include/plaintext.php

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