<?php
/**
- * @file src/Content/Text/Plaintext.php
+ * @copyright Copyright (C) 2010-2021, the Friendica project
+ *
+ * @license GNU AGPL version 3 or any later version
+ *
+ * This program is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU Affero General Public License as
+ * published by the Free Software Foundation, either version 3 of the
+ * License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU Affero General Public License for more details.
+ *
+ * You should have received a copy of the GNU Affero General Public License
+ * along with this program. If not, see <https://www.gnu.org/licenses/>.
+ *
*/
-namespace Friendica\Content\Text;
-use Friendica\App;
-use Friendica\Core\PConfig;
-use Friendica\Object\Image;
-use Friendica\Util\ParseUrl;
+namespace Friendica\Content\Text;
-require_once "include/bbcode.php";
-require_once "include/html2plain.php";
-require_once "include/network.php";
+use Friendica\Core\Protocol;
+use Friendica\DI;
class Plaintext
{
/**
- * @brief Fetches attachment data that were generated the old way
+ * Shortens message
+ *
+ * @param string $msg
+ * @param int $limit
+ * @param int $uid
+ * @return string
*
- * @param string $body Message body
- * @return array
- * 'type' -> Message type ("link", "video", "photo")
- * 'text' -> Text before the shared message
- * 'after' -> Text after the shared message
- * 'image' -> Preview image of the message
- * 'url' -> Url to the attached message
- * 'title' -> Title of the attachment
- * 'description' -> Description of the attachment
+ * @todo For Twitter URLs aren't shortened, but they have to be calculated as if.
*/
- function getOldAttachmentData($body)
+ public static function shorten(string $msg, int $limit, int $uid = 0):string
{
- $post = [];
-
- // Simplify image codes
- $body = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $body);
-
- if (preg_match_all("(\[class=(.*?)\](.*?)\[\/class\])ism", $body, $attached, PREG_SET_ORDER)) {
- foreach ($attached as $data) {
- if (!in_array($data[1], ["type-link", "type-video", "type-photo"])) {
- continue;
- }
-
- $post["type"] = substr($data[1], 5);
-
- $pos = strpos($body, $data[0]);
- if ($pos > 0) {
- $post["text"] = trim(substr($body, 0, $pos));
- $post["after"] = trim(substr($body, $pos + strlen($data[0])));
- } else {
- $post["text"] = trim(str_replace($data[0], "", $body));
- }
-
- $attacheddata = $data[2];
-
- $URLSearchString = "^\[\]";
-
- if (preg_match("/\[img\]([$URLSearchString]*)\[\/img\]/ism", $attacheddata, $matches)) {
-
- $picturedata = Image::getInfoFromURL($matches[1]);
-
- if (($picturedata[0] >= 500) && ($picturedata[0] >= $picturedata[1])) {
- $post["image"] = $matches[1];
- } else {
- $post["preview"] = $matches[1];
- }
- }
+ $ellipsis = html_entity_decode("…", ENT_QUOTES, 'UTF-8');
- if (preg_match("/\[bookmark\=([$URLSearchString]*)\](.*?)\[\/bookmark\]/ism", $attacheddata, $matches)) {
- $post["url"] = $matches[1];
- $post["title"] = $matches[2];
- }
- if (($post["url"] == "") && (in_array($post["type"], ["link", "video"]))
- && preg_match("/\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", $attacheddata, $matches)) {
- $post["url"] = $matches[1];
- }
+ if (!empty($uid) && DI::pConfig()->get($uid, 'system', 'simple_shortening')) {
+ return iconv_substr(iconv_substr(trim($msg), 0, $limit, "UTF-8"), 0, -3, "UTF-8") . $ellipsis;
+ }
- // Search for description
- if (preg_match("/\[quote\](.*?)\[\/quote\]/ism", $attacheddata, $matches)) {
- $post["description"] = $matches[1];
- }
+ $lines = explode("\n", $msg);
+ $msg = "";
+ $recycle = html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8');
+ foreach ($lines as $row => $line) {
+ if (iconv_strlen(trim($msg . "\n" . $line), "UTF-8") <= $limit) {
+ $msg = trim($msg . "\n" . $line);
+ } elseif (($msg == "") || (($row == 1) && (substr($msg, 0, 4) == $recycle))) {
+ // Is the new message empty by now or is it a reshared message?
+ $msg = iconv_substr(iconv_substr(trim($msg . "\n" . $line), 0, $limit, "UTF-8"), 0, -3, "UTF-8") . $ellipsis;
+ } else {
+ break;
}
}
- return $post;
+
+ return $msg;
}
/**
- * @brief Fetches attachment data that were generated with the "attachment" element
+ * Returns the character positions of the provided boundaries, optionally skipping a number of first occurrences
*
- * @param string $body Message body
- * @return array
- * 'type' -> Message type ("link", "video", "photo")
- * 'text' -> Text before the shared message
- * 'after' -> Text after the shared message
- * 'image' -> Preview image of the message
- * 'url' -> Url to the attached message
- * 'title' -> Title of the attachment
- * 'description' -> Description of the attachment
+ * @param string $text Text to search
+ * @param string $open Left boundary
+ * @param string $close Right boundary
+ * @param int $occurrences Number of first occurrences to skip
+ * @return boolean|array
*/
- function getAttachmentData($body)
+ public static function getBoundariesPosition($text, $open, $close, $occurrences = 0)
{
- $data = [];
-
- if (!preg_match("/(.*)\[attachment(.*?)\](.*?)\[\/attachment\](.*)/ism", $body, $match)) {
- return get_old_attachment_data($body);
- }
-
- $attributes = $match[2];
-
- $data["text"] = trim($match[1]);
-
- $type = "";
- preg_match("/type='(.*?)'/ism", $attributes, $matches);
- if (x($matches, 1)) {
- $type = strtolower($matches[1]);
- }
-
- preg_match('/type="(.*?)"/ism', $attributes, $matches);
- if (x($matches, 1)) {
- $type = strtolower($matches[1]);
- }
-
- if ($type == "") {
- return [];
- }
-
- if (!in_array($type, ["link", "audio", "photo", "video"])) {
- return [];
- }
-
- if ($type != "") {
- $data["type"] = $type;
- }
-
- $url = "";
- preg_match("/url='(.*?)'/ism", $attributes, $matches);
- if (x($matches, 1)) {
- $url = $matches[1];
- }
-
- preg_match('/url="(.*?)"/ism', $attributes, $matches);
- if (x($matches, 1)) {
- $url = $matches[1];
- }
-
- if ($url != "") {
- $data["url"] = html_entity_decode($url, ENT_QUOTES, 'UTF-8');
- }
-
- $title = "";
- preg_match("/title='(.*?)'/ism", $attributes, $matches);
- if (x($matches, 1)) {
- $title = $matches[1];
- }
-
- preg_match('/title="(.*?)"/ism', $attributes, $matches);
- if (x($matches, 1)) {
- $title = $matches[1];
- }
-
- if ($title != "") {
- $title = bbcode(html_entity_decode($title, ENT_QUOTES, 'UTF-8'), false, false, true);
- $title = html_entity_decode($title, ENT_QUOTES, 'UTF-8');
- $title = str_replace(["[", "]"], ["[", "]"], $title);
- $data["title"] = $title;
+ if ($occurrences < 0) {
+ $occurrences = 0;
}
- $image = "";
- preg_match("/image='(.*?)'/ism", $attributes, $matches);
- if (x($matches, 1)) {
- $image = $matches[1];
- }
-
- preg_match('/image="(.*?)"/ism', $attributes, $matches);
- if (x($matches, 1)) {
- $image = $matches[1];
- }
-
- if ($image != "") {
- $data["image"] = html_entity_decode($image, ENT_QUOTES, 'UTF-8');
+ $start_pos = -1;
+ for ($i = 0; $i <= $occurrences; $i++) {
+ if ($start_pos !== false) {
+ $start_pos = strpos($text, $open, $start_pos + 1);
+ }
}
- $preview = "";
- preg_match("/preview='(.*?)'/ism", $attributes, $matches);
- if (x($matches, 1)) {
- $preview = $matches[1];
+ if ($start_pos === false) {
+ return false;
}
- preg_match('/preview="(.*?)"/ism', $attributes, $matches);
- if (x($matches, 1)) {
- $preview = $matches[1];
- }
+ $end_pos = strpos($text, $close, $start_pos);
- if ($preview != "") {
- $data["preview"] = html_entity_decode($preview, ENT_QUOTES, 'UTF-8');
+ if ($end_pos === false) {
+ return false;
}
- $data["description"] = trim($match[3]);
-
- $data["after"] = trim($match[4]);
-
- return $data;
- }
-
- function getAttachedData($body, $item = [])
- {
- /*
- - text:
- - type: link, video, photo
- - title:
- - url:
- - image:
- - description:
- - (thumbnail)
- */
-
- $has_title = !empty($item['title']);
- $plink = (!empty($item['plink']) ? $item['plink'] : '');
- $post = get_attachment_data($body);
-
- // if nothing is found, it maybe having an image.
- if (!isset($post["type"])) {
- // Simplify image codes
- $body = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $body);
-
- $URLSearchString = "^\[\]";
- 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
- $data = ParseUrl::getSiteinfoCached($pictures[0][1], true);
-
- // Workaround:
- // Sometimes photo posts to the own album are not detected at the start.
- // So we seem to cannot use the cache for these cases. That's strange.
- if (($data["type"] != "photo") && strstr($pictures[0][1], "/photos/")) {
- $data = ParseUrl::getSiteinfo($pictures[0][1], true);
- }
+ $res = ['start' => $start_pos, 'end' => $end_pos];
- if ($data["type"] == "photo") {
- $post["type"] = "photo";
- if (isset($data["images"][0])) {
- $post["image"] = $data["images"][0]["src"];
- $post["url"] = $data["url"];
- } else {
- $post["image"] = $data["url"];
- }
-
- $post["preview"] = $pictures[0][2];
- $post["text"] = str_replace($pictures[0][0], "", $body);
- } else {
- $imgdata = Image::getInfoFromURL($pictures[0][1]);
- if (substr($imgdata["mime"], 0, 6) == "image/") {
- $post["type"] = "photo";
- $post["image"] = $pictures[0][1];
- $post["preview"] = $pictures[0][2];
- $post["text"] = str_replace($pictures[0][0], "", $body);
- }
- }
- } elseif (count($pictures) > 0) {
- $post["type"] = "link";
- $post["url"] = $plink;
- $post["image"] = $pictures[0][2];
- $post["text"] = $body;
- }
- } elseif (preg_match_all("(\[img\]([$URLSearchString]*)\[\/img\])ism", $body, $pictures, PREG_SET_ORDER)) {
- if ((count($pictures) == 1) && !$has_title) {
- $post["type"] = "photo";
- $post["image"] = $pictures[0][1];
- $post["text"] = str_replace($pictures[0][0], "", $body);
- } elseif (count($pictures) > 0) {
- $post["type"] = "link";
- $post["url"] = $plink;
- $post["image"] = $pictures[0][1];
- $post["text"] = $body;
- }
- }
-
- // Test for the external links
- preg_match_all("(\[url\]([$URLSearchString]*)\[\/url\])ism", $body, $links1, PREG_SET_ORDER);
- preg_match_all("(\[url\=([$URLSearchString]*)\].*?\[\/url\])ism", $body, $links2, PREG_SET_ORDER);
-
- $links = array_merge($links1, $links2);
-
- // If there is only a single one, then use it.
- // This should cover link posts via API.
- if ((count($links) == 1) && !isset($post["preview"]) && !$has_title) {
- $post["type"] = "link";
- $post["text"] = trim($body);
- $post["url"] = $links[0][1];
- }
-
- // Now count the number of external media links
- preg_match_all("(\[vimeo\](.*?)\[\/vimeo\])ism", $body, $links1, PREG_SET_ORDER);
- preg_match_all("(\[youtube\\](.*?)\[\/youtube\\])ism", $body, $links2, PREG_SET_ORDER);
- preg_match_all("(\[video\\](.*?)\[\/video\\])ism", $body, $links3, PREG_SET_ORDER);
- preg_match_all("(\[audio\\](.*?)\[\/audio\\])ism", $body, $links4, PREG_SET_ORDER);
-
- // Add them to the other external links
- $links = array_merge($links, $links1, $links2, $links3, $links4);
-
- // Are there more than one?
- if (count($links) > 1) {
- // The post will be the type "text", which means a blog post
- unset($post["type"]);
- $post["url"] = $plink;
- }
-
- if (!isset($post["type"])) {
- $post["type"] = "text";
- $post["text"] = trim($body);
- }
- } elseif (isset($post["url"]) && ($post["type"] == "video")) {
- $data = ParseUrl::getSiteinfoCached($post["url"], true);
-
- if (isset($data["images"][0])) {
- $post["image"] = $data["images"][0]["src"];
- }
- }
-
- return $post;
+ return $res;
}
/**
- * Shortens message
- *
- * @param type $msg
- * @param type $limit
- * @return type
- *
- * @todo For Twitter URLs aren't shortened, but they have to be calculated as if.
- */
- function shortenMsg($msg, $limit)
- {
- $lines = explode("\n", $msg);
- $msg = "";
- $recycle = html_entity_decode("♲ ", ENT_QUOTES, 'UTF-8');
- $ellipsis = html_entity_decode("…", ENT_QUOTES, 'UTF-8');
- foreach ($lines as $row => $line) {
- if (iconv_strlen(trim($msg . "\n" . $line), "UTF-8") <= $limit) {
- $msg = trim($msg . "\n" . $line);
- } elseif (($msg == "") || (($row == 1) && (substr($msg, 0, 4) == $recycle))) {
- // Is the new message empty by now or is it a reshared message?
- $msg = iconv_substr(iconv_substr(trim($msg . "\n" . $line), 0, $limit, "UTF-8"), 0, -3, "UTF-8") . $ellipsis;
- } else {
- break;
- }
- }
-
- return $msg;
- }
-
- /**
- * @brief Convert a message into plaintext for connectors to other networks
+ * Convert a message into plaintext for connectors to other networks
*
- * @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 array $item 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 controls the behavior of the BBCode conversion
* @param string $target_network Name of the network where the post should go to.
*
- * @return string The converted message
+ * @return array Same array structure than \Friendica\Content\Text\BBCode::getAttachedData
+ * @throws \Friendica\Network\HTTPException\InternalServerErrorException
+ * @see \Friendica\Content\Text\BBCode::getAttachedData
+ *
*/
- function toPlaintext($b, $limit = 0, $includedlinks = false, $htmlmode = 2, $target_network = "")
+ public static function getPost($item, $limit = 0, $includedlinks = false, $htmlmode = BBCode::API, $target_network = '')
{
- // Remove the hash tags
- $URLSearchString = "^\[\]";
- $body = preg_replace("/([#@])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '$1$3', $b["body"]);
+ // Remove hashtags
+ $URLSearchString = '^\[\]';
+ $body = preg_replace("/([#@])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '$1$3', $item['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);
+ $body = preg_replace('/([^\]\=\'"]|^)(https?\:\/\/[a-zA-Z0-9\:\/\-\?\&\;\.\=\_\~\#\%\$\!\+\,]+)/ism',
+ '$1[url]$2[/url]', $body);
// Remove the abstract
- $body = remove_abstract($body);
+ $body = BBCode::stripAbstract($body);
// At first look at data that is attached via "type-..." stuff
// This will hopefully replaced with a dedicated bbcode later
- //$post = get_attached_data($b["body"]);
- $post = get_attached_data($body, $b);
+ //$post = self::getAttachedData($b['body']);
+ $post = BBCode::getAttachedData($body, $item);
- if (($b["title"] != "") && ($post["text"] != "")) {
- $post["text"] = trim($b["title"]."\n\n".$post["text"]);
- } elseif ($b["title"] != "") {
- $post["text"] = trim($b["title"]);
+ if (($item['title'] != '') && ($post['text'] != '')) {
+ $post['text'] = trim($item['title'] . "\n\n" . $post['text']);
+ } elseif ($item['title'] != '') {
+ $post['text'] = trim($item['title']);
}
- $abstract = "";
+ $abstract = '';
// Fetch the abstract from the given target network
- if ($target_network != "") {
- $default_abstract = fetch_abstract($b["body"]);
- $abstract = fetch_abstract($b["body"], $target_network);
+ if ($target_network != '') {
+ $default_abstract = BBCode::getAbstract($item['body']);
+ $abstract = BBCode::getAbstract($item['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 = "";
+ $abstract = '';
}
} else {// Try to guess the correct target network
switch ($htmlmode) {
- case 8:
- $abstract = fetch_abstract($b["body"], NETWORK_TWITTER);
- break;
- case 7:
- $abstract = fetch_abstract($b["body"], NETWORK_STATUSNET);
+ case BBCode::TWITTER:
+ $abstract = BBCode::getAbstract($item['body'], Protocol::TWITTER);
break;
- case 6:
- $abstract = fetch_abstract($b["body"], NETWORK_APPNET);
+
+ case BBCode::OSTATUS:
+ $abstract = BBCode::getAbstract($item['body'], Protocol::STATUSNET);
break;
+
default: // We don't know the exact target.
// We fetch an abstract since there is a posting limit.
if ($limit > 0) {
- $abstract = fetch_abstract($b["body"]);
+ $abstract = BBCode::getAbstract($item['body']);
}
}
}
- if ($abstract != "") {
- $post["text"] = $abstract;
+ if ($abstract != '') {
+ $post['text'] = $abstract;
- if ($post["type"] == "text") {
- $post["type"] = "link";
- $post["url"] = $b["plink"];
+ if ($post['type'] == 'text') {
+ $post['type'] = 'link';
+ $post['url'] = $item['plink'];
}
}
- $html = bbcode($post["text"].$post["after"], false, false, $htmlmode);
- $msg = html2plain($html, 0, true);
+ $html = BBCode::convertForUriId($item['uri-id'], $post['text'] . ($post['after'] ?? ''), $htmlmode);
+ $msg = HTML::toPlaintext($html, 0, true);
$msg = trim(html_entity_decode($msg, ENT_QUOTES, 'UTF-8'));
- $link = "";
+ $link = '';
if ($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 = $post["image"];
+ if ($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['title'])) {
+ $msg = trim($post['title']);
}
- if (($msg == "") && isset($post["description"])) {
- $msg = trim($post["description"]);
+ 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, $link)) {
+ if (($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));
+ $msg = trim(str_replace($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)) {
+ // Only do it with Twitter
+ if (($limit > 0) && (strlen($link) > 23) && ($htmlmode == BBCode::TWITTER)) {
$limit = $limit - 23 + strlen($link);
}
- $link = "";
+ $link = '';
- if ($post["type"] == "text") {
- unset($post["url"]);
+ if ($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(" ", " ", $msg);
+ while (strpos($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) {
+ 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"])) {
+ if (iconv_strlen($msg, 'UTF-8') > $limit) {
+ if (($post['type'] == 'text') && isset($post['url'])) {
+ $post['url'] = $item['plink'];
+ } elseif (!isset($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"];
+ $post['url'] = $item['plink'];
+ } elseif (strpos($item['body'], '[share') !== false) {
+ $post['url'] = $item['plink'];
+ } elseif (DI::pConfig()->get($item['uid'], 'system', 'no_intelligent_shortening')) {
+ $post['url'] = $item['plink'];
}
- $msg = shortenmsg($msg, $limit);
+ $msg = self::shorten($msg, $limit, $item['uid']);
}
}
- $post["text"] = trim($msg);
+ $post['text'] = trim($msg);
- return($post);
+ return $post;
}
}