From e0fe3bf77d33cb589053e9c523fcf33ca3b7ee2d Mon Sep 17 00:00:00 2001 From: Michael Date: Mon, 6 Mar 2023 18:00:58 +0000 Subject: [PATCH] Fix plaintext function by partly reverting function removal --- src/Content/Text/BBCode.php | 177 +++++++++++++++++++++++++++++++++ src/Content/Text/Plaintext.php | 2 +- 2 files changed, 178 insertions(+), 1 deletion(-) diff --git a/src/Content/Text/BBCode.php b/src/Content/Text/BBCode.php index 33f91b43d1..12fef305b1 100644 --- a/src/Content/Text/BBCode.php +++ b/src/Content/Text/BBCode.php @@ -209,6 +209,183 @@ class BBCode ); } + public static function getAttachedData(string $body, array $item = []): array + { + /* + - text: + - type: link, video, photo + - title: + - url: + - image: + - description: + - (thumbnail) + */ + + DI::profiler()->startRecording('rendering'); + $has_title = !empty($item['title']); + $plink = $item['plink'] ?? ''; + $post = self::getAttachmentData($body); + + // Get all linked images with alternative image description + if (preg_match_all("/\[img=(http[^\[\]]*)\]([^\[\]]*)\[\/img\]/Usi", $body, $pictures, PREG_SET_ORDER)) { + foreach ($pictures as $picture) { + if ($id = Photo::getIdForName($picture[1])) { + $post['images'][] = ['url' => str_replace('-1.', '-0.', $picture[1]), 'description' => $picture[2], 'id' => $id]; + } else { + $post['remote_images'][] = ['url' => $picture[1], 'description' => $picture[2]]; + } + } + if (!empty($post['images']) && !empty($post['images'][0]['description'])) { + $post['image_description'] = $post['images'][0]['description']; + } + } + + if (preg_match_all("/\[img\]([^\[\]]*)\[\/img\]/Usi", $body, $pictures, PREG_SET_ORDER)) { + foreach ($pictures as $picture) { + if ($id = Photo::getIdForName($picture[1])) { + $post['images'][] = ['url' => str_replace('-1.', '-0.', $picture[1]), 'description' => '', 'id' => $id]; + } else { + $post['remote_images'][] = ['url' => $picture[1], 'description' => '']; + } + } + } + + if (!isset($post['type'])) { + $post['text'] = $body; + } + + // Simplify image codes + $post['text'] = preg_replace("/\[img\=([0-9]*)x([0-9]*)\](.*?)\[\/img\]/ism", '[img]$3[/img]', $post['text']); + $post['text'] = preg_replace("/\[img\=(.*?)\](.*?)\[\/img\]/ism", '[img]$1[/img]', $post['text']); + + // if nothing is found, it maybe having an image. + if (!isset($post['type'])) { + if (preg_match_all("#\[url=([^\]]+?)\]\s*\[img\]([^\[]+?)\[/img\]\s*\[/url\]#ism", $post['text'], $pictures, PREG_SET_ORDER)) { + if ((count($pictures) == 1) && !$has_title) { + if (!empty($item['object-type']) && ($item['object-type'] == Activity\ObjectType::IMAGE)) { + // Replace the preview picture with the real picture + $url = str_replace('-1.', '-0.', $pictures[0][2]); + $data = ['url' => $url, 'type' => 'photo']; + } else { + // Checking, if the link goes to a picture + $data = ParseUrl::getSiteinfoCached($pictures[0][1]); + } + + // 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]); + } + + 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]; + $post['text'] = trim(str_replace($pictures[0][0], '', $post['text'])); + } else { + $imgdata = Images::getInfoFromURLCached($pictures[0][1]); + if (($imgdata) && substr($imgdata['mime'], 0, 6) == 'image/') { + $post['type'] = 'photo'; + $post['image'] = $pictures[0][1]; + $post['preview'] = $pictures[0][2]; + $post['text'] = trim(str_replace($pictures[0][0], '', $post['text'])); + } + } + } elseif (count($pictures) > 0) { + if (count($pictures) > 4) { + $post['type'] = 'link'; + $post['url'] = $plink; + } else { + $post['type'] = 'photo'; + } + + $post['image'] = $pictures[0][2]; + + foreach ($pictures as $picture) { + $post['text'] = trim(str_replace($picture[0], '', $post['text'])); + } + } + } elseif (preg_match_all("(\[img\](.*?)\[\/img\])ism", $post['text'], $pictures, PREG_SET_ORDER)) { + if ($has_title) { + $post['type'] = 'link'; + $post['url'] = $plink; + } else { + $post['type'] = 'photo'; + } + + $post['image'] = $pictures[0][1]; + foreach ($pictures as $picture) { + $post['text'] = trim(str_replace($picture[0], '', $post['text'])); + } + } + + // Test for the external links + preg_match_all("(\[url\](.*?)\[\/url\])ism", $post['text'], $links1, PREG_SET_ORDER); + preg_match_all("(\[url\=(.*?)\].*?\[\/url\])ism", $post['text'], $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['url'] = $links[0][1]; + } + + // Simplify "video" element + $post['text'] = preg_replace('(\[video.*?\ssrc\s?=\s?([^\s\]]+).*?\].*?\[/video\])ism', '[video]$1[/video]', $post['text']); + + // Now count the number of external media links + preg_match_all("(\[vimeo\](.*?)\[\/vimeo\])ism", $post['text'], $links1, PREG_SET_ORDER); + preg_match_all("(\[youtube\\](.*?)\[\/youtube\\])ism", $post['text'], $links2, PREG_SET_ORDER); + preg_match_all("(\[video\\](.*?)\[\/video\\])ism", $post['text'], $links3, PREG_SET_ORDER); + preg_match_all("(\[audio\\](.*?)\[\/audio\\])ism", $post['text'], $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'])) { + $post['type'] = 'text'; + } + + if (($post['type'] == 'photo') && empty($post['images']) && !empty($post['remote_images'])) { + $post['images'] = $post['remote_images']; + $post['image'] = $post['images'][0]['url']; + if (!empty($post['images']) && !empty($post['images'][0]['description'])) { + $post['image_description'] = $post['images'][0]['description']; + } + } + unset($post['remote_images']); + } elseif (isset($post['url']) && ($post['type'] == 'video')) { + $data = ParseUrl::getSiteinfoCached($post['url']); + + if (isset($data['images'][0])) { + $post['image'] = $data['images'][0]['src']; + } + } elseif (preg_match_all("#\[url=([^\]]+?)\]\s*\[img\]([^\[]+?)\[/img\]\s*\[/url\]#ism", $post['text'], $pictures, PREG_SET_ORDER)) { + foreach ($pictures as $picture) { + $post['text'] = trim(str_replace($picture[0], '', $post['text'])); + } + } + + DI::profiler()->stopRecording(); + return $post; + } + /** * Remove [attachment] BBCode * diff --git a/src/Content/Text/Plaintext.php b/src/Content/Text/Plaintext.php index 7e3263d127..c72dad2be5 100644 --- a/src/Content/Text/Plaintext.php +++ b/src/Content/Text/Plaintext.php @@ -132,7 +132,7 @@ class Plaintext $body = BBCode::stripAbstract($body); // At first look at data that is attached via "type-..." stuff - $post = BBCode::getAttachmentData($body, $item); + $post = BBCode::getAttachedData($body, $item); if (($item['title'] != '') && ($post['text'] != '')) { $post['text'] = trim($item['title'] . "\n\n" . $post['text']);