', '~-{20,}~' => '<hr>', '~#([0-9a-fA-F]{4,6}\\b)~' => '#$1');
// We are starting with HTML, our goal is to convert the best parts of it to BBC,
if ($html) {
// upfront pre-process $tags, mostly for the email template strings
$text = preg_replace(array_keys($tags), array_values($tags), $text);
// Run the parsers on the html
$text = pbe_run_parsers($text);
// Convert the email-HTML to BBC
require_once SUBSDIR . '/Html2BBC.class.php';
$bbc_converter = new Html_2_BBC($text);
$bbc_converter->skip_tags(array('font', 'span'));
$bbc_converter->skip_styles(array('font-family', 'font-size', 'color'));
$text = $bbc_converter->get_bbc();
} else {
// Run the parser to try and remove common mail clients "reply to" stuff
$text = pbe_run_parsers($text);
// Set a gmail flag for special quote processing since its quotes are strange
$gmail = (bool) preg_match('~<div class="gmail_quote">~i', $text);
// Attempt to fix textual ('>') quotes so we also fix wrapping issues first!
$text = pbe_fix_email_quotes($text, $html && !$gmail);
$text = str_replace(array('[quote]', '[/quote]'), array('>blockquote>', '>/blockquote>'), $text);
// Convert this (markup) text to html
$text = preg_replace(array_keys($tags), array_values($tags), $text);
require_once EXTDIR . '/markdown/markdown.php';
$text = Markdown($text);
$text = str_replace(array('>blockquote>', '>/blockquote>'), array('<blockquote>', '</blockquote>'), $text);
// Convert any resulting HTML created by markup style text in the email to BBC
require_once SUBSDIR . '/Html2BBC.class.php';
$bbc_converter = new Html_2_BBC($text, false);
$text = $bbc_converter->get_bbc();
}
// Some tags often end up as just empty tags - remove those.
$emptytags = array('~\\[[bisu]\\]\\s*\\[/[bisu]\\]~' => '', '~\\[[bisu]\\]\\s*\\[/[bisu]\\]~' => '', '~(\\n){3,}~si' => "\n\n");
$text = preg_replace(array_keys($emptytags), array_values($emptytags), $text);
return $text;
}