3 * @copyright Copyright (C) 2020, Friendica
5 * @license GNU AGPL version 3 or any later version
7 * This program is free software: you can redistribute it and/or modify
8 * it under the terms of the GNU Affero General Public License as
9 * published by the Free Software Foundation, either version 3 of the
10 * License, or (at your option) any later version.
12 * This program is distributed in the hope that it will be useful,
13 * but WITHOUT ANY WARRANTY; without even the implied warranty of
14 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
15 * GNU Affero General Public License for more details.
17 * You should have received a copy of the GNU Affero General Public License
18 * along with this program. If not, see <https://www.gnu.org/licenses/>.
22 namespace Friendica\Model;
24 use Friendica\Content\Text;
25 use Friendica\Core\Protocol;
31 * Convert a message into plaintext for connectors to other networks
33 * @param array $item The message array that is about to be posted
34 * @param int $limit The maximum number of characters when posting to that network
35 * @param bool $includedlinks Has an attached link to be included into the message?
36 * @param int $htmlmode This controls the behavior of the BBCode conversion
37 * @param string $target_network Name of the network where the post should go to.
39 * @return array Same array structure than \Friendica\Content\Text\BBCode::getAttachedData
40 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
41 * @see \Friendica\Content\Text\BBCode::getAttachedData
44 public static function getPlaintextPost($item, $limit = 0, $includedlinks = false, $htmlmode = 2, $target_network = '')
47 $URLSearchString = '^\[\]';
48 $body = preg_replace("/([#@])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '$1$3', $item['body']);
50 // Add an URL element if the text contains a raw link
51 $body = preg_replace('/([^\]\=\'"]|^)(https?\:\/\/[a-zA-Z0-9\:\/\-\?\&\;\.\=\_\~\#\%\$\!\+\,]+)/ism',
52 '$1[url]$2[/url]', $body);
54 // Remove the abstract
55 $body = Text\BBCode::stripAbstract($body);
57 // At first look at data that is attached via "type-..." stuff
58 // This will hopefully replaced with a dedicated bbcode later
59 //$post = self::getAttachedData($b['body']);
60 $post = Text\BBCode::getAttachedData($body, $item);
62 if (($item['title'] != '') && ($post['text'] != '')) {
63 $post['text'] = trim($item['title'] . "\n\n" . $post['text']);
64 } elseif ($item['title'] != '') {
65 $post['text'] = trim($item['title']);
70 // Fetch the abstract from the given target network
71 if ($target_network != '') {
72 $default_abstract = Text\BBCode::getAbstract($item['body']);
73 $abstract = Text\BBCode::getAbstract($item['body'], $target_network);
75 // If we post to a network with no limit we only fetch
76 // an abstract exactly for this network
77 if (($limit == 0) && ($abstract == $default_abstract)) {
80 } else {// Try to guess the correct target network
83 $abstract = Text\BBCode::getAbstract($item['body'], Protocol::TWITTER);
87 $abstract = Text\BBCode::getAbstract($item['body'], Protocol::STATUSNET);
90 default: // We don't know the exact target.
91 // We fetch an abstract since there is a posting limit.
93 $abstract = Text\BBCode::getAbstract($item['body']);
98 if ($abstract != '') {
99 $post['text'] = $abstract;
101 if ($post['type'] == 'text') {
102 $post['type'] = 'link';
103 $post['url'] = $item['plink'];
107 $html = Text\BBCode::convert($post['text'] . ($post['after'] ?? ''), false, $htmlmode);
108 $msg = Text\HTML::toPlaintext($html, 0, true);
109 $msg = trim(html_entity_decode($msg, ENT_QUOTES, 'UTF-8'));
112 if ($includedlinks) {
113 if ($post['type'] == 'link') {
114 $link = $post['url'];
115 } elseif ($post['type'] == 'text') {
116 $link = $post['url'] ?? '';
117 } elseif ($post['type'] == 'video') {
118 $link = $post['url'];
119 } elseif ($post['type'] == 'photo') {
120 $link = $post['image'];
123 if (($msg == '') && isset($post['title'])) {
124 $msg = trim($post['title']);
127 if (($msg == '') && isset($post['description'])) {
128 $msg = trim($post['description']);
131 // If the link is already contained in the post, then it neeedn't to be added again
132 // But: if the link is beyond the limit, then it has to be added.
133 if (($link != '') && strstr($msg, $link)) {
134 $pos = strpos($msg, $link);
136 // Will the text be shortened in the link?
137 // Or is the link the last item in the post?
138 if (($limit > 0) && ($pos < $limit) && (($pos + 23 > $limit) || ($pos + strlen($link) == strlen($msg)))) {
139 $msg = trim(str_replace($link, '', $msg));
140 } elseif (($limit == 0) || ($pos < $limit)) {
141 // The limit has to be increased since it will be shortened - but not now
142 // Only do it with Twitter (htmlmode = 8)
143 if (($limit > 0) && (strlen($link) > 23) && ($htmlmode == 8)) {
144 $limit = $limit - 23 + strlen($link);
149 if ($post['type'] == 'text') {
157 // Reduce multiple spaces
158 // When posted to a network with limited space, we try to gain space where possible
159 while (strpos($msg, ' ') !== false) {
160 $msg = str_replace(' ', ' ', $msg);
163 // Twitter is using its own limiter, so we always assume that shortened links will have this length
164 if (iconv_strlen($link, 'UTF-8') > 0) {
165 $limit = $limit - 23;
168 if (iconv_strlen($msg, 'UTF-8') > $limit) {
169 if (($post['type'] == 'text') && isset($post['url'])) {
170 $post['url'] = $item['plink'];
171 } elseif (!isset($post['url'])) {
172 $limit = $limit - 23;
173 $post['url'] = $item['plink'];
174 } elseif (strpos($item['body'], '[share') !== false) {
175 $post['url'] = $item['plink'];
176 } elseif (DI::pConfig()->get($item['uid'], 'system', 'no_intelligent_shortening')) {
177 $post['url'] = $item['plink'];
179 $msg = Text\Plaintext::shorten($msg, $limit);
183 $post['text'] = trim($msg);