public static function parseDay($rows)
{
$result = array();
foreach ($rows as $item) {
$crawler = new Crawler($item);
$anime = new Anime();
$url = $crawler->filter('a[class="link-title"]')->attr('href');
$id = preg_match('/\\/(anime|manga)\\/(\\d+)\\/.*?/', $url, $urlParts);
if ($id !== false || $id !== 0) {
$anime->setId((int) $urlParts[2]);
}
$anime->setTitle(trim($crawler->filter('a[class="link-title"]')->text()));
$producer = $crawler->filter('span[class="producer"] a');
if ($producer->count() > 0) {
$anime->setProducers(explode(', ', $crawler->filter('span[class="producer"] a')->text()));
}
$anime->setEpisodes((int) str_replace(' eps', '', $crawler->filter('div[class="eps"] span')->text()));
$genres = $crawler->filter('div[class="genres-inner js-genre-inner"] a');
$genreArray = array();
foreach ($genres as $genre) {
$genreCrawler = new Crawler($genre);
$genreArray[] = $genreCrawler->text();
}
$anime->setGenres($genreArray);
$anime->setImageUrl($crawler->filter('div[class="image lazyload"]')->attr('data-bg'));
$anime->setSynopsis(trim($crawler->filter('div[class="synopsis js-synopsis"]')->text()));
$detail = explode('-', $crawler->filter('div[class="info"]')->text());
$anime->setType(trim($detail[0]));
$anime->setMembersCount((int) str_replace(',', '', trim($crawler->filter('span[class="member fl-r"]')->text())));
$anime->setMembersScore((double) trim($crawler->filter('span[class="score"]')->text()));
$result[] = $anime;
}
return $result;
}
private static function parseAnime($item)
{
$crawler = new Crawler($item);
$anime = new Anime();
$anime->setId($crawler->filter('id')->text());
$anime->setTitle($crawler->filter('title')->text());
$otherTitles = array();
$english = explode('; ', $crawler->filter('english')->text());
if (count($english) > 0 && $english !== '') {
$otherTitles['english'] = $english;
}
$synonyms = explode('; ', $crawler->filter('synonyms')->text());
if (count($synonyms) > 0 && $synonyms[0] !== '') {
$otherTitles['synonyms'] = $synonyms;
}
$anime->setOtherTitles($otherTitles);
$anime->setEpisodes($crawler->filter('episodes')->text());
$anime->setMembersScore($crawler->filter('score')->text());
$anime->setStatus($crawler->filter('status')->text());
$startDate = $crawler->filter('start_date')->text();
if ($startDate !== '0000-00-00') {
$anime->setStartDate((new \DateTime())->createFromFormat('Y-m-d', $startDate));
}
$EndDate = $crawler->filter('end_date')->text();
if ($EndDate !== '0000-00-00') {
$anime->setEndDate((new \DateTime())->createFromFormat('Y-m-d', $EndDate));
}
$anime->setSynopsis($crawler->filter('synopsis')->text());
$anime->setImageUrl($crawler->filter('image')->text());
return $anime;
}
public static function parse($contents, $apiVersion)
{
$crawler = new Crawler();
$crawler->addHTMLContent($contents, 'UTF-8');
$animerecord = new Anime();
# Anime ID.
# Example:
# <input type="hidden" name="aid" value="790">
$animerecord->setId((int) $crawler->filter('input[name="aid"]')->attr('value'));
# Title and rank.
# Example:
# <span itemprop="name">One Piece</span>
$animerecord->setTitle(trim($crawler->filter('span[itemprop="name"]')->text()));
$rank = $crawler->filterXPath('//span[contains(@class, "ranked")]');
if (count($rank) > 0) {
$animerecord->setRank((int) str_replace('Ranked #', '', $rank->text()));
}
# Title Image
# Example:
# <a href="http://myanimelist.net/anime/16353/Love_Lab/pic&pid=50257"><img src="http://cdn.myanimelist.net/images/anime/12/50257.jpg" alt="Love Lab" align="center"></a>
$animerecord->setImageUrl(str_replace('t.jpg', '.jpg', $crawler->filter('div#content tr td div img')->attr('src')));
# Alternative Titles section.
# Example:
# <h2>Alternative Titles</h2>
# <div class="spaceit_pad"><span class="dark_text">English:</span> Lucky Star/div>
# <div class="spaceit_pad"><span class="dark_text">Synonyms:</span> Lucky Star, Raki ☆ Suta</div>
# <div class="spaceit_pad"><span class="dark_text">Japanese:</span> らき すた</div>
$leftcolumn = $crawler->filterXPath('//div[@id="content"]/table/tr/td[@class="borderClass"]');
# English:
$extracted = $leftcolumn->filterXPath('//span[text()="English:"]');
if ($extracted->count() > 0) {
$text = trim(str_replace($extracted->text(), '', $extracted->parents()->text()));
$other_titles['english'] = explode(', ', $text);
$animerecord->setOtherTitles($other_titles);
}
# Synonyms:
$extracted = $leftcolumn->filterXPath('//span[text()="Synonyms:"]');
if ($extracted->count() > 0) {
$text = trim(str_replace($extracted->text(), '', $extracted->parents()->text()));
$other_titles['synonyms'] = explode(', ', $text);
$animerecord->setOtherTitles($other_titles);
}
# Japanese:
$extracted = $leftcolumn->filterXPath('//span[text()="Japanese:"]');
if ($extracted->count() > 0) {
$text = trim(str_replace($extracted->text(), '', $extracted->parents()->text()));
$other_titles['japanese'] = explode(', ', $text);
$animerecord->setOtherTitles($other_titles);
}
# Information section.
# Example:
# <h2>Information</h2>
# <div><span class="dark_text">Type:</span> TV</div>
# <div class="spaceit"><span class="dark_text">Episodes:</span> 24</div>
# <div><span class="dark_text">Status:</span> Finished Airing</div>
# <div class="spaceit"><span class="dark_text">Aired:</span> Apr 9, 2007 to Sep 17, 2007</div>
# <div>
# <span class="dark_text">Producers:</span>
# <a href="http://myanimelist.net/anime.php?p=2">Kyoto Animation</a>,
# <a href="http://myanimelist.net/anime.php?p=104">Lantis</a>,
# <a href="http://myanimelist.net/anime.php?p=262">Kadokawa Pictures USA</a><sup><small>L</small></sup>,
# <a href="http://myanimelist.net/anime.php?p=286">Bang Zoom! Entertainment</a>
# </div>
# <div class="spaceit">
# <span class="dark_text">Genres:</span>
# <a href="http://myanimelist.net/anime.php?genre[]=4">Comedy</a>,
# <a href="http://myanimelist.net/anime.php?genre[]=20">Parody</a>,
# <a href="http://myanimelist.net/anime.php?genre[]=23">School</a>,
# <a href="http://myanimelist.net/anime.php?genre[]=36">Slice of Life</a>
# </div>
# <div><span class="dark_text">Duration:</span> 24 min. per episode</div>
# <div class="spaceit"><span class="dark_text">Rating:</span> PG-13 - Teens 13 or older</div>
# Type:
$extracted = $leftcolumn->filterXPath('//span[text()="Type:"]');
if ($extracted->count() > 0) {
$animerecord->setType(trim(str_replace($extracted->text(), '', $extracted->parents()->text())));
}
# Episodes:
$extracted = $leftcolumn->filterXPath('//span[text()="Episodes:"]');
if ($extracted->count() > 0) {
$episodeCount = trim(str_replace($extracted->text(), '', $extracted->parents()->text()));
if (is_numeric($episodeCount)) {
$animerecord->setEpisodes((int) $episodeCount);
}
}
# Status:
$extracted = $leftcolumn->filterXPath('//span[text()="Status:"]');
if ($extracted->count() > 0) {
$animerecord->setStatus(strtolower(trim(str_replace($extracted->text(), '', $extracted->parents()->text()))));
}
# Aired:
$extracted = $leftcolumn->filterXPath('//span[text()="Aired:"]');
if ($extracted->count() > 0) {
/*
* NOTE: The Ruby API has a bug where yet-to-air shows that only have one date
* get that listed as the "end_date", not the "start_date". The code below fixes
* this and in doing so delibrately breaks compatibility in order to present the
* data properly.
*/
$daterange = explode(' to ', trim(str_replace($extracted->text(), '', $extracted->parents()->text())));
//MAL always provides record dates in US-style format.
if (strpos($daterange[0], ',') === false) {
if (strlen($daterange[0]) === 4) {
$animerecord->setStartDate(DateTime::createFromFormat('Y m d', $daterange[0] . ' 01 01'), 'year');
//Example ID 6535 or 9951
} elseif ($daterange[0] !== 'Not available') {
$animerecord->setStartDate(DateTime::createFromFormat('M Y d', $daterange[0] . ' 01'), 'month');
//Example ID 22535 (check upcoming list)
}
} else {
if (count(explode(' ', $daterange[0])) == 2) {
//MAL has been showing a comma with month and year (Jan, 2016), so catch that
$dateComponents = explode(' ', $daterange[0]);
$month = substr($dateComponents[0], 0, -1);
$year = $dateComponents[1];
$animerecord->setStartDate(DateTime::createFromFormat('M Y d', $month . ' ' . $year . ' 01'), 'month');
} elseif (strlen($daterange[0]) !== 7 && strlen($daterange[0]) !== 8) {
$animerecord->setStartDate(DateTime::createFromFormat('M j, Y', $daterange[0]), 'day');
}
}
//Series not yet to air won't list a range at all while currently airing series will use a "?"
//For these, we should return a null
if (count($daterange) > 1 && $daterange[1] !== '?') {
//MAL always provides record dates in US-style format.
//Not all dates are full, so we have to figure out how to properly parse them
$dateParts = explode(' ', $daterange[1]);
$firstIsNumber = is_numeric($dateParts[0]);
$hasComma = strpos($dateParts[0], ',');
if (count($dateParts) == 3) {
//Full date, normal processing
$endDate = DateTime::createFromFormat('M j, Y', $daterange[1]);
$animerecord->setEndDate($endDate, 'day');
} elseif (count($dateParts) == 2) {
//We only have two parts, figure out what we were given
if ($firstIsNumber === false && $hasComma !== false) {
//So, it looks like month and year, because MAL adds the comma regardless.
$endDate = DateTime::createFromFormat('M, Y d', $daterange[1] . ' 01');
//Example ID 21275
$animerecord->setEndDate($endDate, 'month');
}
} else {
if (count($dateParts) == 1 && $firstIsNumber) {
//Most likely just a year.
$endDate = DateTime::createFromFormat('Y m d', $daterange[1] . ' 01 01');
$animerecord->setEndDate($endDate, 'year');
//Example ID 11836
}
}
}
}
# Producers:
$extracted = $leftcolumn->filterXPath('//span[text()="Producers:"]');
if (strpos($extracted->parents()->text(), 'None found') === false && $extracted->count() > 0) {
$records = $extracted->parents()->first()->filter('a');
foreach ($records as $rItem) {
$producers[] = $rItem->nodeValue;
}
$animerecord->setProducers($producers);
}
# Genres:
$extracted = $leftcolumn->filterXPath('//span[text()="Genres:"]');
if ($extracted->count() > 0) {
$genres = array();
$records = $extracted->parents()->first()->filter('a');
foreach ($records as $rItem) {
$genres[] = $rItem->nodeValue;
}
if (count($genres) > 0) {
$animerecord->setGenres($genres);
}
}
# Classification:
$extracted = $leftcolumn->filterXPath('//span[text()="Rating:"]');
if ($extracted->count() > 0) {
$animerecord->setClassification(trim(str_replace($extracted->text(), '', $extracted->parents()->text())));
}
# Statistics
# Example:
# <h2>Statistics</h2>
# <div>
# <span class="dark_text">Score:</span> 8.41<sup><small>1</small></sup>
# <small>(scored by 22601 users)</small>
# </div>
# <div class="spaceit"><span class="dark_text">Ranked:</span> #96<sup><small>2</small></sup></div>
# <div><span class="dark_text">Popularity:</span> #15</div>
# <div class="spaceit"><span class="dark_text">Members:</span> 36,961</div>
# <div><span class="dark_text">Favorites:</span> 2,874</div>
//TODO: Rewrite to properly clean up excess tags.
# Score:
$extracted = $leftcolumn->filterXPath('//span[text()="Score:"]');
if ($extracted->count() > 0) {
$extracted = str_replace($extracted->text(), '', $extracted->parents()->text());
//Remove the parenthetical at the end of the string
$extracted = trim(str_replace(strstr($extracted, '('), '', $extracted));
//Sometimes there is a superscript number at the end from a note.
//Scores are only two decimals, so number_format should chop off the excess, hopefully.
if (strpos($extracted, 'N/A') === false) {
$animerecord->setMembersScore((double) number_format($extracted, 2));
}
}
# Popularity:
$extracted = $leftcolumn->filterXPath('//span[text()="Popularity:"]');
if ($extracted->count() > 0) {
$extracted = str_replace($extracted->text(), '', $extracted->parents()->text());
//Remove the hash at the front of the string and trim whitespace. Needed so we can cast to an int.
$extracted = trim(str_replace('#', '', $extracted));
$animerecord->setPopularityRank((int) $extracted);
}
# Members:
$extracted = $leftcolumn->filterXPath('//span[text()="Members:"]');
if ($extracted->count() > 0) {
$extracted = str_replace($extracted->text(), '', $extracted->parents()->text());
//PHP doesn't like commas in integers. Remove it.
$extracted = trim(str_replace(',', '', $extracted));
$animerecord->setMembersCount((int) $extracted);
}
# Members:
$extracted = $leftcolumn->filterXPath('//span[text()="Favorites:"]');
if ($extracted->count() > 0) {
$extracted = str_replace($extracted->text(), '', $extracted->parents()->text());
//PHP doesn't like commas in integers. Remove it.
$extracted = trim(str_replace(',', '', $extracted));
$animerecord->setFavoritedCount((int) $extracted);
}
# -
# Extract from sections on the right column: Synopsis, Related Anime, Characters & Voice Actors, Reviews
# Recommendations.
# -
$rightcolumn = $crawler->filterXPath('//div[@id="content"]/table/tr/td[2]');
# Synopsis
# Example:
# <td>
# <h2>Synopsis</h2>
# Having fun in school, doing homework together, cooking and eating, playing videogames, watching anime. All those little things make up the daily life of the anime- and chocolate-loving Izumi Konata and her friends. Sometimes relaxing but more than often simply funny! <br />
# -From AniDB
$extracted = $crawler->filterXPath('//span[@itemprop="description"]');
//Compatibility Note: We don't convert extended characters to HTML entities, we just
//use the output directly from MAL. This should be okay as our return charset is UTF-8.
$animerecord->setSynopsis('There is currently no synopsis for this title.');
if ($extracted->count() > 0) {
$animerecord->setSynopsis($extracted->html());
}
# Related Anime
# Example:
#<table class="anime_detail_related_anime" style="border-spacing:0px;">
# <tr>
# <td class="ar fw-n borderClass" nowrap="" valign="top">Adaptation:</td>
# <td class="borderClass" width="100%"><a href="/manga/587/Lucky☆Star">Lucky☆Star</a></td>
# </tr>
# <tr>
# <td class="ar fw-n borderClass" nowrap="" valign="top">Character:</td>
# <td class="borderClass" width="100%"><a href="/anime/3080/Anime_Tenchou">Anime Tenchou</a></td>
# </tr>
#</table>
$related = $rightcolumn->filter('table.anime_detail_related_anime');
//NOTE: Not all relations are currently supported.
if (iterator_count($related)) {
$rows = $related->children();
foreach ($rows as $row) {
$rowItem = $row->firstChild;
$relationType = strtr(strtolower(rtrim($rowItem->nodeValue, ':')), ' ', '_');
//This gets the next td containing the items
$relatedItem = $rowItem->nextSibling->firstChild;
do {
if ($relatedItem->nodeType !== XML_TEXT_NODE && $relatedItem->tagName == 'a') {
$url = $relatedItem->attributes->getNamedItem('href')->nodeValue;
$id = preg_match('/\\/(anime|manga)\\/(\\d+)\\/.*?/', $url, $urlParts);
if (count($urlParts) > 2) {
if ($id !== false || $id !== 0) {
$itemId = (int) $urlParts[2];
$itemTitle = $relatedItem->textContent;
$itemUrl = $url;
}
$itemArray = array();
if ($urlParts[1] == 'anime') {
$itemArray['anime_id'] = $itemId;
} else {
$itemArray['manga_id'] = $itemId;
}
$itemArray['title'] = $itemTitle;
$itemArray['url'] = 'https://myanimelist.net' . $itemUrl;
$animerecord->addRelation($itemArray, $relationType);
}
}
//Grab next item
$relatedItem = $relatedItem->nextSibling;
} while ($relatedItem !== null);
}
}
# Personal Info
$userPersonalInfo = $crawler->filterXPath('//h2[text()="Edit Status"]');
// Only try to parse personal info if the box is there
if ($userPersonalInfo->count() > 0) {
#Watched Status - Only available when user is authenticated
$my_data = $crawler->filter('select#myinfo_status');
if (iterator_count($my_data) && iterator_count($my_data->filter('option[selected="selected"]'))) {
$animerecord->setWatchedStatus($my_data->filter('option[selected="selected"]')->attr('value'));
}
#Watched Episodes - Only available when user is authenticated
$my_data = $crawler->filter('input#myinfo_watchedeps');
if (iterator_count($my_data)) {
$animerecord->setWatchedEpisodes((int) $my_data->attr('value'));
}
#User's Score - Only available when user is authenticated
$my_data = $crawler->filter('select#myinfo_score');
if (iterator_count($my_data) && iterator_count($my_data->filter('option[selected="selected"]'))) {
$animerecord->setScore((int) $my_data->filter('option[selected="selected"]')->attr('value'));
}
#Listed ID (?) - Only available when user is authenticated
$my_data = $crawler->filterXPath('//a[text()="Edit Details"]');
if (iterator_count($my_data)) {
if (preg_match('/id=(\\d+)/', $my_data->attr('href'), $my_data)) {
$animerecord->setListedAnimeId((int) $my_data[1]);
}
}
}
if ($apiVersion >= '2.1') {
# Background
preg_match('/div>Background<\\/h2>(.+?)<div/s', $crawler->filter('span[itemprop="description"]')->parents()->html(), $matches);
if (strpos($matches[0], 'No background information') !== false) {
$animerecord->setBackground('No background information has been added to this title.');
} else {
$animerecord->setBackground(trim($matches[1]));
}
# Broadcast:
$extracted = $leftcolumn->filterXPath('//span[text()="Broadcast:"]');
if ($extracted->count() > 0) {
$animerecord->setBroadcast(trim(preg_replace('/(\\w.+)s at(\\s\\d.+)\\((\\w.+)\\)/', '$1$2$3', str_replace($extracted->text(), '', $extracted->parents()->text()))));
}
# Duration:
$extracted = $leftcolumn->filterXPath('//span[text()="Duration:"]');
if ($extracted->count() > 0) {
$duration = trim(str_replace($extracted->text(), '', $extracted->parents()->text()));
// Handle varations that include minutes
if (strpos($duration, 'min.') !== false) {
if (strpos($duration, 'hr.') !== false) {
//contains hours and minutes
preg_match('/([0-9]+) hr\\. ([0-9]+) min\\./', $duration, $durationParts);
//This could all be done in one line, but it's more understandable and maintainable broken up.
$hours = (int) $durationParts[1];
$minutes = (int) $durationParts[2];
$animerecord->setDuration($hours * 60 + $minutes);
} else {
//contains only minutes
preg_match('/([0-9]+) min\\./', $duration, $durationParts);
$animerecord->setDuration((int) $durationParts[1]);
}
//Handle hour-only durations
} elseif (strpos($duration, 'hr.') !== false) {
preg_match('/([0-9]+) hr\\./', $duration, $durationParts);
$animerecord->setDuration((int) $durationParts[1] * 60);
}
// Any other format (such as just "Unknown") isn't understood and is ignored
}
// External links is only visible when an user has logged in any may be hidden on some records.
$externalLinks = $crawler->filterXPath('//h2[text()="External Links"]');
if ($externalLinks->count() > 0) {
$extracted = $externalLinks->nextAll()->filter('a');
# External Links:
foreach ($extracted as $externalLinkRow) {
$animerecord->setExternalLinks($externalLinkRow->nodeValue, $externalLinkRow->getAttribute('href'));
}
}
# Preview:
$extracted = $crawler->filter('div[class="video-promotion"] a');
if ($extracted->count() > 0) {
$animerecord->setPreview(preg_replace('/\\?(.+?)$/', '$2', $extracted->attr('href')));
}
# Opening Theme:
$extracted = $crawler->filter('div[class="theme-songs js-theme-songs opnening"] span');
foreach ($extracted as $openingRow) {
$animerecord->setOpeningTheme($openingRow->nodeValue);
}
# Ending Theme:
$extracted = $crawler->filter('div[class="theme-songs js-theme-songs ending"] span');
foreach ($extracted as $endingRow) {
$animerecord->setEndingTheme($endingRow->nodeValue);
}
# Recommendations
$extracted = $crawler->filter('div[id="anime_recommendation"] li[class="btn-anime"]');
foreach ($extracted as $recommendationsRow) {
$recommendationsCrawler = new Crawler($recommendationsRow);
$anime = new Anime();
// ID can be on either side of the hyphen
$tempId = preg_replace('/.*\\//', '', $recommendationsCrawler->filter('a')->attr('href'));
// Put the 2 IDs into $tempId
$anime->setId(preg_replace('/(-|' . $animerecord->getId() . ')/', '', $tempId));
// Remove the hyphen and the current anime's id, leaving the desired id
$anime->setTitle($recommendationsCrawler->filter('span')->text());
$anime->setImageUrl(preg_replace('/r(.+?)\\/(.+?)\\?(.+?)$/', '$2', $recommendationsCrawler->filter('img')->attr('data-src')));
$animerecord->setRecommendations($anime);
}
}
return $animerecord;
}
public function testEpisodes()
{
$episodes = rand();
$anime = new Anime();
$anime->setEpisodes($episodes);
$this->assertEquals($episodes, $anime->getEpisodes());
}