From 792f069c6e328b20ff198a68ad699befc04972a9 Mon Sep 17 00:00:00 2001 From: Adam Magness Date: Fri, 26 Jan 2018 19:04:04 -0500 Subject: [PATCH 1/5] Create Plaintext class create Plaintext class with functions from include/plaintext --- src/Content/Text/Plaintext.php | 507 +++++++++++++++++++++++++++++++++ 1 file changed, 507 insertions(+) create mode 100644 src/Content/Text/Plaintext.php diff --git a/src/Content/Text/Plaintext.php b/src/Content/Text/Plaintext.php new file mode 100644 index 0000000000..1463e9f349 --- /dev/null +++ b/src/Content/Text/Plaintext.php @@ -0,0 +1,507 @@ + 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 + */ + function getOldAttachmentData($body) + { + $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]; + } + } + + 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]; + } + + // Search for description + if (preg_match("/\[quote\](.*?)\[\/quote\]/ism", $attacheddata, $matches)) { + $post["description"] = $matches[1]; + } + } + } + return $post; + } + + /** + * @brief Fetches attachment data that were generated with the "attachment" element + * + * @param string $body Message body + * @return array + * 'type' -> Message type ("link", "video", "photo") + * 'text' -> Text before the shared message + * 'after' -> Text after the shared message + * 'image' -> Preview image of the message + * 'url' -> Url to the attached message + * 'title' -> Title of the attachment + * 'description' -> Description of the attachment + */ + function getAttachmentData($body) + { + $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; + } + + $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'); + } + + $preview = ""; + preg_match("/preview='(.*?)'/ism", $attributes, $matches); + if (x($matches, 1)) { + $preview = $matches[1]; + } + + preg_match('/preview="(.*?)"/ism', $attributes, $matches); + if (x($matches, 1)) { + $preview = $matches[1]; + } + + if ($preview != "") { + $data["preview"] = html_entity_decode($preview, ENT_QUOTES, 'UTF-8'); + } + + $data["description"] = trim($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); + } + + 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; + } + + /** + * 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 + * + * @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 string $target_network Name of the network where the post should go to. + * + * @return string The converted message + */ + function toPlaintext($b, $limit = 0, $includedlinks = false, $htmlmode = 2, $target_network = "") + { + // Remove the hash tags + $URLSearchString = "^\[\]"; + $body = preg_replace("/([#@])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '$1$3', $b["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); + + // Remove the abstract + $body = remove_abstract($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, $b); + + if (($b["title"] != "") && ($post["text"] != "")) { + $post["text"] = trim($b["title"]."\n\n".$post["text"]); + } elseif ($b["title"] != "") { + $post["text"] = trim($b["title"]); + } + + $abstract = ""; + + // Fetch the abstract from the given target network + if ($target_network != "") { + $default_abstract = fetch_abstract($b["body"]); + $abstract = fetch_abstract($b["body"], $target_network); + + // If we post to a network with no limit we only fetch + // an abstract exactly for this network + if (($limit == 0) && ($abstract == $default_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); + break; + case 6: + $abstract = fetch_abstract($b["body"], NETWORK_APPNET); + 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"]); + } + } + } + + if ($abstract != "") { + $post["text"] = $abstract; + + if ($post["type"] == "text") { + $post["type"] = "link"; + $post["url"] = $b["plink"]; + } + } + + $html = bbcode($post["text"].$post["after"], false, false, $htmlmode); + $msg = html2plain($html, 0, true); + $msg = trim(html_entity_decode($msg, ENT_QUOTES, 'UTF-8')); + + $link = ""; + if ($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 = $post["image"]; + } + + if (($msg == "") && isset($post["title"])) { + $msg = trim($post["title"]); + } + + 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)) { + $pos = strpos($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, "", $msg)); + } elseif (($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)) { + $limit = $limit - 23 + strlen($link); + } + + $link = ""; + + if ($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(" ", " ", $msg); + } + + // Twitter is using its own limiter, so we always assume that shortened links will have this length + if (iconv_strlen($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"])) { + $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"]; + } + $msg = shortenmsg($msg, $limit); + } + } + + $post["text"] = trim($msg); + + return($post); + } +} From 21881f28855969c211427e0695e3a5f5fe718172 Mon Sep 17 00:00:00 2001 From: Adam Magness Date: Fri, 26 Jan 2018 19:14:47 -0500 Subject: [PATCH 2/5] Update function calls update to new function calls --- include/api.php | 4 ++-- include/bbcode.php | 6 +++--- mod/item.php | 5 +++-- src/Content/Text/Plaintext.php | 20 ++++++++++---------- src/Protocol/OStatus.php | 5 +++-- 5 files changed, 21 insertions(+), 19 deletions(-) diff --git a/include/api.php b/include/api.php index fa4e7f7758..07ff900832 100644 --- a/include/api.php +++ b/include/api.php @@ -8,6 +8,7 @@ use Friendica\App; use Friendica\Content\ContactSelector; use Friendica\Content\Feature; +use Friendica\Content\Text\Plaintext; use Friendica\Core\Addon; use Friendica\Core\System; use Friendica\Core\Config; @@ -47,7 +48,6 @@ require_once 'include/html2bbcode.php'; require_once 'mod/wall_upload.php'; require_once 'mod/proxy.php'; require_once 'include/like.php'; -require_once 'include/plaintext.php'; define('API_METHOD_ANY', '*'); define('API_METHOD_GET', 'GET'); @@ -5196,7 +5196,7 @@ function api_clean_plain_items($Text) */ function api_clean_attachments($body) { - $data = get_attachment_data($body); + $data = Plaintext::getAttachmentData($body); if (!$data) { return $body; diff --git a/include/bbcode.php b/include/bbcode.php index 79412f1a3f..2223eac634 100644 --- a/include/bbcode.php +++ b/include/bbcode.php @@ -5,6 +5,7 @@ use Friendica\App; use Friendica\Content\Smilies; use Friendica\Content\OEmbed; +use Friendica\Content\Text\Plaintext; use Friendica\Core\Addon; use Friendica\Core\Cache; use Friendica\Core\L10n; @@ -15,7 +16,6 @@ use Friendica\Util\Map; require_once 'include/event.php'; require_once 'mod/proxy.php'; -require_once 'include/plaintext.php'; function bb_PictureCacheExt($matches) { if (strpos($matches[3], "data:image/") === 0) { @@ -57,7 +57,7 @@ function bb_map_location($match) { */ function bb_attachment($return, $simplehtml = false, $tryoembed = true) { - $data = get_attachment_data($return); + $data = Plaintext::getAttachmentData($return); if (!$data) { return $return; } @@ -120,7 +120,7 @@ function bb_attachment($return, $simplehtml = false, $tryoembed = true) function bb_remove_share_information($Text, $plaintext = false, $nolink = false) { - $data = get_attachment_data($Text); + $data = Plaintext::getAttachmentData($Text); if (!$data) { return $Text; diff --git a/mod/item.php b/mod/item.php index eaced3e467..2cdb9dd85a 100644 --- a/mod/item.php +++ b/mod/item.php @@ -15,6 +15,7 @@ * information. */ use Friendica\App; +use Friendica\Content\Text\Plaintext; use Friendica\Core\Addon; use Friendica\Core\Config; use Friendica\Core\L10n; @@ -506,7 +507,7 @@ function item_post(App $a) { // embedded bookmark or attachment in post? set bookmark flag $bookmark = 0; - $data = get_attachment_data($body); + $data = Plaintext::getAttachmentData($body); if (preg_match_all("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", $body, $match, PREG_SET_ORDER) || isset($data["type"])) { $objecttype = ACTIVITY_OBJ_BOOKMARK; $bookmark = 1; @@ -524,7 +525,7 @@ function item_post(App $a) { if (!$objecttype) { $objecttype = ACTIVITY_OBJ_NOTE; // Default value require_once 'include/plaintext.php'; - $objectdata = get_attached_data($body); + $objectdata = Plaintext::getAttachedData($body); if ($objectdata["type"] == "link") { $objecttype = ACTIVITY_OBJ_BOOKMARK; diff --git a/src/Content/Text/Plaintext.php b/src/Content/Text/Plaintext.php index 1463e9f349..a72d20ea7a 100644 --- a/src/Content/Text/Plaintext.php +++ b/src/Content/Text/Plaintext.php @@ -28,7 +28,7 @@ class Plaintext * 'title' -> Title of the attachment * 'description' -> Description of the attachment */ - function getOldAttachmentData($body) + private static function getOldAttachmentData($body) { $post = []; @@ -97,12 +97,12 @@ class Plaintext * 'title' -> Title of the attachment * 'description' -> Description of the attachment */ - function getAttachmentData($body) + public static function getAttachmentData($body) { $data = []; if (!preg_match("/(.*)\[attachment(.*?)\](.*?)\[\/attachment\](.*)/ism", $body, $match)) { - return get_old_attachment_data($body); + return self::getOldAttachmentData($body); } $attributes = $match[2]; @@ -202,7 +202,7 @@ class Plaintext return $data; } - function getAttachedData($body, $item = []) + public static function getAttachedData($body, $item = []) { /* - text: @@ -216,7 +216,7 @@ class Plaintext $has_title = !empty($item['title']); $plink = (!empty($item['plink']) ? $item['plink'] : ''); - $post = get_attachment_data($body); + $post = self::getAttachmentData($body); // if nothing is found, it maybe having an image. if (!isset($post["type"])) { @@ -329,7 +329,7 @@ class Plaintext * * @todo For Twitter URLs aren't shortened, but they have to be calculated as if. */ - function shortenMsg($msg, $limit) + public static function shortenMsg($msg, $limit) { $lines = explode("\n", $msg); $msg = ""; @@ -360,7 +360,7 @@ class Plaintext * * @return string The converted message */ - function toPlaintext($b, $limit = 0, $includedlinks = false, $htmlmode = 2, $target_network = "") + public static function toPlaintext($b, $limit = 0, $includedlinks = false, $htmlmode = 2, $target_network = "") { // Remove the hash tags $URLSearchString = "^\[\]"; @@ -374,8 +374,8 @@ class Plaintext // 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"]); + $post = self::getAttachedData($body, $b); if (($b["title"] != "") && ($post["text"] != "")) { $post["text"] = trim($b["title"]."\n\n".$post["text"]); @@ -496,7 +496,7 @@ class Plaintext } elseif (PConfig::get($b["uid"], "system", "no_intelligent_shortening")) { $post["url"] = $b["plink"]; } - $msg = shortenmsg($msg, $limit); + $msg = self::shortenMsg($msg, $limit); } } diff --git a/src/Protocol/OStatus.php b/src/Protocol/OStatus.php index 0e695416f3..05fac90708 100644 --- a/src/Protocol/OStatus.php +++ b/src/Protocol/OStatus.php @@ -5,6 +5,7 @@ namespace Friendica\Protocol; use Friendica\App; +use Friendica\Content\Text\Plaintext; use Friendica\Core\Cache; use Friendica\Core\Config; use Friendica\Core\L10n; @@ -1205,7 +1206,7 @@ class OStatus */ private static function formatPicturePost($body) { - $siteinfo = get_attached_data($body); + $siteinfo = Plaintext::getAttachedData($body); if (($siteinfo["type"] == "photo")) { if (isset($siteinfo["preview"])) { @@ -1327,7 +1328,7 @@ class OStatus private static function getAttachment($doc, $root, $item) { $o = ""; - $siteinfo = get_attached_data($item["body"]); + $siteinfo = Plaintext::getAttachedData($item["body"]); switch ($siteinfo["type"]) { case 'photo': From 11c7441e94a99ce1ca3e1575b88e95ac367a4d1c Mon Sep 17 00:00:00 2001 From: Adam Magness Date: Fri, 26 Jan 2018 19:15:57 -0500 Subject: [PATCH 3/5] Remove requires remove require_once statements --- include/items.php | 1 - mod/item.php | 1 - 2 files changed, 2 deletions(-) diff --git a/include/items.php b/include/items.php index d9e96ba814..edc757dc2e 100644 --- a/include/items.php +++ b/include/items.php @@ -28,7 +28,6 @@ require_once 'include/bbcode.php'; require_once 'include/tags.php'; require_once 'include/text.php'; require_once 'include/threads.php'; -require_once 'include/plaintext.php'; require_once 'mod/share.php'; require_once 'include/enotify.php'; diff --git a/mod/item.php b/mod/item.php index 2cdb9dd85a..10b4d47c4a 100644 --- a/mod/item.php +++ b/mod/item.php @@ -524,7 +524,6 @@ function item_post(App $a) { // Setting the object type if not defined before if (!$objecttype) { $objecttype = ACTIVITY_OBJ_NOTE; // Default value - require_once 'include/plaintext.php'; $objectdata = Plaintext::getAttachedData($body); if ($objectdata["type"] == "link") { From c437b0c87103fa4d40b3f0c5529dfbb0f93d0834 Mon Sep 17 00:00:00 2001 From: Adam Magness Date: Fri, 26 Jan 2018 19:16:26 -0500 Subject: [PATCH 4/5] Remove old file remove include/plaintext.php --- include/plaintext.php | 486 ------------------------------------------ 1 file changed, 486 deletions(-) delete mode 100644 include/plaintext.php diff --git a/include/plaintext.php b/include/plaintext.php deleted file mode 100644 index a5031e7d81..0000000000 --- a/include/plaintext.php +++ /dev/null @@ -1,486 +0,0 @@ - Message type ("link", "video", "photo") - * 'text' -> Text before the shared message - * 'after' -> Text after the shared message - * 'image' -> Preview image of the message - * 'url' -> Url to the attached message - * 'title' -> Title of the attachment - * 'description' -> Description of the attachment - */ -function get_old_attachment_data($body) { - - $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]; - } - - 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]; - } - - // Search for description - if (preg_match("/\[quote\](.*?)\[\/quote\]/ism", $attacheddata, $matches)) - $post["description"] = $matches[1]; - - } - } - return $post; -} - -/** - * @brief Fetches attachment data that were generated with the "attachment" element - * - * @param string $body Message body - * @return array - * 'type' -> Message type ("link", "video", "photo") - * 'text' -> Text before the shared message - * 'after' -> Text after the shared message - * 'image' -> Preview image of the message - * 'url' -> Url to the attached message - * 'title' -> Title of the attachment - * 'description' -> Description of the attachment - */ -function get_attachment_data($body) -{ - $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; - } - - $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'); - } - - $preview = ""; - preg_match("/preview='(.*?)'/ism", $attributes, $matches); - if (x($matches, 1)) { - $preview = $matches[1]; - } - - preg_match('/preview="(.*?)"/ism', $attributes, $matches); - if (x($matches, 1)) { - $preview = $matches[1]; - } - - if ($preview != "") { - $data["preview"] = html_entity_decode($preview, ENT_QUOTES, 'UTF-8'); - } - - $data["description"] = trim($match[3]); - - $data["after"] = trim($match[4]); - - return $data; -} - -function get_attached_data($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); - - 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; -} - -/** - * 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 - * - * @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 string $target_network Name of the network where the post should go to. - * - * @return string The converted message - */ -function plaintext($b, $limit = 0, $includedlinks = false, $htmlmode = 2, $target_network = "") { - - // Remove the hash tags - $URLSearchString = "^\[\]"; - $body = preg_replace("/([#@])\[url\=([$URLSearchString]*)\](.*?)\[\/url\]/ism", '$1$3', $b["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); - - // Remove the abstract - $body = remove_abstract($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, $b); - - if (($b["title"] != "") && ($post["text"] != "")) - $post["text"] = trim($b["title"]."\n\n".$post["text"]); - elseif ($b["title"] != "") - $post["text"] = trim($b["title"]); - - $abstract = ""; - - // Fetch the abstract from the given target network - if ($target_network != "") { - $default_abstract = fetch_abstract($b["body"]); - $abstract = fetch_abstract($b["body"], $target_network); - - // If we post to a network with no limit we only fetch - // an abstract exactly for this network - if (($limit == 0) && ($abstract == $default_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); - break; - case 6: - $abstract = fetch_abstract($b["body"], NETWORK_APPNET); - 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"]); - } - - if ($abstract != "") { - $post["text"] = $abstract; - - if ($post["type"] == "text") { - $post["type"] = "link"; - $post["url"] = $b["plink"]; - } - } - - $html = bbcode($post["text"].$post["after"], false, false, $htmlmode); - $msg = html2plain($html, 0, true); - $msg = trim(html_entity_decode($msg,ENT_QUOTES,'UTF-8')); - - $link = ""; - if ($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 = $post["image"]; - - if (($msg == "") && isset($post["title"])) - $msg = trim($post["title"]); - - 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)) { - $pos = strpos($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, "", $msg)); - elseif (($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)) - $limit = $limit - 23 + strlen($link); - - $link = ""; - - if ($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(" ", " ", $msg); - - // Twitter is using its own limiter, so we always assume that shortened links will have this length - if (iconv_strlen($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"])) { - $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"]; - } - $msg = shortenmsg($msg, $limit); - } - } - - $post["text"] = trim($msg); - - return($post); -} From 177edd2b6e0ef0ddc3cb4c812350dc62ebd0e766 Mon Sep 17 00:00:00 2001 From: Adam Magness Date: Fri, 26 Jan 2018 20:01:32 -0500 Subject: [PATCH 5/5] Rename class to BBCode Rename Plaintext class to BBCode --- include/api.php | 4 ++-- include/bbcode.php | 6 +++--- mod/item.php | 6 +++--- src/Content/Text/{Plaintext.php => BBCode.php} | 4 ++-- src/Protocol/OStatus.php | 6 +++--- 5 files changed, 13 insertions(+), 13 deletions(-) rename src/Content/Text/{Plaintext.php => BBCode.php} (99%) diff --git a/include/api.php b/include/api.php index 07ff900832..8579d2d185 100644 --- a/include/api.php +++ b/include/api.php @@ -8,7 +8,7 @@ use Friendica\App; use Friendica\Content\ContactSelector; use Friendica\Content\Feature; -use Friendica\Content\Text\Plaintext; +use Friendica\Content\Text\BBCode; use Friendica\Core\Addon; use Friendica\Core\System; use Friendica\Core\Config; @@ -5196,7 +5196,7 @@ function api_clean_plain_items($Text) */ function api_clean_attachments($body) { - $data = Plaintext::getAttachmentData($body); + $data = BBCode::getAttachmentData($body); if (!$data) { return $body; diff --git a/include/bbcode.php b/include/bbcode.php index 2223eac634..dfaed8ef78 100644 --- a/include/bbcode.php +++ b/include/bbcode.php @@ -5,7 +5,7 @@ use Friendica\App; use Friendica\Content\Smilies; use Friendica\Content\OEmbed; -use Friendica\Content\Text\Plaintext; +use Friendica\Content\Text\BBCode; use Friendica\Core\Addon; use Friendica\Core\Cache; use Friendica\Core\L10n; @@ -57,7 +57,7 @@ function bb_map_location($match) { */ function bb_attachment($return, $simplehtml = false, $tryoembed = true) { - $data = Plaintext::getAttachmentData($return); + $data = BBCode::getAttachmentData($return); if (!$data) { return $return; } @@ -120,7 +120,7 @@ function bb_attachment($return, $simplehtml = false, $tryoembed = true) function bb_remove_share_information($Text, $plaintext = false, $nolink = false) { - $data = Plaintext::getAttachmentData($Text); + $data = BBCode::getAttachmentData($Text); if (!$data) { return $Text; diff --git a/mod/item.php b/mod/item.php index 10b4d47c4a..d1186a2a0a 100644 --- a/mod/item.php +++ b/mod/item.php @@ -15,7 +15,7 @@ * information. */ use Friendica\App; -use Friendica\Content\Text\Plaintext; +use Friendica\Content\Text\BBCode; use Friendica\Core\Addon; use Friendica\Core\Config; use Friendica\Core\L10n; @@ -507,7 +507,7 @@ function item_post(App $a) { // embedded bookmark or attachment in post? set bookmark flag $bookmark = 0; - $data = Plaintext::getAttachmentData($body); + $data = BBCode::getAttachmentData($body); if (preg_match_all("/\[bookmark\=([^\]]*)\](.*?)\[\/bookmark\]/ism", $body, $match, PREG_SET_ORDER) || isset($data["type"])) { $objecttype = ACTIVITY_OBJ_BOOKMARK; $bookmark = 1; @@ -524,7 +524,7 @@ function item_post(App $a) { // Setting the object type if not defined before if (!$objecttype) { $objecttype = ACTIVITY_OBJ_NOTE; // Default value - $objectdata = Plaintext::getAttachedData($body); + $objectdata = BBCode::getAttachedData($body); if ($objectdata["type"] == "link") { $objecttype = ACTIVITY_OBJ_BOOKMARK; diff --git a/src/Content/Text/Plaintext.php b/src/Content/Text/BBCode.php similarity index 99% rename from src/Content/Text/Plaintext.php rename to src/Content/Text/BBCode.php index a72d20ea7a..b4a3106833 100644 --- a/src/Content/Text/Plaintext.php +++ b/src/Content/Text/BBCode.php @@ -1,6 +1,6 @@