Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[YoutubeBridge] Several fixes and switch maintainer #2115

Merged
merged 5 commits into from
May 16, 2021
Merged

[YoutubeBridge] Several fixes and switch maintainer #2115

Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
589c97c
[YoutubeBridge] Fix incorrectly working method for quering video info
em92 May 16, 2021
e6bca0d
[YoutubeBridge] Partially fix playlist mode.
em92 May 16, 2021
6a95bf0
[YoutubeBridge] Switch maintainer.
em92 May 16, 2021
404353b
better video description in feed and fixed playlist title
em92 May 16, 2021
96bf0fb
100 -> 15
em92 May 16, 2021
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
98 changes: 78 additions & 20 deletions bridges/YoutubeBridge.php
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,7 +12,7 @@ class YoutubeBridge extends BridgeAbstract {
const URI = 'https://www.youtube.com/';
const CACHE_TIMEOUT = 10800; // 3h
const DESCRIPTION = 'Returns the 10 newest videos by username/channel/playlist or search';
const MAINTAINER = 'mitsukarenai';
const MAINTAINER = 'em92';

const PARAMETERS = array(
'By username' => array(
Expand Down Expand Up @@ -68,29 +68,35 @@ private function ytBridgeQueryVideoInfo($vid, &$author, &$desc, &$time){
$html = $this->ytGetSimpleHTMLDOM(self::URI . "watch?v=$vid", true);

// Skip unavailable videos
if(!strpos($html->innertext, 'IS_UNAVAILABLE_PAGE')) {
if(strpos($html->innertext, 'IS_UNAVAILABLE_PAGE') !== false) {
return;
}

foreach($html->find('script') as $script) {
$data = trim($script->innertext);

if(strpos($data, '{') !== 0)
continue; // Wrong script
$elAuthor = $html->find('span[itemprop=author] > link[itemprop=name]', 0);
if (!is_null($elAuthor)) {
$author = $elAuthor->getAttribute('content');
}

$json = json_decode($data);
$elDatePublished = $html->find('meta[itemprop=datePublished]', 0);
if(!is_null($elDatePublished))
$time = strtotime($elDatePublished->getAttribute('content'));

if(!isset($json->itemListElement))
continue; // Wrong script
$scriptRegex = '/var ytInitialData = (.*);<\/script>/';
preg_match($scriptRegex, $html, $matches) or returnServerError('Could not find ytInitialData');
$jsonData = json_decode($matches[1]);
$jsonData = $jsonData->contents->twoColumnWatchNextResults->results->results->contents;

$author = $json->itemListElement[0]->item->name;
$videoSecondaryInfo = null;
foreach($jsonData as $item) {
if (isset($item->videoSecondaryInfoRenderer)) {
$videoSecondaryInfo = $item->videoSecondaryInfoRenderer;
break;
}
}

if(!is_null($html->find('#watch-description-text', 0)))
$desc = $html->find('#watch-description-text', 0)->innertext;

if(!is_null($html->find('meta[itemprop=datePublished]', 0)))
$time = strtotime($html->find('meta[itemprop=datePublished]', 0)->getAttribute('content'));
if (!$videoSecondaryInfo) {
returnServerError('Could not find videoSecondaryInfoRenderer');
}
$desc = nl2br($videoSecondaryInfo->description->runs[0]->text);
}

private function ytBridgeAddItem($vid, $title, $author, $desc, $time){
Expand Down Expand Up @@ -241,16 +247,28 @@ public function collectData(){
returnServerError("Could not request YouTube. Tried:\n - $url_feed\n - $url_listing");
}
} elseif($this->getInput('p')) { /* playlist mode */
// TODO: this mode makes a lot of excess video query requests.
// To make less requests, we need to cache following dictionary "videoId -> datePublished, duration"
// This cache will be used to find out, which videos to fetch
// to make feed of 15 items or more, if there a lot of videos published on that date.
$this->request = $this->getInput('p');
$url_feed = self::URI . 'feeds/videos.xml?playlist_id=' . urlencode($this->request);
$url_listing = self::URI . 'playlist?list=' . urlencode($this->request);
$html = $this->ytGetSimpleHTMLDOM($url_listing)
or returnServerError("Could not request YouTube. Tried:\n - $url_listing");
$item_count = $this->ytBridgeParseHtmlListing($html, 'tr.pl-video', '.pl-video-title a', false);
$scriptRegex = '/var ytInitialData = (.*);<\/script>/';
preg_match($scriptRegex, $html, $matches) or returnServerError('Could not find ytInitialData');
// TODO: this method returns only first 100 video items
// if it has more videos, playlistVideoListRenderer will have continuationItemRenderer as last element
$jsonData = json_decode($matches[1]);
$jsonData = $jsonData->contents->twoColumnBrowseResultsRenderer->tabs[0];
$jsonData = $jsonData->tabRenderer->content->sectionListRenderer->contents[0]->itemSectionRenderer;
$jsonData = $jsonData->contents[0]->playlistVideoListRenderer->contents;
$item_count = count($jsonData);
if ($item_count <= 15 && !$this->skipFeeds() && ($xml = $this->ytGetSimpleHTMLDOM($url_feed))) {
$this->ytBridgeParseXmlFeed($xml);
} else {
$this->ytBridgeParseHtmlListing($html, 'tr.pl-video', '.pl-video-title a');
$this->parseJsonPlaylist($jsonData);
}
$this->feedName = 'Playlist: ' . str_replace(' - YouTube', '', $html->find('title', 0)->plaintext); // feedName will be used by getName()
usort($this->items, function ($item1, $item2) {
Expand Down Expand Up @@ -284,16 +302,56 @@ private function skipFeeds() {
return ($this->getInput('duration_min') || $this->getInput('duration_max'));
}

public function getURI()
{
if (!is_null($this->getInput('p'))) {
return static::URI . 'playlist?list=' . $this->getInput('p');
}

return parent::getURI();
}

public function getName(){
// Name depends on queriedContext:
switch($this->queriedContext) {
case 'By username':
case 'By channel id':
case 'By playlist Id':
case 'Search result':
return $this->feedName . ' - YouTube'; // We already know it's a bridge, right?
return htmlspecialchars_decode($this->feedName) . ' - YouTube'; // We already know it's a bridge, right?
default:
return parent::getName();
}
}

private function parseJsonPlaylist($jsonData) {
$duration_min = $this->getInput('duration_min') ?: -1;
$duration_min = $duration_min * 60;

$duration_max = $this->getInput('duration_max') ?: INF;
$duration_max = $duration_max * 60;

if($duration_max < $duration_min) {
returnClientError('Max duration must be greater than min duration!');
}

foreach($jsonData as $item) {
if (!isset($item->playlistVideoRenderer)) {
continue;
}
$vid = $item->playlistVideoRenderer->videoId;
$title = $item->playlistVideoRenderer->title->runs[0]->text;

$author = '';
$desc = '';
$time = 0;
$duration = intval($item->playlistVideoRenderer->lengthSeconds);
if($duration < $duration_min || $duration > $duration_max) {
continue;
}

$this->ytBridgeQueryVideoInfo($vid, $author, $desc, $time);
$this->ytBridgeAddItem($vid, $title, $author, $desc, $time);
}
}
}