include/plaintext.php

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