<?php
-
/**
* @file src/Content/Text/BBCode.php
*/
use Friendica\Core\Cache;
use Friendica\Core\Config;
use Friendica\Core\L10n;
-use Friendica\Core\PConfig;
use Friendica\Core\Protocol;
use Friendica\Core\System;
use Friendica\Model\Contact;
use Friendica\Util\Map;
use Friendica\Util\Network;
use Friendica\Util\ParseUrl;
-use League\HTMLToMarkdown\HtmlConverter;
-
-require_once "mod/proxy.php";
+use Friendica\Util\Proxy as ProxyUtils;
class BBCode extends BaseObject
{
$post["url"] = $matches[1];
$post["title"] = $matches[2];
}
- if (($post["url"] == "") && (in_array($post["type"], ["link", "video"]))
+ if (!empty($post["url"]) && (in_array($post["type"], ["link", "video"]))
&& preg_match("/\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", $attacheddata, $matches)) {
$post["url"] = $matches[1];
}
*/
public static function toPlaintext($text, $keep_urls = true)
{
- $naked_text = preg_replace('/\[(.+?)\]/','', $text);
+ $naked_text = preg_replace('/\[(.+?)\]\s*/','', $text);
if (!$keep_urls) {
$naked_text = preg_replace('#https?\://[^\s<]+[^\s\.\)]#i', '', $naked_text);
}
return $naked_text;
}
+ private static function proxyUrl($image, $simplehtml = false)
+ {
+ // Only send proxied pictures to API and for internal display
+ if (in_array($simplehtml, [false, 2])) {
+ return ProxyUtils::proxifyUrl($image);
+ } else {
+ return $image;
+ }
+ }
+
public static function scaleExternalImages($srctext, $include_link = true, $scale_replace = false)
{
// Suppress "view full size"
*/
public static function limitBodySize($body)
{
- $maxlen = get_max_import_size();
+ $maxlen = Config::get('config', 'max_import_size', 0);
// If the length of the body, including the embedded images, is smaller
// than the maximum, then don't waste time looking for the images
private static function convertAttachment($return, $simplehtml = false, $tryoembed = true)
{
$data = self::getAttachmentData($return);
- if (!$data) {
+ if (empty($data) || empty($data["url"])) {
return $return;
}
if (isset($data["title"])) {
$data["title"] = strip_tags($data["title"]);
$data["title"] = str_replace(["http://", "https://"], "", $data["title"]);
+ } else {
+ $data["title"] = null;
}
- if (((strpos($data["text"], "[img=") !== false) || (strpos($data["text"], "[img]") !== false) || Config::get('system', 'always_show_preview')) && ($data["image"] != "")) {
+ if (((strpos($data["text"], "[img=") !== false) || (strpos($data["text"], "[img]") !== false) || Config::get('system', 'always_show_preview')) && !empty($data["image"])) {
$data["preview"] = $data["image"];
$data["image"] = "";
}
throw new Exception('OEmbed is disabled for this attachment.');
}
} catch (Exception $e) {
+ $data["title"] = defaults($data, 'title', $data['url']);
+
if ($simplehtml != 4) {
$return = sprintf('<div class="type-%s">', $data["type"]);
}
- if ($data["image"] != "") {
- $return .= sprintf('<a href="%s" target="_blank"><img src="%s" alt="" title="%s" class="attachment-image" /></a><br />', $data["url"], proxy_url($data["image"]), $data["title"]);
- } elseif ($data["preview"] != "") {
- $return .= sprintf('<a href="%s" target="_blank"><img src="%s" alt="" title="%s" class="attachment-preview" /></a><br />', $data["url"], proxy_url($data["preview"]), $data["title"]);
- }
-
- if (($data["type"] == "photo") && ($data["url"] != "") && ($data["image"] != "")) {
- $return .= sprintf('<a href="%s" target="_blank"><img src="%s" alt="" title="%s" class="attachment-image" /></a>', $data["url"], proxy_url($data["image"]), $data["title"]);
- } else {
- $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'], $data['title']);
+ }
}
- if ($data["description"] != "" && $data["description"] != $data["title"]) {
+ if (!empty($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 ($data["type"] == "link") {
+
+ if (!empty($data['url'])) {
$return .= sprintf('<sup><a href="%s">%s</a></sup>', $data['url'], parse_url($data['url'], PHP_URL_HOST));
}
}
}
- return trim($data["text"] . ' ' . $return . ' ' . $data["after"]);
+ return trim(defaults($data, 'text', '') . ' ' . $return . ' ' . defaults($data, 'after', ''));
}
public static function removeShareInformation($Text, $plaintext = false, $nolink = false)
if (!$data) {
return $Text;
} elseif ($nolink) {
- return $data["text"] . $data["after"];
+ return $data["text"] . defaults($data, 'after', '');
}
- $title = htmlentities($data["title"], ENT_QUOTES, 'UTF-8', false);
+ $title = htmlentities(defaults($data, 'title', ''), ENT_QUOTES, 'UTF-8', false);
$text = htmlentities($data["text"], ENT_QUOTES, 'UTF-8', false);
if ($plaintext || (($title != "") && strstr($text, $title))) {
$data["title"] = $data["url"];
}
// If the link already is included in the post, don't add it again
- if (($data["url"] != "") && strpos($data["text"], $data["url"])) {
+ if (!empty($data["url"]) && strpos($data["text"], $data["url"])) {
return $data["text"] . $data["after"];
}
$text = $data["text"];
- if (($data["url"] != "") && ($data["title"] != "")) {
+ if (!empty($data["url"]) && !empty($data["title"])) {
$text .= "\n[url=" . $data["url"] . "]" . $data["title"] . "[/url]";
- } elseif (($data["url"] != "")) {
+ } elseif (!empty($data["url"])) {
$text .= "\n[url]" . $data["url"] . "[/url]";
}
// it loops over the array starting from the first element and going sequentially
// to the last element
$newbody = str_replace('[$#saved_image' . $cnt . '#$]',
- '<img src="' . proxy_url($image) . '" alt="' . L10n::t('Image/photo') . '" />', $newbody);
+ '<img src="' . self::proxyUrl($image) . '" alt="' . L10n::t('Image/photo') . '" />', $newbody);
$cnt++;
}
case 5:
$text = $preshare . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . ' ' . $userid_compact . ": <br />" . $share[3];
break;
- case 6: // app.net
- $text = $preshare . ">> @" . $userid_compact . ": <br />" . $share[3];
- break;
case 7: // statusnet/GNU Social
$text = $preshare . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . " @" . $userid_compact . ": " . $share[3];
break;
case 8: // twitter
$text = $preshare . "RT @" . $userid_compact . ": " . $share[3];
break;
- case 9: // Google+/Facebook
+ case 9: // Google+
$text = $preshare . html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8') . ' ' . $userid_compact . ": <br />" . $share[3];
if ($link != "") {
} else {
$text = trim($share[1]) . "\n";
- $avatar = proxy_url($avatar, false, PROXY_SIZE_THUMB);
+ $avatar = ProxyUtils::proxifyUrl($avatar, false, ProxyUtils::SIZE_THUMB);
$tpl = get_markup_template('shared_content.tpl');
$text .= replace_macros($tpl, [
$ch = @curl_init($match[1]);
@curl_setopt($ch, CURLOPT_NOBODY, true);
@curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
- @curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent());
+ @curl_setopt($ch, CURLOPT_USERAGENT, $a->getUserAgent());
@curl_exec($ch);
$curl_info = @curl_getinfo($ch);
- $a->save_timestamp($stamp1, "network");
+ $a->saveTimestamp($stamp1, "network");
if (substr($curl_info["content_type"], 0, 6) == "image/") {
$text = "[url=" . $match[1] . "]" . $match[1] . "[/url]";
$ch = @curl_init($match[1]);
@curl_setopt($ch, CURLOPT_NOBODY, true);
@curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
- @curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent());
+ @curl_setopt($ch, CURLOPT_USERAGENT, $a->getUserAgent());
@curl_exec($ch);
$curl_info = @curl_getinfo($ch);
- $a->save_timestamp($stamp1, "network");
+ $a->saveTimestamp($stamp1, "network");
// if its a link to a picture then embed this picture
if (substr($curl_info["content_type"], 0, 6) == "image/") {
return $return;
}
- private static function textHighlightCallback($match)
- {
- // Fallback in case the language doesn't exist
- $return = '[code]' . $match[2] . '[/code]';
-
- if (in_array(strtolower($match[1]),
- ['php', 'css', 'mysql', 'sql', 'abap', 'diff', 'html', 'perl', 'ruby',
- 'vbscript', 'avrc', 'dtd', 'java', 'xml', 'cpp', 'python', 'javascript', 'js', 'sh', 'bash'])
- ) {
- $return = text_highlight($match[2], strtolower($match[1]));
- }
-
- return $return;
- }
-
/**
* @brief Converts a BBCode message to HTML message
*
* Simple HTML values meaning:
* - 0: Friendica display
* - 1: Unused
- * - 2: Used for Facebook, Google+, Windows Phone push, Friendica API
+ * - 2: Used for Google+, Windows Phone push, Friendica API
* - 3: Used before converting to Markdown in bb2diaspora.php
* - 4: Used for WordPress, Libertree (before Markdown), pump.io and tumblr
* - 5: Unused
- * - 6: Used for Appnet
+ * - 6: Unused
* - 7: Used for dfrn, OStatus
* - 8: Used for WP backlink text setting
*
return $return;
};
+ // Extracting multi-line code blocks before the whitespace processing
+ $codeblocks = [];
+
+ $text = preg_replace_callback("#\[code(?:=([^\]]*))?\](.*?)\[\/code\]#is",
+ function ($matches) use (&$codeblocks) {
+ $return = $matches[0];
+ if (strpos($matches[2], "\n") !== false) {
+ $return = '#codeblock-' . count($codeblocks) . '#';
+
+ $codeblocks[] = '<pre><code class="language-' . trim($matches[1]) . '">' . trim($matches[2], "\n\r") . '</code></pre>';
+ }
+ return $return;
+ },
+ $text
+ );
+
// Hide all [noparse] contained bbtags by spacefying them
// POSSIBLE BUG --> Will the 'preg' functions crash if there's an embedded image?
$text = preg_replace("/\[share(.*?)avatar\s?=\s?'.*?'\s?(.*?)\]\s?(.*?)\s?\[\/share\]\s?/ism", "\n[share$1$2]$3[/share]", $text);
}
- // Check for [code] text here, before the linefeeds are messed with.
- // The highlighter will unescape and re-escape the content.
- if (strpos($text, '[code=') !== false) {
- $text = preg_replace_callback("/\[code=(.*?)\](.*?)\[\/code\]/ism", 'self::textHighlightCallback', $text);
- }
// Convert new line chars to html <br /> tags
// nlbr seems to be hopelessly messed up
}, $text
);
+ $text = preg_replace_callback(
+ "&\[url=/people\?q\=(.*)\](.*)\[\/url\]&Usi",
+ function ($match) {
+ return "[url=" . System::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";
// [img=widthxheight]image source[/img]
$text = preg_replace_callback(
"/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism",
- function ($matches) {
+ function ($matches) use ($simple_html) {
if (strpos($matches[3], "data:image/") === 0) {
return $matches[0];
}
- $matches[3] = proxy_url($matches[3]);
+ $matches[3] = self::proxyUrl($matches[3], $simple_html);
return "[img=" . $matches[1] . "x" . $matches[2] . "]" . $matches[3] . "[/img]";
},
$text
$text = preg_replace("/\[zmg\=([0-9]*)x([0-9]*)\](.*?)\[\/zmg\]/ism", '<img class="zrl" src="$3" style="width: $1px;" >', $text);
$text = preg_replace_callback("/\[img\=([$URLSearchString]*)\](.*?)\[\/img\]/ism",
- function ($matches) {
- $matches[1] = proxy_url($matches[1]);
+ function ($matches) use ($simple_html) {
+ $matches[1] = self::proxyUrl($matches[1], $simple_html);
$matches[2] = htmlspecialchars($matches[2], ENT_COMPAT);
return '<img src="' . $matches[1] . '" alt="' . $matches[2] . '">';
},
// [img]pathtoimage[/img]
$text = preg_replace_callback(
"/\[img\](.*?)\[\/img\]/ism",
- function ($matches) {
+ function ($matches) use ($simple_html) {
if (strpos($matches[1], "data:image/") === 0) {
return $matches[0];
}
- $matches[1] = proxy_url($matches[1]);
+ $matches[1] = self::proxyUrl($matches[1], $simple_html);
return "[img]" . $matches[1] . "[/img]";
},
$text
// Try to Oembed
if ($try_oembed) {
- $text = preg_replace("/\[video\](.*?\.(ogg|ogv|oga|ogm|webm|mp4))\[\/video\]/ism", '<video src="$1" controls="controls" width="' . $a->videowidth . '" height="' . $a->videoheight . '" loop="true"><a href="$1">$1</a></video>', $text);
- $text = preg_replace("/\[audio\](.*?\.(ogg|ogv|oga|ogm|webm|mp4|mp3))\[\/audio\]/ism", '<audio src="$1" controls="controls"><a href="$1">$1</a></audio>', $text);
+ $text = preg_replace("/\[video\](.*?\.(ogg|ogv|oga|ogm|webm|mp4).*?)\[\/video\]/ism", '<video src="$1" controls="controls" width="' . $a->videowidth . '" height="' . $a->videoheight . '" loop="true"><a href="$1">$1</a></video>', $text);
+ $text = preg_replace("/\[audio\](.*?\.(ogg|ogv|oga|ogm|webm|mp4|mp3).*?)\[\/audio\]/ism", '<audio src="$1" controls="controls"><a href="$1">$1</a></audio>', $text);
$text = preg_replace_callback("/\[video\](.*?)\[\/video\]/ism", $try_oembed_callback, $text);
$text = preg_replace_callback("/\[audio\](.*?)\[\/audio\]/ism", $try_oembed_callback, $text);
} else {
- $text = preg_replace("/\[video\](.*?)\[\/video\]/",
+ $text = preg_replace("/\[video\](.*?)\[\/video\]/ism",
'<a href="$1" target="_blank">$1</a>', $text);
- $text = preg_replace("/\[audio\](.*?)\[\/audio\]/",
+ $text = preg_replace("/\[audio\](.*?)\[\/audio\]/ism",
'<a href="$1" target="_blank">$1</a>', $text);
}
$text = self::interpolateSavedImagesIntoItemBody($text, $saved_image);
}
+ // Restore code blocks
+ $text = preg_replace_callback('/#codeblock-([0-9]+)#/iU',
+ function ($matches) use ($codeblocks) {
+ $return = $matches[0];
+ if (isset($codeblocks[intval($matches[1])])) {
+ $return = $codeblocks[$matches[1]];
+ }
+ return $return;
+ },
+ $text
+ );
+
// Clean up the HTML by loading and saving the HTML with the DOM.
// Bad structured html can break a whole page.
// For performance reasons do it only with ativated item cache or at export.
// Converting images with size parameters to simple images. Markdown doesn't know it.
$text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $text);
- // Extracting multi-line code blocks before the whitespace processing/code highlighter in self::convert()
- $codeblocks = [];
-
- $text = preg_replace_callback("#\[code(?:=([^\]]*))?\](.*?)\[\/code\]#is",
- function ($matches) use (&$codeblocks) {
- $return = $matches[0];
- if (strpos($matches[2], "\n") !== false) {
- $return = '#codeblock-' . count($codeblocks) . '#';
-
- $prefix = '````' . $matches[1] . PHP_EOL;
- $codeblocks[] = $prefix . trim($matches[2]) . PHP_EOL . '````';
- }
- return $return;
- },
- $text
- );
-
// Convert it to HTML - don't try oembed
if ($for_diaspora) {
$text = self::convert($text, false, 3);
$stamp1 = microtime(true);
// Now convert HTML to Markdown
- $converter = new HtmlConverter();
- $text = $converter->convert($text);
+ $text = HTML::toMarkdown($text);
// unmask the special chars back to HTML
$text = str_replace(['&\_lt\_;', '&\_gt\_;', '&\_amp\_;'], ['<', '>', '&'], $text);
- $a->save_timestamp($stamp1, "parser");
+ $a->saveTimestamp($stamp1, "parser");
// Libertree has a problem with escaped hashtags.
$text = str_replace(['\#'], ['#'], $text);
);
}
- // Restore code blocks
- $text = preg_replace_callback('/#codeblock-([0-9]+)#/iU',
- function ($matches) use ($codeblocks) {
- $return = '';
- if (isset($codeblocks[intval($matches[1])])) {
- $return = $codeblocks[$matches[1]];
- }
- return $return;
- },
- $text
- );
-
Addon::callHooks('bb2diaspora', $text);
return $text;