use DOMDocument;
use DOMXPath;
use Exception;
-use Friendica\BaseObject;
use Friendica\Content\OEmbed;
use Friendica\Content\Smilies;
use Friendica\Core\Cache;
use Friendica\Core\Protocol;
use Friendica\Core\Renderer;
use Friendica\Core\System;
+use Friendica\DI;
use Friendica\Model\Contact;
use Friendica\Model\Event;
use Friendica\Model\Photo;
use Friendica\Network\Probe;
use Friendica\Object\Image;
use Friendica\Protocol\Activity;
+use Friendica\Util\Images;
use Friendica\Util\Map;
use Friendica\Util\Network;
use Friendica\Util\ParseUrl;
use Friendica\Util\Strings;
use Friendica\Util\XML;
-class BBCode extends BaseObject
+class BBCode
{
/**
* @brief Fetches attachment data that were generated the old way
if (preg_match("/\[img\](.*?)\[\/img\]/ism", $attacheddata, $matches)) {
- $picturedata = Image::getInfoFromURL($matches[1]);
+ $picturedata = Images::getInfoFromURLCached($matches[1]);
if ($picturedata) {
if (($picturedata[0] >= 500) && ($picturedata[0] >= $picturedata[1])) {
$post['preview'] = $pictures[0][2];
$post['text'] = trim(str_replace($pictures[0][0], '', $body));
} else {
- $imgdata = Image::getInfoFromURL($pictures[0][1]);
+ $imgdata = Images::getInfoFromURLCached($pictures[0][1]);
if ($imgdata && substr($imgdata['mime'], 0, 6) == 'image/') {
$post['type'] = 'photo';
$post['image'] = $pictures[0][1];
$post['url'] = $links[0][1];
}
+ // Simplify "video" element
+ $post['text'] = preg_replace('(\[video.*?\ssrc\s?=\s?([^\s\]]+).*?\].*?\[/video\])ism', '[video]$1[/video]', $post['text']);
+
// Now count the number of external media links
preg_match_all("(\[vimeo\](.*?)\[\/vimeo\])ism", $post['text'], $links1, PREG_SET_ORDER);
preg_match_all("(\[youtube\\](.*?)\[\/youtube\\])ism", $post['text'], $links2, PREG_SET_ORDER);
return $post;
}
+ /**
+ * Remove [attachment] BBCode and replaces it with a regular [url]
+ *
+ * @param string $body
+ * @param boolean $no_link_desc No link description
+ *
+ * @return string with replaced body
+ */
+ public static function removeAttachment($body, $no_link_desc = false)
+ {
+ return preg_replace_callback("/\s*\[attachment (.*)\](.*?)\[\/attachment\]\s*/ism",
+ function ($match) use ($no_link_desc) {
+ $attach_data = self::getAttachmentData($match[0]);
+ if (empty($attach_data['url'])) {
+ return $match[0];
+ } elseif (empty($attach_data['title']) || $no_link_desc) {
+ return "\n[url]" . $attach_data['url'] . "[/url]\n";
+ } else {
+ return "\n[url=" . $attach_data['url'] . ']' . $attach_data['title'] . "[/url]\n";
+ }
+ }, $body);
+ }
+
/**
* @brief Converts a BBCode text into plaintext
*
}
}
- public static function scaleExternalImages($srctext, $include_link = true, $scale_replace = false)
+ public static function scaleExternalImages($srctext)
{
- // Suppress "view full size"
- if (intval(Config::get('system', 'no_view_full_size'))) {
- $include_link = false;
- }
-
- // Picture addresses can contain special characters
- $s = htmlspecialchars_decode($srctext);
+ $s = $srctext;
$matches = null;
$c = preg_match_all('/\[img.*?\](.*?)\[\/img\]/ism', $s, $matches, PREG_SET_ORDER);
foreach ($matches as $mtch) {
Logger::log('scale_external_image: ' . $mtch[1]);
- $hostname = str_replace('www.', '', substr(System::baseUrl(), strpos(System::baseUrl(), '://') + 3));
+ $hostname = str_replace('www.', '', substr(DI::baseUrl(), strpos(DI::baseUrl(), '://') + 3));
if (stristr($mtch[1], $hostname)) {
continue;
}
- // $scale_replace, if passed, is an array of two elements. The
- // first is the name of the full-size image. The second is the
- // name of a remote, scaled-down version of the full size image.
- // This allows Friendica to display the smaller remote image if
- // one exists, while still linking to the full-size image
- if ($scale_replace) {
- $scaled = str_replace($scale_replace[0], $scale_replace[1], $mtch[1]);
- } else {
- $scaled = $mtch[1];
- }
- $i = Network::fetchUrl($scaled);
+ $i = Network::fetchUrl($mtch[1]);
if (!$i) {
return $srctext;
}
// guess mimetype from headers or filename
- $type = Image::guessType($mtch[1], true);
+ $type = Images::guessType($mtch[1], true);
if ($i) {
$Image = new Image($i, $type);
Logger::log('scale_external_images: ' . $orig_width . '->' . $new_width . 'w ' . $orig_height . '->' . $new_height . 'h' . ' match: ' . $mtch[0], Logger::DEBUG);
$s = str_replace(
$mtch[0],
- '[img=' . $new_width . 'x' . $new_height. ']' . $scaled . '[/img]'
- . "\n" . (($include_link)
- ? '[url=' . $mtch[1] . ']' . L10n::t('view full size') . '[/url]' . "\n"
- : ''),
+ '[img=' . $new_width . 'x' . $new_height. ']' . $mtch[1] . '[/img]'
+ . "\n",
$s
);
Logger::log('scale_external_images: new string: ' . $s, Logger::DEBUG);
}
}
- // replace the special char encoding
- $s = htmlspecialchars($s, ENT_NOQUOTES, 'UTF-8');
return $s;
}
* Note: Can produce a [bookmark] tag in the returned string
*
* @brief Processes [attachment] tags
- * @param string $return
+ * @param string $text
* @param bool|int $simplehtml
* @param bool $tryoembed
* @return string
* @throws \Friendica\Network\HTTPException\InternalServerErrorException
*/
- private static function convertAttachment($return, $simplehtml = false, $tryoembed = true)
+ private static function convertAttachment($text, $simplehtml = false, $tryoembed = true)
{
- $data = self::getAttachmentData($return);
+ $data = self::getAttachmentData($text);
if (empty($data) || empty($data['url'])) {
- return $return;
+ return $text;
}
if (isset($data['title'])) {
}
$return = '';
- if (in_array($simplehtml, [7, 9])) {
- $return = self::convertUrlForActivityPub($data['url']);
- } elseif (($simplehtml != 4) && ($simplehtml != 0)) {
- $return = sprintf('<a href="%s" target="_blank">%s</a><br>', $data['url'], $data['title']);
- } else {
- try {
- if ($tryoembed && OEmbed::isAllowedURL($data['url'])) {
- $return = OEmbed::getHTML($data['url'], $data['title']);
- } else {
- throw new Exception('OEmbed is disabled for this attachment.');
- }
- } catch (Exception $e) {
- $data['title'] = ($data['title'] ?? '') ?: $data['url'];
+ try {
+ if ($tryoembed && OEmbed::isAllowedURL($data['url'])) {
+ $return = OEmbed::getHTML($data['url'], $data['title']);
+ } else {
+ throw new Exception('OEmbed is disabled for this attachment.');
+ }
+ } catch (Exception $e) {
+ $data['title'] = ($data['title'] ?? '') ?: $data['url'];
- if ($simplehtml != 4) {
- $return = sprintf('<div class="type-%s">', $data['type']);
- }
+ if ($simplehtml != 4) {
+ $return = sprintf('<div class="type-%s">', $data['type']);
+ }
- if (!empty($data['title']) && !empty($data['url'])) {
- if (!empty($data['image']) && empty($data['text']) && ($data['type'] == 'photo')) {
- $return .= sprintf('<a href="%s" target="_blank"><img src="%s" alt="" title="%s" class="attachment-image" /></a>', $data['url'], self::proxyUrl($data['image'], $simplehtml), $data['title']);
- } else {
- if (!empty($data['image'])) {
- $return .= sprintf('<a href="%s" target="_blank"><img src="%s" alt="" title="%s" class="attachment-image" /></a><br />', $data['url'], self::proxyUrl($data['image'], $simplehtml), $data['title']);
- } elseif (!empty($data['preview'])) {
- $return .= sprintf('<a href="%s" target="_blank"><img src="%s" alt="" title="%s" class="attachment-preview" /></a><br />', $data['url'], self::proxyUrl($data['preview'], $simplehtml), $data['title']);
- }
- $return .= sprintf('<h4><a href="%s">%s</a></h4>', $data['url'], $data['title']);
+ if (!empty($data['title']) && !empty($data['url'])) {
+ if (!empty($data['image']) && empty($data['text']) && ($data['type'] == 'photo')) {
+ $return .= sprintf('<a href="%s" target="_blank"><img src="%s" alt="" title="%s" class="attachment-image" /></a>', $data['url'], self::proxyUrl($data['image'], $simplehtml), $data['title']);
+ } else {
+ if (!empty($data['image'])) {
+ $return .= sprintf('<a href="%s" target="_blank"><img src="%s" alt="" title="%s" class="attachment-image" /></a><br />', $data['url'], self::proxyUrl($data['image'], $simplehtml), $data['title']);
+ } elseif (!empty($data['preview'])) {
+ $return .= sprintf('<a href="%s" target="_blank"><img src="%s" alt="" title="%s" class="attachment-preview" /></a><br />', $data['url'], self::proxyUrl($data['preview'], $simplehtml), $data['title']);
}
+ $return .= sprintf('<h4><a href="%s">%s</a></h4>', $data['url'], $data['title']);
}
+ }
- if (!empty($data['description']) && $data['description'] != $data['title']) {
- // Sanitize the HTML by converting it to BBCode
- $bbcode = HTML::toBBCode($data['description']);
- $return .= sprintf('<blockquote>%s</blockquote>', trim(self::convert($bbcode)));
- }
+ if (!empty($data['description']) && $data['description'] != $data['title']) {
+ // Sanitize the HTML by converting it to BBCode
+ $bbcode = HTML::toBBCode($data['description']);
+ $return .= sprintf('<blockquote>%s</blockquote>', trim(self::convert($bbcode)));
+ }
- if (!empty($data['url'])) {
- $return .= sprintf('<sup><a href="%s">%s</a></sup>', $data['url'], parse_url($data['url'], PHP_URL_HOST));
- }
+ if (!empty($data['url'])) {
+ $return .= sprintf('<sup><a href="%s">%s</a></sup>', $data['url'], parse_url($data['url'], PHP_URL_HOST));
+ }
- if ($simplehtml != 4) {
- $return .= '</div>';
- }
+ if ($simplehtml != 4) {
+ $return .= '</div>';
}
}
public static function convertShare($text, callable $callback)
{
$return = preg_replace_callback(
- "/(.*?)\[share(.*?)\](.*?)\[\/share\]/ism",
+ "/(.*?)\[share(.*?)\](.*)\[\/share\]/ism",
function ($match) use ($callback) {
$attribute_string = $match[2];
-
$attributes = [];
- foreach(['author', 'profile', 'avatar', 'link', 'posted'] as $field) {
+ foreach (['author', 'profile', 'avatar', 'link', 'posted'] as $field) {
preg_match("/$field=(['\"])(.+?)\\1/ism", $attribute_string, $matches);
$attributes[$field] = html_entity_decode($matches[2] ?? '', ENT_QUOTES, 'UTF-8');
}
$text = ($is_quote_share? '<br />' : '') . '<p>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' ' . $author_contact['addr'] . ': </p>' . "\n" . $content;
break;
case 7: // statusnet/GNU Social
- case 9: // ActivityPub
$text = ($is_quote_share? '<br />' : '') . '<p>' . html_entity_decode('♲ ', ENT_QUOTES, 'UTF-8') . ' @' . $author_contact['addr'] . ': ' . $content . '</p>' . "\n";
break;
+ case 9: // ActivityPub
+ $author = '@<span class="vcard"><a href="' . $author_contact['url'] . '" class="url u-url mention" title="' . $author_contact['addr'] . '"><span class="fn nickname mention">' . $author_contact['addr'] . '</span></a>:</span>';
+ $text = '<div><a href="' . $attributes['link'] . '">' . html_entity_decode('♲', ENT_QUOTES, 'UTF-8') . '</a> ' . $author . '<blockquote>' . $content . '</blockquote></div>' . "\n";
+ break;
default:
// Transforms quoted tweets in rich attachments to avoid nested tweets
if (stripos(Strings::normaliseLink($attributes['link']), 'http://twitter.com/') === 0 && OEmbed::isAllowedURL($attributes['link'])) {
$text = Cache::get($cache_key);
if (is_null($text)) {
- $a = self::getApp();
+ $a = DI::app();
$stamp1 = microtime(true);
@curl_exec($ch);
$curl_info = @curl_getinfo($ch);
- $a->getProfiler()->saveTimestamp($stamp1, "network", System::callstack());
+ DI::profiler()->saveTimestamp($stamp1, "network", System::callstack());
if (substr($curl_info['content_type'], 0, 6) == 'image/') {
$text = "[url=" . $match[1] . ']' . $match[1] . "[/url]";
private static function cleanPictureLinksCallback($match)
{
- $a = self::getApp();
+ $a = DI::app();
// When the picture link is the own photo path then we can avoid fetching the link
- $own_photo_url = preg_quote(Strings::normaliseLink($a->getBaseURL()) . '/photos/');
+ $own_photo_url = preg_quote(Strings::normaliseLink(DI::baseUrl()->get()) . '/photos/');
if (preg_match('|' . $own_photo_url . '.*?/image/|', Strings::normaliseLink($match[1]))) {
if (!empty($match[3])) {
$text = '[img=' . str_replace('-1.', '-0.', $match[2]) . ']' . $match[3] . '[/img]';
@curl_exec($ch);
$curl_info = @curl_getinfo($ch);
- $a->getProfiler()->saveTimestamp($stamp1, "network", System::callstack());
+ DI::profiler()->saveTimestamp($stamp1, "network", System::callstack());
// if its a link to a picture then embed this picture
if (substr($curl_info['content_type'], 0, 6) == 'image/') {
* - 5: Unused
* - 6: Unused
* - 7: Used for dfrn, OStatus
- * - 8: Used for WP backlink text setting
+ * - 8: Used for Twitter, WP backlink text setting
* - 9: ActivityPub
*
* @param string $text
*/
public static function convert($text, $try_oembed = true, $simple_html = 0, $for_plaintext = false)
{
- $a = self::getApp();
+ $a = DI::app();
/*
* preg_match_callback function to replace potential Oembed tags with Oembed content
function ($matches) use (&$codeblocks) {
$return = '#codeblock-' . count($codeblocks) . '#';
if (strpos($matches[2], "\n") !== false) {
- $codeblocks[] = '<pre><code class="language-' . trim($matches[1]) . '">' . trim($matches[2], "\n\r") . '</code></pre>';
+ $codeblocks[] = '<pre><code class="language-' . trim($matches[1]) . '">' . htmlspecialchars(trim($matches[2], "\n\r"), ENT_NOQUOTES, 'UTF-8') . '</code></pre>';
} else {
- $codeblocks[] = '<code>' . $matches[2] . '</code>';
+ $codeblocks[] = '<code>' . htmlspecialchars($matches[2], ENT_NOQUOTES, 'UTF-8') . '</code>';
}
return $return;
} while ($oldtext != $text);
}
+ /// @todo Have a closer look at the different html modes
// Handle attached links or videos
- $text = self::convertAttachment($text, $simple_html, $try_oembed);
+ if (in_array($simple_html, [9])) {
+ $text = self::removeAttachment($text);
+ } elseif (!in_array($simple_html, [0, 4])) {
+ $text = self::removeAttachment($text, true);
+ } else {
+ $text = self::convertAttachment($text, $simple_html, $try_oembed);
+ }
// leave open the posibility of [map=something]
// this is replaced in Item::prepareBody() which has knowledge of the item location
$text = str_replace('[hr]', '<hr />', $text);
if (!$for_plaintext) {
+ $escaped = [];
+
+ // Escaping BBCodes susceptible to contain rogue URL we don'' want the autolinker to catch
+ $text = preg_replace_callback('#\[(url|img|audio|video|youtube|vimeo|share|attachment|iframe|bookmark).+?\[/\1\]#ism',
+ function ($matches) use (&$escaped) {
+ $return = '{escaped-' . count($escaped) . '}';
+ $escaped[] = $matches[0];
+
+ return $return;
+ },
+ $text
+ );
+
// Autolinker for isolated URLs
$text = preg_replace(Strings::autoLinkRegEx(), '[url]$1[/url]', $text);
+
+ // Restoring escaped blocks
+ $text = preg_replace_callback('/{escaped-([0-9]+)}/iU',
+ function ($matches) use ($escaped) {
+ return $escaped[intval($matches[1])] ?? $matches[0];
+ },
+ $text
+ );
}
// This is actually executed in Item::prepareBody()
$text = preg_replace("/\[img\](.*?)\[\/img\]/ism", '<img src="$1" alt="' . L10n::t('Image/photo') . '" />', $text);
$text = preg_replace("/\[zmg\](.*?)\[\/zmg\]/ism", '<img src="$1" alt="' . L10n::t('Image/photo') . '" />', $text);
- $text = preg_replace("/\[crypt\](.*?)\[\/crypt\]/ism", '<br/><img src="' .System::baseUrl() . '/images/lock_icon.gif" alt="' . L10n::t('Encrypted content') . '" title="' . L10n::t('Encrypted content') . '" /><br />', $text);
- $text = preg_replace("/\[crypt(.*?)\](.*?)\[\/crypt\]/ism", '<br/><img src="' .System::baseUrl() . '/images/lock_icon.gif" alt="' . L10n::t('Encrypted content') . '" title="' . '$1' . ' ' . L10n::t('Encrypted content') . '" /><br />', $text);
- //$Text = preg_replace("/\[crypt=(.*?)\](.*?)\[\/crypt\]/ism", '<br/><img src="' .System::baseUrl() . '/images/lock_icon.gif" alt="' . L10n::t('Encrypted content') . '" title="' . '$1' . ' ' . L10n::t('Encrypted content') . '" /><br />', $Text);
+ $text = preg_replace("/\[crypt\](.*?)\[\/crypt\]/ism", '<br/><img src="' .DI::baseUrl() . '/images/lock_icon.gif" alt="' . L10n::t('Encrypted content') . '" title="' . L10n::t('Encrypted content') . '" /><br />', $text);
+ $text = preg_replace("/\[crypt(.*?)\](.*?)\[\/crypt\]/ism", '<br/><img src="' .DI::baseUrl() . '/images/lock_icon.gif" alt="' . L10n::t('Encrypted content') . '" title="' . '$1' . ' ' . L10n::t('Encrypted content') . '" /><br />', $text);
+ //$Text = preg_replace("/\[crypt=(.*?)\](.*?)\[\/crypt\]/ism", '<br/><img src="' .DI::baseUrl() . '/images/lock_icon.gif" alt="' . L10n::t('Encrypted content') . '" title="' . '$1' . ' ' . L10n::t('Encrypted content') . '" /><br />', $Text);
+
+ // Simplify "video" element
+ $text = preg_replace('(\[video.*?\ssrc\s?=\s?([^\s\]]+).*?\].*?\[/video\])ism', '[video]$1[/video]', $text);
// Try to Oembed
if ($try_oembed) {
$text = str_replace(["\r","\n"], ['<br />', '<br />'], $text);
// Remove all hashtag addresses
- if ((!$try_oembed || $simple_html) && !in_array($simple_html, [3, 7, 9])) {
+ if ($simple_html && !in_array($simple_html, [3, 7, 9])) {
$text = preg_replace("/([#@!])\[url\=(.*?)\](.*?)\[\/url\]/ism", '$1$3', $text);
} elseif ($simple_html == 3) {
// The ! is converted to @ since Diaspora only understands the @
$text = preg_replace_callback(
"&\[url=/?posts/([^\[\]]*)\](.*)\[\/url\]&Usi",
function ($match) {
- return "[url=" . System::baseUrl() . "/display/" . $match[1] . "]" . $match[2] . "[/url]";
+ return "[url=" . DI::baseUrl() . "/display/" . $match[1] . "]" . $match[2] . "[/url]";
}, $text
);
$text = preg_replace_callback(
"&\[url=/people\?q\=(.*)\](.*)\[\/url\]&Usi",
function ($match) {
- return "[url=" . System::baseUrl() . "/search?search=%40" . $match[1] . "]" . $match[2] . "[/url]";
+ return "[url=" . DI::baseUrl() . "/search?search=%40" . $match[1] . "]" . $match[2] . "[/url]";
}, $text
);
// Server independent link to posts and comments
// See issue: https://github.com/diaspora/diaspora_federation/issues/75
$expression = "=diaspora://.*?/post/([0-9A-Za-z\-_@.:]{15,254}[0-9A-Za-z])=ism";
- $text = preg_replace($expression, System::baseUrl()."/display/$1", $text);
+ $text = preg_replace($expression, DI::baseUrl()."/display/$1", $text);
/* Tag conversion
* Supports:
* - #[url=<anything>]<term>[/url]
* - [url=<anything>]#<term>[/url]
*/
- $text = preg_replace_callback("/(?:#\[url\=.*?\]|\[url\=.*?\]#)(.*?)\[\/url\]/ism", function($matches) {
+ $text = preg_replace_callback("/(?:#\[url\=[^\[\]]*\]|\[url\=[^\[\]]*\]#)(.*?)\[\/url\]/ism", function($matches) {
return '#<a href="'
- . System::baseUrl() . '/search?tag=' . rawurlencode($matches[1])
- . '" class="tag" title="' . XML::escape($matches[1]) . '">'
+ . DI::baseUrl() . '/search?tag=' . rawurlencode($matches[1])
+ . '" class="tag" rel="tag" title="' . XML::escape($matches[1]) . '">'
. XML::escape($matches[1])
. '</a>';
}, $text);
// We need no target="_blank" for local links
- // convert links start with System::baseUrl() as local link without the target="_blank" attribute
- $escapedBaseUrl = preg_quote(System::baseUrl(), '/');
+ // convert links start with DI::baseUrl() as local link without the target="_blank" attribute
+ $escapedBaseUrl = preg_quote(DI::baseUrl(), '/');
$text = preg_replace("/\[url\](".$escapedBaseUrl.".*?)\[\/url\]/ism", '<a href="$1">$1</a>', $text);
$text = preg_replace("/\[url\=(".$escapedBaseUrl.".*?)\](.*?)\[\/url\]/ism", '<a href="$1">$2</a>', $text);
// we may need to restrict this further if it picks up too many strays
// link acct:user@host to a webfinger profile redirector
- $text = preg_replace('/acct:([^@]+)@((?!\-)(?:[a-zA-Z\d\-]{0,62}[a-zA-Z\d]\.){1,126}(?!\d+)[a-zA-Z\d]{1,63})/', '<a href="' . System::baseUrl() . '/acctlink?addr=$1@$2" target="extlink">acct:$1@$2</a>', $text);
+ $text = preg_replace('/acct:([^@]+)@((?!\-)(?:[a-zA-Z\d\-]{0,62}[a-zA-Z\d]\.){1,126}(?!\d+)[a-zA-Z\d]{1,63})/', '<a href="' . DI::baseUrl() . '/acctlink?addr=$1@$2" target="extlink">acct:$1@$2</a>', $text);
// Perform MAIL Search
$text = preg_replace("/\[mail\](.*?)\[\/mail\]/", '<a href="mailto:$1">$1</a>', $text);
*/
public static function toMarkdown($text, $for_diaspora = true)
{
- $a = self::getApp();
+ $a = DI::app();
$original_text = $text;
$text = self::convert($text, false, 4);
}
- // mask some special HTML chars from conversation to markdown
- $text = str_replace(['<', '>', '&'], ['&_lt_;', '&_gt_;', '&_amp_;'], $text);
-
// If a link is followed by a quote then there should be a newline before it
// Maybe we should make this newline at every time before a quote.
$text = str_replace(['</a><blockquote>'], ['</a><br><blockquote>'], $text);
// Now convert HTML to Markdown
$text = HTML::toMarkdown($text);
- // unmask the special chars back to HTML
- $text = str_replace(['&\_lt\_;', '&\_gt\_;', '&\_amp\_;'], ['<', '>', '&'], $text);
-
- $a->getProfiler()->saveTimestamp($stamp1, "parser", System::callstack());
+ DI::profiler()->saveTimestamp($stamp1, "parser", System::callstack());
// Libertree has a problem with escaped hashtags.
$text = str_replace(['\#'], ['#'], $text);