X-Git-Url: https://git.mxchange.org/?a=blobdiff_plain;f=src%2FContent%2FText%2FBBCode.php;h=6c441bac6fb152621a948c8a98da59cee72b4020;hb=a66e9b81bafdb94fcb5405818f721c3a0dc2a418;hp=13d4e1b055f66b71c1e369fa5e6151819aee47b8;hpb=c845415a99ebc348103815a7b2c55b15c75cdd24;p=friendica.git
diff --git a/src/Content/Text/BBCode.php b/src/Content/Text/BBCode.php
index 13d4e1b055..6c441bac6f 100644
--- a/src/Content/Text/BBCode.php
+++ b/src/Content/Text/BBCode.php
@@ -1,34 +1,36 @@
= 500) && ($picturedata[0] >= $picturedata[1])) {
- $post["image"] = $matches[1];
- } else {
- $post["preview"] = $matches[1];
+ if ($picturedata) {
+ if (($picturedata[0] >= 500) && ($picturedata[0] >= $picturedata[1])) {
+ $post["image"] = $matches[1];
+ } else {
+ $post["preview"] = $matches[1];
+ }
}
}
@@ -174,7 +178,7 @@ class BBCode
}
if ($title != "") {
- $title = BBCode::convert(html_entity_decode($title, ENT_QUOTES, 'UTF-8'), false, true);
+ $title = self::convert(html_entity_decode($title, ENT_QUOTES, 'UTF-8'), false, true);
$title = html_entity_decode($title, ENT_QUOTES, 'UTF-8');
$title = str_replace(["[", "]"], ["[", "]"], $title);
$data["title"] = $title;
@@ -239,6 +243,9 @@ class BBCode
$body = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $body);
$URLSearchString = "^\[\]";
+
+ $body = preg_replace("/\[img\=([$URLSearchString]*)\](.*?)\[\/img\]/ism", '[img]$1[/img]', $body);
+
if (preg_match_all("(\[url=([$URLSearchString]*)\]\s*\[img\]([$URLSearchString]*)\[\/img\]\s*\[\/url\])ism", $body, $pictures, PREG_SET_ORDER)) {
if ((count($pictures) == 1) && !$has_title) {
// Checking, if the link goes to a picture
@@ -264,7 +271,7 @@ class BBCode
$post["text"] = str_replace($pictures[0][0], "", $body);
} else {
$imgdata = Image::getInfoFromURL($pictures[0][1]);
- if (substr($imgdata["mime"], 0, 6) == "image/") {
+ if ($imgdata && substr($imgdata["mime"], 0, 6) == "image/") {
$post["type"] = "photo";
$post["image"] = $pictures[0][1];
$post["preview"] = $pictures[0][2];
@@ -336,159 +343,20 @@ class BBCode
}
/**
- * @brief Convert a message into plaintext for connectors to other networks
+ * @brief Converts a BBCode text into plaintext
*
- * @param array $b The message array that is about to be posted
- * @param int $limit The maximum number of characters when posting to that network
- * @param bool $includedlinks Has an attached link to be included into the message?
- * @param int $htmlmode This triggers the behaviour of the bbcode conversion
- * @param string $target_network Name of the network where the post should go to.
+ * @param bool $keep_urls Whether to keep URLs in the resulting plaintext
*
- * @return string The converted message
+ * @return string
*/
- public static function toPlaintext($b, $limit = 0, $includedlinks = false, $htmlmode = 2, $target_network = "")
+ public static function toPlaintext($text, $keep_urls = true)
{
- // Remove the hash tags
- $URLSearchString = "^\[\]";
- $body = preg_replace("/([#@])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '$1$3', $b["body"]);
-
- // Add an URL element if the text contains a raw link
- $body = preg_replace("/([^\]\='".'"'."]|^)(https?\:\/\/[a-zA-Z0-9\:\/\-\?\&\;\.\=\_\~\#\%\$\!\+\,]+)/ism", '$1[url]$2[/url]', $body);
-
- // Remove the abstract
- $body = self::stripAbstract($body);
-
- // At first look at data that is attached via "type-..." stuff
- // This will hopefully replaced with a dedicated bbcode later
- //$post = self::getAttachedData($b["body"]);
- $post = self::getAttachedData($body, $b);
-
- if (($b["title"] != "") && ($post["text"] != "")) {
- $post["text"] = trim($b["title"]."\n\n".$post["text"]);
- } elseif ($b["title"] != "") {
- $post["text"] = trim($b["title"]);
- }
-
- $abstract = "";
-
- // Fetch the abstract from the given target network
- if ($target_network != "") {
- $default_abstract = self::getAbstract($b["body"]);
- $abstract = self::getAbstract($b["body"], $target_network);
-
- // If we post to a network with no limit we only fetch
- // an abstract exactly for this network
- if (($limit == 0) && ($abstract == $default_abstract)) {
- $abstract = "";
- }
- } else {// Try to guess the correct target network
- switch ($htmlmode) {
- case 8:
- $abstract = self::getAbstract($b["body"], NETWORK_TWITTER);
- break;
- case 7:
- $abstract = self::getAbstract($b["body"], NETWORK_STATUSNET);
- break;
- case 6:
- $abstract = self::getAbstract($b["body"], NETWORK_APPNET);
- break;
- default: // We don't know the exact target.
- // We fetch an abstract since there is a posting limit.
- if ($limit > 0) {
- $abstract = self::getAbstract($b["body"]);
- }
- }
- }
-
- if ($abstract != "") {
- $post["text"] = $abstract;
-
- if ($post["type"] == "text") {
- $post["type"] = "link";
- $post["url"] = $b["plink"];
- }
+ $naked_text = preg_replace('/\[(.+?)\]/','', $text);
+ if (!$keep_urls) {
+ $naked_text = preg_replace('#https?\://[^\s<]+[^\s\.\)]#i', '', $naked_text);
}
- $html = BBCode::convert($post["text"].$post["after"], false, $htmlmode);
- $msg = html2plain($html, 0, true);
- $msg = trim(html_entity_decode($msg, ENT_QUOTES, 'UTF-8'));
-
- $link = "";
- if ($includedlinks) {
- if ($post["type"] == "link") {
- $link = $post["url"];
- } elseif ($post["type"] == "text") {
- $link = $post["url"];
- } elseif ($post["type"] == "video") {
- $link = $post["url"];
- } elseif ($post["type"] == "photo") {
- $link = $post["image"];
- }
-
- if (($msg == "") && isset($post["title"])) {
- $msg = trim($post["title"]);
- }
-
- if (($msg == "") && isset($post["description"])) {
- $msg = trim($post["description"]);
- }
-
- // If the link is already contained in the post, then it neeedn't to be added again
- // But: if the link is beyond the limit, then it has to be added.
- if (($link != "") && strstr($msg, $link)) {
- $pos = strpos($msg, $link);
-
- // Will the text be shortened in the link?
- // Or is the link the last item in the post?
- if (($limit > 0) && ($pos < $limit) && (($pos + 23 > $limit) || ($pos + strlen($link) == strlen($msg)))) {
- $msg = trim(str_replace($link, "", $msg));
- } elseif (($limit == 0) || ($pos < $limit)) {
- // The limit has to be increased since it will be shortened - but not now
- // Only do it with Twitter (htmlmode = 8)
- if (($limit > 0) && (strlen($link) > 23) && ($htmlmode == 8)) {
- $limit = $limit - 23 + strlen($link);
- }
-
- $link = "";
-
- if ($post["type"] == "text") {
- unset($post["url"]);
- }
- }
- }
- }
-
- if ($limit > 0) {
- // Reduce multiple spaces
- // When posted to a network with limited space, we try to gain space where possible
- while (strpos($msg, " ") !== false) {
- $msg = str_replace(" ", " ", $msg);
- }
-
- // Twitter is using its own limiter, so we always assume that shortened links will have this length
- if (iconv_strlen($link, "UTF-8") > 0) {
- $limit = $limit - 23;
- }
-
- if (iconv_strlen($msg, "UTF-8") > $limit) {
- if (($post["type"] == "text") && isset($post["url"])) {
- $post["url"] = $b["plink"];
- } elseif (!isset($post["url"])) {
- $limit = $limit - 23;
- $post["url"] = $b["plink"];
- // Which purpose has this line? It is now uncommented, but left as a reminder
- //} elseif (strpos($b["body"], "[share") !== false) {
- // $post["url"] = $b["plink"];
- } elseif (PConfig::get($b["uid"], "system", "no_intelligent_shortening")) {
- $post["url"] = $b["plink"];
- }
- $msg = Plaintext::shorten($msg, $limit);
- }
- }
-
- $post["text"] = trim($msg);
-
- return($post);
+ return $naked_text;
}
public static function scaleExternalImages($srctext, $include_link = true, $scale_replace = false)
@@ -707,7 +575,7 @@ class BBCode
if ($data["description"] != "" && $data["description"] != $data["title"]) {
// Sanitize the HTML by converting it to BBCode
- $bbcode = html2bbcode($data["description"]);
+ $bbcode = HTML::toBBCode($data["description"]);
$return .= sprintf('%s
', trim(self::convert($bbcode)));
}
if ($data["type"] == "link") {
@@ -762,27 +630,6 @@ class BBCode
return $text . "\n" . $data["after"];
}
- private static function cleanCss($input)
- {
- $cleaned = "";
-
- $input = strtolower($input);
-
- for ($i = 0; $i < strlen($input); $i++) {
- $char = substr($input, $i, 1);
-
- if (($char >= "a") && ($char <= "z")) {
- $cleaned .= $char;
- }
-
- if (!(strpos(" #;:0123456789-_.%", $char) === false)) {
- $cleaned .= $char;
- }
- }
-
- return $cleaned;
- }
-
/**
* Converts [url] BBCodes in a format that looks fine on Mastodon. (callback function)
*
@@ -1197,7 +1044,7 @@ class BBCode
$text = Cache::get($match[1]);
if (is_null($text)) {
- $a = get_app();
+ $a = self::getApp();
$stamp1 = microtime(true);
@@ -1220,7 +1067,7 @@ class BBCode
$doc = new DOMDocument();
@$doc->loadHTML($body);
- $xpath = new DomXPath($doc);
+ $xpath = new DOMXPath($doc);
$list = $xpath->query("//meta[@name]");
foreach ($list as $node) {
$attr = [];
@@ -1256,7 +1103,7 @@ class BBCode
$text = Cache::get($match[1]);
if (is_null($text)) {
- $a = get_app();
+ $a = self::getApp();
$stamp1 = microtime(true);
@@ -1280,7 +1127,7 @@ class BBCode
$doc = new DOMDocument();
@$doc->loadHTML($body);
- $xpath = new DomXPath($doc);
+ $xpath = new DOMXPath($doc);
$list = $xpath->query("//meta[@name]");
foreach ($list as $node) {
$attr = [];
@@ -1309,13 +1156,17 @@ class BBCode
private static function textHighlightCallback($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'])
+ 'vbscript', 'avrc', 'dtd', 'java', 'xml', 'cpp', 'python', 'javascript', 'js', 'sh', 'bash'])
) {
- return text_highlight($match[2], strtolower($match[1]));
+ $return = text_highlight($match[2], strtolower($match[1]));
}
- return $match[0];
+
+ return $return;
}
/**
@@ -1343,7 +1194,7 @@ class BBCode
*/
public static function convert($text, $try_oembed = true, $simple_html = false, $for_plaintext = false)
{
- $a = get_app();
+ $a = self::getApp();
/*
* preg_match_callback function to replace potential Oembed tags with Oembed content
@@ -1392,7 +1243,7 @@ class BBCode
// After we're finished processing the bbcode we'll
// replace all of the event code with a reformatted version.
- $ev = bbtoevent($text);
+ $ev = Event::fromBBCode($text);
// Replace any html brackets with HTML Entities to prevent executing HTML or script
// Don't use strip_tags here because it breaks [url] search by replacing & with amp
@@ -1593,7 +1444,7 @@ class BBCode
$text = preg_replace("(\[u\](.*?)\[\/u\])ism", '$1', $text);
// Check for strike-through text
- $text = preg_replace("(\[s\](.*?)\[\/s\])ism", '$1', $text);
+ $text = preg_replace("(\[s\](.*?)\[\/s\])ism", '$1', $text);
// Check for over-line text
$text = preg_replace("(\[o\](.*?)\[\/o\])ism", '$1', $text);
@@ -1616,7 +1467,7 @@ class BBCode
$text = preg_replace_callback(
"(\[style=(.*?)\](.*?)\[\/style\])ism",
function ($match) {
- return "" . $match[2] . "";
+ return "" . $match[2] . "";
},
$text
);
@@ -1625,7 +1476,7 @@ class BBCode
$text = preg_replace_callback(
"(\[class=(.*?)\](.*?)\[\/class\])ism",
function ($match) {
- return "" . $match[2] . "";
+ return "" . $match[2] . "";
},
$text
);
@@ -1734,6 +1585,14 @@ class BBCode
$text = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '
', $text);
$text = preg_replace("/\[zmg\=([0-9]*)x([0-9]*)\](.*?)\[\/zmg\]/ism", '', $text);
+ $text = preg_replace_callback("/\[img\=([$URLSearchString]*)\](.*?)\[\/img\]/ism",
+ function ($matches) {
+ $matches[1] = proxy_url($matches[1]);
+ $matches[2] = htmlspecialchars($matches[2], ENT_COMPAT);
+ return '![' . $matches[2] . '](' . $matches[1] . ')
';
+ },
+ $text);
+
// Images
// [img]pathtoimage[/img]
$text = preg_replace_callback(
@@ -1830,7 +1689,7 @@ class BBCode
// start which is always required). Allow desc with a missing summary for compatibility.
if ((x($ev, 'desc') || x($ev, 'summary')) && x($ev, 'start')) {
- $sub = format_event_html($ev, $simple_html);
+ $sub = Event::getHTML($ev, $simple_html);
$text = preg_replace("/\[event\-summary\](.*?)\[\/event\-summary\]/ism", '', $text);
$text = preg_replace("/\[event\-description\](.*?)\[\/event\-description\]/ism", '', $text);
@@ -1865,10 +1724,12 @@ class BBCode
$text = preg_replace_callback("/\[nobb\](.*?)\[\/nobb\]/ism", 'self::unescapeNoparseCallback', $text);
$text = preg_replace_callback("/\[pre\](.*?)\[\/pre\]/ism", 'self::unescapeNoparseCallback', $text);
-
+ /// @todo What is the meaning of these lines?
$text = preg_replace('/\[\&\;([#a-z0-9]+)\;\]/', '&$1;', $text);
$text = preg_replace('/\&\#039\;/', '\'', $text);
- $text = preg_replace('/\"\;/', '"', $text);
+
+ // Currently deactivated, it made problems with " inside of alt texts.
+ //$text = preg_replace('/\"\;/', '"', $text);
// fix any escaped ampersands that may have been converted into links
$text = preg_replace('/\<([^>]*?)(src|href)=(.*?)\&\;(.*?)\>/ism', '<$1$2=$3&$4>', $text);
@@ -1947,7 +1808,7 @@ class BBCode
* @param string $addon The addon for which the abstract is meant for
* @return string The abstract
*/
- private static function getAbstract($text, $addon = "")
+ public static function getAbstract($text, $addon = "")
{
$abstract = "";
$abstracts = [];
@@ -1969,4 +1830,148 @@ class BBCode
return $abstract;
}
+
+ /**
+ * @brief Callback function to replace a Friendica style mention in a mention for Diaspora
+ *
+ * @param array $match Matching values for the callback
+ * @return string Replaced mention
+ */
+ private static function bbCodeMention2DiasporaCallback($match)
+ {
+ $contact = Contact::getDetailsByURL($match[3]);
+
+ if (empty($contact['addr'])) {
+ $contact = Probe::uri($match[3]);
+ }
+
+ if (empty($contact['addr'])) {
+ return $match[0];
+ }
+
+ $mention = '@{' . $match[2] . '; ' . $contact['addr'] . '}';
+ return $mention;
+ }
+
+ /**
+ * @brief Converts a BBCode text into Markdown
+ *
+ * This function converts a BBCode item body to be sent to Markdown-enabled
+ * systems like Diaspora and Libertree
+ *
+ * @param string $text
+ * @param bool $for_diaspora Diaspora requires more changes than Libertree
+ * @return string
+ */
+ public static function toMarkdown($text, $for_diaspora = true)
+ {
+ $a = self::getApp();
+
+ $original_text = $text;
+
+ // Since Diaspora is creating a summary for links, this function removes them before posting
+ if ($for_diaspora) {
+ $text = self::removeShareInformation($text);
+ }
+
+ /**
+ * Transform #tags, strip off the [url] and replace spaces with underscore
+ */
+ $url_search_string = "^\[\]";
+ $text = preg_replace_callback("/#\[url\=([$url_search_string]*)\](.*?)\[\/url\]/i",
+ function ($matches) {
+ return '#' . str_replace(' ', '_', $matches[2]);
+ },
+ $text
+ );
+
+ // 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);
+
+ // Add all tags that maybe were removed
+ if (preg_match_all("/#\[url\=([$url_search_string]*)\](.*?)\[\/url\]/ism", $original_text, $tags)) {
+ $tagline = "";
+ foreach ($tags[2] as $tag) {
+ $tag = html_entity_decode($tag, ENT_QUOTES, 'UTF-8');
+ if (!strpos(html_entity_decode($text, ENT_QUOTES, 'UTF-8'), '#' . $tag)) {
+ $tagline .= '#' . $tag . ' ';
+ }
+ }
+ $text = $text . " " . $tagline;
+ }
+ } else {
+ $text = self::convert($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([""], ["
"], $text);
+
+ $stamp1 = microtime(true);
+
+ // Now convert HTML to Markdown
+ $converter = new HtmlConverter();
+ $text = $converter->convert($text);
+
+ // unmask the special chars back to HTML
+ $text = str_replace(['&\_lt\_;', '&\_gt\_;', '&\_amp\_;'], ['<', '>', '&'], $text);
+
+ $a->save_timestamp($stamp1, "parser");
+
+ // Libertree has a problem with escaped hashtags.
+ $text = str_replace(['\#'], ['#'], $text);
+
+ // Remove any leading or trailing whitespace, as this will mess up
+ // the Diaspora signature verification and cause the item to disappear
+ $text = trim($text);
+
+ if ($for_diaspora) {
+ $url_search_string = "^\[\]";
+ $text = preg_replace_callback(
+ "/([@]\[(.*?)\])\(([$url_search_string]*?)\)/ism",
+ ['self', 'bbCodeMention2DiasporaCallback'],
+ $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;
+ }
}