Hello World!
` becomes `Hello World!`).
* @return object A simplehtmldom object of the remaining contents.
*
* @todo Check if this implementation is still necessary, because simplehtmldom
* already removes some of the tags (search for `remove_noise` in simple_html_dom.php).
* @todo Rename parameters to make more sense. `$textToSanitize` must be HTML,
* `$removedTags`, `$keptAttributes` and `$keptText` are past tense.
* @todo Clarify the meaning of `*[!b38fd2b1fe7f4747d6b1c1254ccd055e]`, which
* looks like a SHA1 hash (does simplehtmldom not support `find('*')`?).
*/
function sanitize($textToSanitize,
$removedTags = array('script', 'iframe', 'input', 'form'),
$keptAttributes = array('title', 'href', 'src'),
$keptText = array()){
$htmlContent = str_get_html($textToSanitize);
foreach($htmlContent->find('*[!b38fd2b1fe7f4747d6b1c1254ccd055e]') as $element) {
if(in_array($element->tag, $keptText)) {
$element->outertext = $element->plaintext;
} elseif(in_array($element->tag, $removedTags)) {
$element->outertext = '';
} else {
foreach($element->getAllAttributes() as $attributeName => $attribute) {
if(!in_array($attributeName, $keptAttributes))
$element->removeAttribute($attributeName);
}
}
}
return $htmlContent;
}
/**
* Replace background by image
*
* Replaces tags with styles of `backgroud-image` by `` tags.
*
* For example:
*
* ```HTML
*
*
* Hello world!
*
*
* ```
*
* results in this output:
*
* ```HTML
*
*
*
* ```
*
* @param string $htmlContent The HTML content
* @return string The HTML content with all ocurrences replaced
*
* @todo Clarify the meaning of `*[!b38fd2b1fe7f4747d6b1c1254ccd055e]`, which
* looks like a SHA1 hash (does simplehtmldom not support `find('*')`?).
*/
function backgroundToImg($htmlContent) {
$regex = '/background-image[ ]{0,}:[ ]{0,}url\([\'"]{0,}(.*?)[\'"]{0,}\)/';
$htmlContent = str_get_html($htmlContent);
foreach($htmlContent->find('*[!b38fd2b1fe7f4747d6b1c1254ccd055e]') as $element) {
if(preg_match($regex, $element->style, $matches) > 0) {
$element->outertext = '';
}
}
return $htmlContent;
}
/**
* Convert relative links in HTML into absolute links
*
* This function is based on `php-urljoin`.
*
* @link https://github.com/plaidfluff/php-urljoin php-urljoin
*
* @param string|object $content The HTML content. Supports HTML objects or string objects
* @param string $server Fully qualified URL to the page containing relative links
* @return object Content with fixed URLs.
*
* @todo If the input type was a string, this function should return a string as
* well. This is currently done implicitly by how the simplehtmldom object works.
*/
function defaultLinkTo($content, $server){
$string_convert = false;
if (is_string($content)) {
$string_convert = true;
$content = str_get_html($content);
}
foreach($content->find('img') as $image) {
$image->src = urljoin($server, $image->src);
}
foreach($content->find('a') as $anchor) {
$anchor->href = urljoin($server, $anchor->href);
}
if ($string_convert) {
$content = $content->outertext;
}
return $content;
}
/**
* Extract the first part of a string matching the specified start and end delimiters
*
* @param string $string Input string, e.g. `Post author: John Doe
`
* @param string $start Start delimiter, e.g. `author: `
* @param string $end End delimiter, e.g. `<`
* @return string|bool Extracted string, e.g. `John Doe`, or false if the
* delimiters were not found.
*
* @todo This function can possibly be simplified to use a single `substr` command.
*/
function extractFromDelimiters($string, $start, $end) {
if (strpos($string, $start) !== false) {
$section_retrieved = substr($string, strpos($string, $start) + strlen($start));
$section_retrieved = substr($section_retrieved, 0, strpos($section_retrieved, $end));
return $section_retrieved;
} return false;
}
/**
* Remove one or more part(s) of a string using a start and end delmiters
*
* @param string $string Input string, e.g. `foobar`
* @param string $start Start delimiter, e.g. `