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

Jump to bottom

fix episode listing #319

Open
wants to merge 9 commits into
base: master
Choose a base branch
from
Open

fix episode listing #319

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
9 commits
Select commit Hold shift + click to select a range
74f2036
Compatibility with the new version of the episode list
mam4dali Aug 5, 2023
0352a93
improvement
mam4dali Aug 6, 2023
99fffff
Resolving the issue of retrieving _next information
mam4dali Aug 7, 2023
8af516a
fix error on php 8.x
mam4dali Sep 22, 2023
4b689c3
solving the issue of not receiving episodes
mam4dali Sep 24, 2023
0326a30
fix error on get episodes
mam4dali Jan 21, 2024
a599133
Merge remote-tracking branch 'upstream/master' into dev
mam4dali Sep 27, 2024
7fe3575
Fixing the issue of receiving episode information
mam4dali Sep 27, 2024
08b634e
fix lang code in episodes()
mam4dali Sep 27, 2024
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
9 changes: 8 additions & 1 deletion src/Imdb/Request.php
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -154,7 +154,14 @@ public function getStatus()
if (empty($headers[0])) {
return null;
}

//TODO: check if this is the correct way to get the status code
/*
* The last time I saw this header was during the redirect:
* Array
* [0] => HTTP/1.1 200 Connection established
* [2] => HTTP/1.1 308 Permanent Redirect
* [3] => Transfer-Encoding: chunked
*/
if (!preg_match("#^HTTP/[\d\.]+ (\d+)#i", $headers[0], $matches)) {
return null;
}
Expand Down
175 changes: 123 additions & 52 deletions src/Imdb/Title.php
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -109,6 +109,7 @@ class Title extends MdbBase
protected $episodeEpisode = null;
protected $jsonLD = null;
protected $XmlNextJson = null;
protected $buildId = null;

protected $pageUrls = array(
"AlternateVersions" => '/alternateversions',
Expand Down Expand Up @@ -193,9 +194,9 @@ protected function getUrlSuffix($pageName)

if (preg_match('!^Episodes-(-?\d+)$!', $pageName, $match)) {
if (strlen($match[1]) == 4) {
return '/episodes?year=' . $match[1];
return '/episodes/?year=' . $match[1];
} else {
return '/episodes?season=' . $match[1];
return '/episodes/?season=' . $match[1];
}
}

Expand Down Expand Up @@ -2041,7 +2042,11 @@ public function episodes()
return array();
}
}
$page = $this->getPage("Episodes");

$seasons = $this->seasons();
$seasons = $seasons > 0 ? $seasons : 1;

$page = $this->getPage("Episodes-{$seasons}");
if (empty($page)) {
return $this->season_episodes;
}
Expand All @@ -2054,61 +2059,118 @@ public function episodes()
*
* default to year based
*/
$selectId = 'id="byYear"';
if (preg_match('!<select id="bySeason"(.*?)</select!ims', $page, $matchSeason)) {
preg_match_all('#<\s*?option\b[^>]*>(.*?)</option\b[^>]*>#s', $matchSeason[1], $matchOptionSeason);
if (is_numeric(trim($matchOptionSeason[1][0]))) {
//season based
$selectId = 'id="bySeason"';

$xpath = $this->getXpathPage("Episodes-{$seasons}");
$is_new_version = $xpath->query('//a[@data-testid="tab-season-entry"]')->length > 0 || $xpath->query('//a[@data-testid="tab-year-entry"]')->length > 0;
if($is_new_version){
$selectId = 'bySeason';
$liElements = $xpath->query('//a[@data-testid="tab-season-entry"]');
if($liElements->length == 0){
$selectId = 'byYear';
$liElements = $xpath->query('//a[@data-testid="tab-year-entry"]');
}
}

if (preg_match('!<select ' . $selectId . '(.*?)</select!ims', $page, $match)) {
preg_match_all('!<option\s+(selected="selected" |)value="([^"]+)">!i', $match[1], $matches);
$count = count($matches[0]);
for ($i = 0; $i < $count; ++$i) {
$s = $matches[2][$i];
$page = $this->getPage("Episodes-$s");
if (empty($page)) {
continue; // no such page
$buildId = $this->getBuildId();

$lang = !empty($this->config->language) && preg_match('/^[a-z]{2}-[A-Z]{2}$/', $this->config->language) ? $this->config->language : 'en-US';

foreach ($liElements as $li) {
$textContent = $li->textContent;
if(!is_numeric($textContent)){
continue;
}
// fetch episodes images
preg_match_all('!<div class="image">\s*(?<img>.*?)\s*</div>\s*!ims', $page, $img);
$urlIndex = 0;
$preg = '!<div class="info" itemprop="episodes".+?>\s*<meta itemprop="episodeNumber" content="(?<episodeNumber>-?\d+)"/>\s*'
. '<div class="airdate">\s*(?<airdate>.*?)\s*</div>\s*'
. '.+?\shref="/title/tt(?<imdbid>\d{7,8})/[^"]+?"\s+title="(?<title>[^"]+?)"\s+itemprop="name"'
. '.+?<div class="item_description" itemprop="description">(?<plot>.*?)</div>!ims';
preg_match_all($preg, $page, $eps, PREG_SET_ORDER);
foreach ($eps as $ep) {
//Fetch episodes image url
if (preg_match('/(?<!_)src=([\'"])?(.*?)\\1/', $img['img'][$urlIndex], $foundUrl)) {
$image_url = $foundUrl[2];
} else {
$image_url = "";
$id = 'tt'.$this->imdbid();
$action = $selectId == 'byYear' ? 'year' : 'season';

$url = "/_next/data/{$buildId}/{$lang}/title/{$id}/episodes.json?{$action}={$textContent}&tconst={$id}";
$req = new Request("https://" . $this->imdbsite . $url, $this->config);
$success = $req->sendRequest();
if ($success) {
$response = $req->getResponseBody();
$json = json_decode($response, true);
$data = @$json['pageProps']['contentData']['section']['episodes']['items'];
if(!empty($data) && is_array($data)){
foreach ($data as $ep) {
$episode = array(
'imdbid' => str_ireplace('tt', '', $ep['id']),
'type' => @$ep['type'],
'title' => !empty($ep['titleText']) ? trim($ep['titleText']) : '',
'airdate' => @$ep['releaseDate'],
'releaseYear' => @$ep['releaseYear'],
'plot' => !empty($ep['plot']) ? strip_tags($ep['plot']) : '',
'season' => (int)@$ep['season'],
'episode' => (int)@$ep['episode'],
'image_url' => @$ep['image']['url'],
'aggregateRating' => @$ep['aggregateRating'],
'voteCount' => @$ep['voteCount'],
'isReleased' => @$ep['isReleased'],
);

if ($ep['episode'] == -1) {
$this->season_episodes[$textContent][] = $episode;
} else {
$this->season_episodes[$textContent][$ep['episode']] = $episode;
}
}
}
$plot = preg_replace('#<a href="[^"]+"\s+>Add a Plot</a>#', '', trim($ep['plot']));
$plot = preg_replace(
'#Know what this is about\?<br>\s*<a href="[^"]+"\s*> Be the first one to add a plot.\s*</a>#ims',
'',
$plot
);
}
}
} else {
$selectId = 'id="byYear"';
$selectElement = $xpath->query('//select[@id="bySeason"]')->item(0);
if ($selectElement) {
$options = $xpath->query('.//option', $selectElement);
if ($options->length > 0 && is_numeric(trim($options->item(0)->nodeValue))) {
$selectId = 'id="bySeason"';
}
}
if (preg_match('!<select ' . $selectId . '(.*?)</select!ims', $page, $match)) {
preg_match_all('!<option\s+(selected="selected" |)value="([^"]+)">!i', $match[1], $matches);
$count = count($matches[0]);
for ($i = 0; $i < $count; ++$i) {
$s = $matches[2][$i];
$page = $this->getPage("Episodes-$s");
if (empty($page)) {
continue; // no such page
}
// fetch episodes images
preg_match_all('!<div class="image">\s*(?<img>.*?)\s*</div>\s*!ims', $page, $img);
$urlIndex = 0;
$preg = '!<div class="info" itemprop="episodes".+?>\s*<meta itemprop="episodeNumber" content="(?<episodeNumber>-?\d+)"/>\s*'
. '<div class="airdate">\s*(?<airdate>.*?)\s*</div>\s*'
. '.+?\shref="/title/tt(?<imdbid>\d{7,8})/[^"]+?"\s+title="(?<title>[^"]+?)"\s+itemprop="name"'
. '.+?<div class="item_description" itemprop="description">(?<plot>.*?)</div>!ims';
preg_match_all($preg, $page, $eps, PREG_SET_ORDER);
foreach ($eps as $ep) {
//Fetch episodes image url
if (preg_match('/(?<!_)src=([\'"])?(.*?)\\1/', $img['img'][$urlIndex], $foundUrl)) {
$image_url = $foundUrl[2];
} else {
$image_url = "";
}
$plot = preg_replace('#<a href="[^"]+"\s+>Add a Plot</a>#', '', trim($ep['plot']));
$plot = preg_replace(
'#Know what this is about\?<br>\s*<a href="[^"]+"\s*> Be the first one to add a plot.\s*</a>#ims',
'',
$plot
);

$episode = array(
'imdbid' => $ep['imdbid'],
'title' => trim($ep['title']),
'airdate' => $ep['airdate'],
'plot' => strip_tags($plot),
'season' => (int)$s,
'episode' => (int)$ep['episodeNumber'],
'image_url' => $image_url
);
$urlIndex = $urlIndex + 1;
$episode = array(
'imdbid' => $ep['imdbid'],
'title' => trim($ep['title']),
'airdate' => $ep['airdate'],
'plot' => strip_tags($plot),
'season' => (int)$s,
'episode' => (int)$ep['episodeNumber'],
'image_url' => $image_url
);
$urlIndex = $urlIndex + 1;

if ($ep['episodeNumber'] == -1) {
$this->season_episodes[$s][] = $episode;
} else {
$this->season_episodes[$s][$ep['episodeNumber']] = $episode;
if ($ep['episodeNumber'] == -1) {
$this->season_episodes[$s][] = $episode;
} else {
$this->season_episodes[$s][$ep['episodeNumber']] = $episode;
}
}
}
}
Expand Down Expand Up @@ -3268,4 +3330,13 @@ protected function graphQlGetAll($queryName, $fieldName, $nodeQuery)

return $edges;
}
protected function getBuildId(){
if (empty($this->buildId)) {
$query = $this->XmlNextJson()->xpath("//buildId");
if (!empty($query[0])) {
$this->buildId = trim($query[0]);
}
}
return $this->buildId;
}
}