include/plaintext.php

   1 <?php
   2
   3 /**
   4  * @file include/plaintext.php
   5  */
   6
   7 use \Friendica\ParseUrl;
   8
   9 require_once("include/Photo.php");
  10 require_once("include/bbcode.php");
  11 require_once("include/html2plain.php");
  12 require_once("include/network.php");
  13
  14 /**
  15  * @brief Fetches attachment data that were generated the old way
  16  *
  17  * @param string $body Message body
  18  * @return array
  19  * 'type' -> Message type ("link", "video", "photo")
  20  * 'text' -> Text before the shared message
  21  * 'after' -> Text after the shared message
  22  * 'image' -> Preview image of the message
  23  * 'url' -> Url to the attached message
  24  * 'title' -> Title of the attachment
  25  * 'description' -> Description of the attachment
  26  */
  27 function get_old_attachment_data($body) {
  28
  29         $post = array();
  30
  31         // Simplify image codes
  32         $body = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $body);
  33
  34         if (preg_match_all("(\[class=(.*?)\](.*?)\[\/class\])ism",$body, $attached,  PREG_SET_ORDER)) {
  35                 foreach ($attached AS $data) {
  36                         if (!in_array($data[1], array("type-link", "type-video", "type-photo")))
  37                                 continue;
  38
  39                         $post["type"] = substr($data[1], 5);
  40
  41                         $pos = strpos($body, $data[0]);
  42                         if ($pos > 0) {
  43                                 $post["text"] = trim(substr($body, 0, $pos));
  44                                 $post["after"] = trim(substr($body, $pos + strlen($data[0])));
  45                         } else
  46                                 $post["text"] = trim(str_replace($data[0], "", $body));
  47
  48                         $attacheddata = $data[2];
  49
  50                         $URLSearchString = "^\[\]";
  51
  52                         if (preg_match("/\[img\]([$URLSearchString]*)\[\/img\]/ism", $attacheddata, $matches)) {
  53
  54                                 $picturedata = get_photo_info($matches[1]);
  55
  56                                 if (($picturedata[0] >= 500) AND ($picturedata[0] >= $picturedata[1]))
  57                                         $post["image"] = $matches[1];
  58                                 else
  59                                         $post["preview"] = $matches[1];
  60                         }
  61
  62                         if (preg_match("/\[bookmark\=([$URLSearchString]*)\](.*?)\[\/bookmark\]/ism", $attacheddata, $matches)) {
  63                                 $post["url"] = $matches[1];
  64                                 $post["title"] = $matches[2];
  65                         }
  66
  67                         // Search for description
  68                         if (preg_match("/\[quote\](.*?)\[\/quote\]/ism", $attacheddata, $matches))
  69                                 $post["description"] = $matches[1];
  70
  71                 }
  72         }
  73
  74         return $post;
  75 }
  76
  77 /**
  78  * @brief Fetches attachment data that were generated with the "attachment" element
  79  *
  80  * @param string $body Message body
  81  * @return array
  82  * 'type' -> Message type ("link", "video", "photo")
  83  * 'text' -> Text before the shared message
  84  * 'after' -> Text after the shared message
  85  * 'image' -> Preview image of the message
  86  * 'url' -> Url to the attached message
  87  * 'title' -> Title of the attachment
  88  * 'description' -> Description of the attachment
  89  */
  90 function get_attachment_data($body) {
  91
  92         $data = array();
  93
  94         if (!preg_match("/(.*)\[attachment(.*?)\](.*?)\[\/attachment\](.*)/ism", $body, $match))
  95                 return get_old_attachment_data($body);
  96
  97         $attributes = $match[2];
  98
  99         $data["text"] = trim($match[1]);
 100
 101         $type = "";
 102         preg_match("/type='(.*?)'/ism", $attributes, $matches);
 103         if ($matches[1] != "")
 104                 $type = strtolower($matches[1]);
 105
 106         preg_match('/type="(.*?)"/ism', $attributes, $matches);
 107         if ($matches[1] != "")
 108                 $type = strtolower($matches[1]);
 109
 110         if ($type == "")
 111                 return(array());
 112
 113         if (!in_array($type, array("link", "audio", "photo", "video")))
 114                 return(array());
 115
 116         if ($type != "")
 117                 $data["type"] = $type;
 118
 119         $url = "";
 120         preg_match("/url='(.*?)'/ism", $attributes, $matches);
 121         if ($matches[1] != "")
 122                 $url = $matches[1];
 123
 124         preg_match('/url="(.*?)"/ism', $attributes, $matches);
 125         if ($matches[1] != "")
 126                 $url = $matches[1];
 127
 128         if ($url != "")
 129                 $data["url"] = html_entity_decode($url, ENT_QUOTES, 'UTF-8');
 130
 131         $title = "";
 132         preg_match("/title='(.*?)'/ism", $attributes, $matches);
 133         if ($matches[1] != "")
 134                 $title = $matches[1];
 135
 136         preg_match('/title="(.*?)"/ism', $attributes, $matches);
 137         if ($matches[1] != "")
 138                 $title = $matches[1];
 139
 140         if ($title != "") {
 141                 $title = bbcode(html_entity_decode($title, ENT_QUOTES, 'UTF-8'), false, false, true);
 142                 $title = html_entity_decode($title, ENT_QUOTES, 'UTF-8');
 143                 $title = str_replace(array("[", "]"), array("&#91;", "&#93;"), $title);
 144                 $data["title"] = $title;
 145         }
 146
 147         $image = "";
 148         preg_match("/image='(.*?)'/ism", $attributes, $matches);
 149         if ($matches[1] != "")
 150                 $image = $matches[1];
 151
 152         preg_match('/image="(.*?)"/ism', $attributes, $matches);
 153         if ($matches[1] != "")
 154                 $image = $matches[1];
 155
 156         if ($image != "")
 157                 $data["image"] = html_entity_decode($image, ENT_QUOTES, 'UTF-8');
 158
 159         $preview = "";
 160         preg_match("/preview='(.*?)'/ism", $attributes, $matches);
 161         if ($matches[1] != "")
 162                 $preview = $matches[1];
 163
 164         preg_match('/preview="(.*?)"/ism', $attributes, $matches);
 165         if ($matches[1] != "")
 166                 $preview = $matches[1];
 167
 168         if ($preview != "")
 169                 $data["preview"] = html_entity_decode($preview, ENT_QUOTES, 'UTF-8');
 170
 171         $data["description"] = trim($match[3]);
 172
 173         $data["after"] = trim($match[4]);
 174
 175         return($data);
 176 }
 177
 178 function get_attached_data($body) {
 179 /*
 180  - text:
 181  - type: link, video, photo
 182  - title:
 183  - url:
 184  - image:
 185  - description:
 186  - (thumbnail)
 187 */
 188
 189         $post = get_attachment_data($body);
 190
 191         // if nothing is found, it maybe having an image.
 192         if (!isset($post["type"])) {
 193                 $URLSearchString = "^\[\]";
 194                 if (preg_match_all("(\[url=([$URLSearchString]*)\]\s*\[img\]([$URLSearchString]*)\[\/img\]\s*\[\/url\])ism", $body, $pictures,  PREG_SET_ORDER)) {
 195                         if (count($pictures) == 1) {
 196                                 // Checking, if the link goes to a picture
 197                                 $data = ParseUrl::getSiteinfoCached($pictures[0][1], true);
 198
 199                                 // Workaround:
 200                                 // Sometimes photo posts to the own album are not detected at the start.
 201                                 // So we seem to cannot use the cache for these cases. That's strange.
 202                                 if (($data["type"] != "photo") AND strstr($pictures[0][1], "/photos/"))
 203                                         $data = ParseUrl::getSiteinfo($pictures[0][1], true);
 204
 205                                 if ($data["type"] == "photo") {
 206                                         $post["type"] = "photo";
 207                                         if (isset($data["images"][0])) {
 208                                                 $post["image"] = $data["images"][0]["src"];
 209                                                 $post["url"] = $data["url"];
 210                                         } else
 211                                                 $post["image"] = $data["url"];
 212
 213                                         $post["preview"] = $pictures[0][2];
 214                                         $post["text"] = str_replace($pictures[0][0], "", $body);
 215                                 } else {
 216                                         $imgdata = get_photo_info($pictures[0][1]);
 217                                         if (substr($imgdata["mime"], 0, 6) == "image/") {
 218                                                 $post["type"] = "photo";
 219                                                 $post["image"] = $pictures[0][1];
 220                                                 $post["preview"] = $pictures[0][2];
 221                                                 $post["text"] = str_replace($pictures[0][0], "", $body);
 222                                         }
 223                                 }
 224                         } elseif (count($pictures) > 1) {
 225                                 $post["type"] = "link";
 226                                 $post["url"] = $b["plink"];
 227                                 $post["image"] = $pictures[0][2];
 228                                 $post["text"] = $body;
 229                         }
 230                 } elseif (preg_match_all("(\[img\]([$URLSearchString]*)\[\/img\])ism", $body, $pictures,  PREG_SET_ORDER)) {
 231                         if (count($pictures) == 1) {
 232                                 $post["type"] = "photo";
 233                                 $post["image"] = $pictures[0][1];
 234                                 $post["text"] = str_replace($pictures[0][0], "", $body);
 235                         } elseif (count($pictures) > 1) {
 236                                 $post["type"] = "link";
 237                                 $post["url"] = $b["plink"];
 238                                 $post["image"] = $pictures[0][1];
 239                                 $post["text"] = $body;
 240                         }
 241                 }
 242
 243                 if (preg_match_all("(\[url\]([$URLSearchString]*)\[\/url\])ism", $body, $links,  PREG_SET_ORDER)) {
 244                         if (count($links) == 1) {
 245                                 $post["type"] = "text";
 246                                 $post["url"] = $links[0][1];
 247                                 $post["text"] = $body;
 248                         }
 249                 }
 250                 if (!isset($post["type"])) {
 251                         $post["type"] = "text";
 252                         $post["text"] = trim($body);
 253                 }
 254         } elseif (isset($post["url"]) AND ($post["type"] == "video")) {
 255                 $data = ParseUrl::getSiteinfoCached($post["url"], true);
 256
 257                 if (isset($data["images"][0]))
 258                         $post["image"] = $data["images"][0]["src"];
 259         }
 260
 261         return($post);
 262 }
 263
 264 function shortenmsg($msg, $limit, $twitter = false) {
 265         /// @TODO
 266         /// For Twitter URLs aren't shortened, but they have to be calculated as if.
 267
 268         $lines = explode("\n", $msg);
 269         $msg = "";
 270         $recycle = html_entity_decode("&#x2672; ", ENT_QUOTES, 'UTF-8');
 271         foreach ($lines AS $row=>$line) {
 272                 if (iconv_strlen(trim($msg."\n".$line), "UTF-8") <= $limit)
 273                         $msg = trim($msg."\n".$line);
 274                 // Is the new message empty by now or is it a reshared message?
 275                 elseif (($msg == "") OR (($row == 1) AND (substr($msg, 0, 4) == $recycle)))
 276                         $msg = iconv_substr(iconv_substr(trim($msg."\n".$line), 0, $limit, "UTF-8"), 0, -3, "UTF-8")."...";
 277                 else
 278                         break;
 279         }
 280         return($msg);
 281 }
 282
 283 /**
 284  * @brief Convert a message into plaintext for connectors to other networks
 285  *
 286  * @param App $a The application class
 287  * @param array $b The message array that is about to be posted
 288  * @param int $limit The maximum number of characters when posting to that network
 289  * @param bool $includedlinks Has an attached link to be included into the message?
 290  * @param int $htmlmode This triggers the behaviour of the bbcode conversion
 291  * @param string $target_network Name of the network where the post should go to.
 292  *
 293  * @return string The converted message
 294  */
 295 function plaintext($a, $b, $limit = 0, $includedlinks = false, $htmlmode = 2, $target_network = "") {
 296
 297         // Remove the hash tags
 298         $URLSearchString = "^\[\]";
 299         $body = preg_replace("/([#@])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '$1$3', $b["body"]);
 300
 301         // Add an URL element if the text contains a raw link
 302         $body = preg_replace("/([^\]\='".'"'."]|^)(https?\:\/\/[a-zA-Z0-9\:\/\-\?\&\;\.\=\_\~\#\%\$\!\+\,]+)/ism", '$1[url]$2[/url]', $body);
 303
 304         // Remove the abstract
 305         $body = remove_abstract($body);
 306
 307         // At first look at data that is attached via "type-..." stuff
 308         // This will hopefully replaced with a dedicated bbcode later
 309         //$post = get_attached_data($b["body"]);
 310         $post = get_attached_data($body);
 311
 312         if (($b["title"] != "") AND ($post["text"] != ""))
 313                 $post["text"] = trim($b["title"]."\n\n".$post["text"]);
 314         elseif ($b["title"] != "")
 315                 $post["text"] = trim($b["title"]);
 316
 317         $abstract = "";
 318
 319         // Fetch the abstract from the given target network
 320         if ($target_network != "") {
 321                 $default_abstract = fetch_abstract($b["body"]);
 322                 $abstract = fetch_abstract($b["body"], $target_network);
 323
 324                 // If we post to a network with no limit we only fetch
 325                 // an abstract exactly for this network
 326                 if (($limit == 0) AND ($abstract == $default_abstract))
 327                         $abstract = "";
 328
 329         } else // Try to guess the correct target network
 330                 switch ($htmlmode) {
 331                         case 8:
 332                                 $abstract = fetch_abstract($b["body"], NETWORK_TWITTER);
 333                                 break;
 334                         case 7:
 335                                 $abstract = fetch_abstract($b["body"], NETWORK_STATUSNET);
 336                                 break;
 337                         case 6:
 338                                 $abstract = fetch_abstract($b["body"], NETWORK_APPNET);
 339                                 break;
 340                         default: // We don't know the exact target.
 341                                  // We fetch an abstract since there is a posting limit.
 342                                 if ($limit > 0)
 343                                         $abstract = fetch_abstract($b["body"]);
 344                 }
 345
 346         if ($abstract != "") {
 347                 $post["text"] = $abstract;
 348
 349                 if ($post["type"] == "text") {
 350                         $post["type"] = "link";
 351                         $post["url"] = $b["plink"];
 352                 }
 353         }
 354
 355         $html = bbcode($post["text"].$post["after"], false, false, $htmlmode);
 356         $msg = html2plain($html, 0, true);
 357         $msg = trim(html_entity_decode($msg,ENT_QUOTES,'UTF-8'));
 358
 359         $link = "";
 360         if ($includedlinks) {
 361                 if ($post["type"] == "link")
 362                         $link = $post["url"];
 363                 elseif ($post["type"] == "text")
 364                         $link = $post["url"];
 365                 elseif ($post["type"] == "video")
 366                         $link = $post["url"];
 367                 elseif ($post["type"] == "photo")
 368                         $link = $post["image"];
 369
 370                 if (($msg == "") AND isset($post["title"]))
 371                         $msg = trim($post["title"]);
 372
 373                 if (($msg == "") AND isset($post["description"]))
 374                         $msg = trim($post["description"]);
 375
 376                 // If the link is already contained in the post, then it neeedn't to be added again
 377                 // But: if the link is beyond the limit, then it has to be added.
 378                 if (($link != "") AND strstr($msg, $link)) {
 379                         $pos = strpos($msg, $link);
 380
 381                         // Will the text be shortened in the link?
 382                         // Or is the link the last item in the post?
 383                         if (($limit > 0) AND ($pos < $limit) AND (($pos + 23 > $limit) OR ($pos + strlen($link) == strlen($msg))))
 384                                 $msg = trim(str_replace($link, "", $msg));
 385                         elseif (($limit == 0) OR ($pos < $limit)) {
 386                                 // The limit has to be increased since it will be shortened - but not now
 387                                 // Only do it with Twitter (htmlmode = 8)
 388                                 if (($limit > 0) AND (strlen($link) > 23) AND ($htmlmode == 8))
 389                                         $limit = $limit - 23 + strlen($link);
 390
 391                                 $link = "";
 392
 393                                 if ($post["type"] == "text")
 394                                         unset($post["url"]);
 395                         }
 396                 }
 397         }
 398
 399         if ($limit > 0) {
 400                 // Reduce multiple spaces
 401                 // When posted to a network with limited space, we try to gain space where possible
 402                 while (strpos($msg, "  ") !== false)
 403                         $msg = str_replace("  ", " ", $msg);
 404
 405                 // Twitter is using its own limiter, so we always assume that shortened links will have this length
 406                 if (iconv_strlen($link, "UTF-8") > 0)
 407                         $limit = $limit - 23;
 408
 409                 if (iconv_strlen($msg, "UTF-8") > $limit) {
 410
 411                         if (($post["type"] == "text") AND isset($post["url"]))
 412                                 $post["url"] = $b["plink"];
 413                         elseif (!isset($post["url"])) {
 414                                 $limit = $limit - 23;
 415                                 $post["url"] = $b["plink"];
 416                         } elseif (strpos($b["body"], "[share") !== false)
 417                                 $post["url"] = $b["plink"];
 418                         elseif (get_pconfig($b["uid"], "system", "no_intelligent_shortening"))
 419                                 $post["url"] = $b["plink"];
 420
 421                         $msg = shortenmsg($msg, $limit);
 422                 }
 423         }
 424
 425         $post["text"] = trim($msg);
 426
 427         return($post);
 428 }
 429 ?>