public static function parse($contents)
{
$crawler = new Crawler();
$crawler->addHTMLContent($contents, 'UTF-8');
$personrecord = new Person();
# Person ID.
# Example:
# <input type="hidden" name="vaid" value="185">
$personrecord->setId((int) $crawler->filter('input[name="vaid"]')->attr('value'));
# Name
# Example:
# <div><h1 class="h1">Hanazawa, Kana</h1></div>
$personrecord->setName(trim($crawler->filter('div h1')->text()));
# Image
# Example:
# <a href="/people/185/Kana_Hanazawa/pictures"><img src="http://cdn.myanimelist.net/images/voiceactors/3/43500.jpg" alt="Hanazawa, Kana"></a>
$personrecord->setImageUrl($crawler->filter('div#content tr td div img')->attr('src'));
$leftcolumn = $crawler->filterXPath('//div[@id="content"]/table/tr/td[@class="borderClass"]');
# Given name
$extracted = $leftcolumn->filterXPath('//span[text()="Given name:"]');
if ($extracted->count() > 0) {
$personrecord->setGivenName(trim(str_replace($extracted->text(), '', $extracted->parents()->text())));
}
# Family name
# MAL messed this field up. It's not wrapped in a div, so the text is floating out in the td.
$extracted = $leftcolumn->filterXPath('//span[text()="Family name:"]');
if ($extracted->count() > 0) {
$matches = array();
# This regex matches "Family name:..." until it hits Birthday/Website/Alternate [name], one of which should
# be the field following the Family name field
preg_match('/Family name:.*?(?:(?!Birthday|Website|Alternate).)*/', $leftcolumn->text(), $matches);
if (count($matches) > 0) {
$personrecord->setFamilyName(trim(str_replace($extracted->text(), '', $matches[0])));
}
}
# Alternate names
$extracted = $leftcolumn->filterXPath('//span[text()="Alternate names:"]');
if ($extracted->count() > 0) {
$text = trim(str_replace($extracted->text(), '', $extracted->parents()->text()));
$alternateNames = explode(', ', $text);
$personrecord->setAlternateNames($alternateNames);
}
# Birthday
$extracted = $leftcolumn->filterXPath('//span[text()="Birthday:"]');
if ($extracted->count() > 0) {
$dateStr = trim(str_replace($extracted->text(), '', $extracted->parents()->text()));
$dateStr = str_replace(' ', ' ', $dateStr);
// Replace 2 spaces with 1, MAL seems to add 2 spaces in some dates
if (strpos($dateStr, ',') === false) {
if (strlen($dateStr) === 4) {
// Only a year, Example ID 11746
$personrecord->setBirthday(DateTime::createFromFormat('Y m d', $dateStr . ' 01 01'), 'year');
} elseif (count(explode(' ', $dateStr)) == 2) {
// Month + Year, i.e. "Feb 1999", Example ID 7277
$dateComponents = explode(' ', $dateStr);
$month = $dateComponents[0];
$year = $dateComponents[1];
$personrecord->setBirthday(DateTime::createFromFormat('M Y d', $month . ' ' . $year . ' 01'), 'month');
}
} else {
$dateComponents = explode(' ', $dateStr);
if (count($dateComponents) == 2) {
// Month + Day, i.e. "Jun 15,", Example ID 2608
$month = $dateComponents[0];
$day = substr($dateComponents[1], 0, -1);
$personrecord->setBirthday(DateTime::createFromFormat('M d Y', $month . ' ' . $day . ' 1970'), 'dayMonth');
} elseif (count($dateComponents) == 3) {
// Full date, i.e. "Feb 25, 1989", Example ID 185
$personrecord->setBirthday(DateTime::createFromFormat('M j, Y', $dateStr), 'day');
}
}
}
# Website
# This isn't in a div, but the anchor element is the next sibling of the span
$extracted = $leftcolumn->filterXPath('//span[text()="Website:"]');
if ($extracted->count() > 0) {
$personrecord->setWebsiteUrl(trim(str_replace($extracted->text(), '', $extracted->nextAll()->attr("href"))));
}
# Favorites count
$extracted = $leftcolumn->filterXPath('//span[text()="Member Favorites:"]');
if ($extracted->count() > 0) {
$personrecord->setFavoritedCount(trim(str_replace($extracted->text(), '', $extracted->parents()->text())));
}
# More Details
# Note: CSS classes are misspelled, need to keep an eye on this
$extracted = $leftcolumn->filter('div[class="people-informantion-more js-people-informantion-more"]');
if ($extracted->count() > 0) {
$personrecord->setMoreDetails($extracted->html());
}
# Extract from sections on the right column: Voice acting roles, anime staff positions, published manga
$rightcolumn = $crawler->filter('div[id="content"] td[style="padding-left: 5px;"]');
// Voice acting roles
$vaRoles = $rightcolumn->filterXPath('//div[text()="Voice Acting Roles"]');
if ($vaRoles->count() > 0) {
$rolesArray = array();
// Iterate over each row in the table
foreach ($vaRoles->nextAll()->children() as $item) {
$node = new Crawler($item);
// Fill in the character details
$characterAnchor = $node->filterXPath('//td[3]/a');
$characterName = $characterAnchor->text();
$characterUrl = $characterAnchor->attr('href');
$isMainCharacer = strpos($node->filterXPath('//td[3]/div')->text(), "Main") !== false;
$characterImage = preg_replace('/r(.+?)\\/(.+?)\\?(.+?)$/', '$2', $node->filterXPath('//td[4]/div/a/img')->attr('data-src'));
$match = preg_match('/\\/(character)\\/(\\d+)\\/.*?/', $characterUrl, $urlParts);
if ($match !== false && $match !== 0) {
$characterId = (int) $urlParts[2];
}
$itemArray = array('id' => $characterId, 'name' => $characterName, 'image_url' => $characterImage, 'main_role' => $isMainCharacer);
$itemArray['anime'] = new Anime();
$animeDetails = $node->filterXPath('//td[2]/a');
// Fill in the anime details
$itemArray['anime']->setImageUrl(preg_replace('/r(.+?)\\/(.+?)\\?(.+?)$/', '$2', $node->filterXPath('//td[1]/div/a/img')->attr('data-src')));
$itemArray['anime']->setTitle($animeDetails->text());
$match = preg_match('/\\/(anime)\\/(\\d+)\\/.*?/', $animeDetails->attr('href'), $urlParts);
if ($match !== false && $match !== 0) {
$itemArray['anime']->setId($urlParts[2]);
}
$rolesArray[] = $itemArray;
}
$personrecord->setVoiceActingRoles($rolesArray);
}
// Anime staff positions
$staffPositions = $rightcolumn->filterXPath('//div[text()="Anime Staff Positions"]');
if ($staffPositions->count() > 0) {
$positionsArray = array();
// Iterate over each row in the table
foreach ($staffPositions->nextAll()->children() as $item) {
$node = new Crawler($item);
// Fill in the position details
$position = $node->filterXPath('//td[2]/div/small')->text();
$positionDetails = $node->filterXPath('//td[2]/div')->text();
// Details are wrapped in parenthesis, so we'll find those parens and grab what's inside
// Sometimes what comes back is malformed - i.e. "ep. 1-15)," - it appears malformed on the site too
$match = preg_match('/\\((.*)\\)/', $positionDetails, $positionDetailsParts);
if ($match !== false && $match !== 0) {
$positionDetails = $positionDetailsParts[1];
} else {
$positionDetails = null;
}
$itemArray = array('position' => $position, 'details' => $positionDetails);
$itemArray['anime'] = new Anime();
$animeDetails = $node->filterXPath('//td[2]/a');
// Fill in the anime details
$itemArray['anime']->setImageUrl(preg_replace('/r(.+?)\\/(.+?)\\?(.+?)$/', '$2', $node->filterXPath('//td[1]/div/a/img')->attr('data-src')));
$itemArray['anime']->setTitle($animeDetails->text());
$match = preg_match('/\\/(anime)\\/(\\d+)\\/.*?/', $animeDetails->attr('href'), $urlParts);
if ($match !== false && $match !== 0) {
$itemArray['anime']->setId($urlParts[2]);
}
$positionsArray[] = $itemArray;
}
$personrecord->setAnimeStaffPositions($positionsArray);
}
// Published manga
$mangaPositions = $rightcolumn->filterXPath('//div[text()="Published Manga"]');
if ($mangaPositions->count() > 0) {
$positionsArray = array();
// Iterate over each row in the table
foreach ($mangaPositions->nextAll()->children() as $item) {
$node = new Crawler($item);
// Fill in the position
$position = $node->filterXPath('//td[2]/div/small')->text();
$itemArray = array('position' => $position);
$itemArray['manga'] = new Manga();
$mangaDetails = $node->filterXPath('//td[2]/a');
// Fill in the manga details
$itemArray['manga']->setImageUrl(preg_replace('/r(.+?)\\/(.+?)\\?(.+?)$/', '$2', $node->filterXPath('//td[1]/div/a/img')->attr('data-src')));
$itemArray['manga']->setTitle($mangaDetails->text());
$match = preg_match('/\\/(manga)\\/(\\d+)\\/.*?/', $mangaDetails->attr('href'), $urlParts);
if ($match !== false && $match !== 0) {
$itemArray['manga']->setId($urlParts[2]);
}
$positionsArray[] = $itemArray;
}
$personrecord->setPublishedManga($positionsArray);
}
return $personrecord;
}
public function testPublishedManga()
{
$person = new Person();
$itemArray = array('position' => 'Story');
$itemArray['manga'] = new Manga();
$itemArray['manga']->setId(rand());
$itemArray['manga']->setTitle('The Title');
$itemArray['manga']->setImageUrl('http://www.example.com/image.jpg');
$publishedManga = array($itemArray);
$person->setPublishedManga($publishedManga);
$this->assertEquals($publishedManga, $person->getPublishedManga());
}
