Add Bilibili, Youku and NetEase Music adapters

CHANGELOG.md

@@ -7,6 +7,10 @@ and this project adheres to [Semantic Versioning](http://semver.org/).
 ## [4.6.0] - 2026-10-30
 ### Added
 - `mastodon:iframe` setting (off by default). Mastodon oEmbed html (`blockquote.mastodon-embed` with `data-embed-url`) becomes a sandboxed `<iframe>` to the post's `/embed` page, so no `embed.js` is needed. Only same-host `https` embed URLs are used
+- Bilibili adapter: `bilibili.com/video/BV...` and `av...` URLs (also `b23.tv` short links) get a `player.bilibili.com` iframe built from the URL. The title, description and image of Bilibili's error pages ("出错啦!" and "视频去哪了呢？") are ignored, and so is a description template that was never filled [#243]
+- Youku adapter: `v.youku.com/v_show/id_<ID>.html` URLs get a `player.youku.com/embed` iframe built from the URL [#243]
+- NetEase Music adapter: songs, albums and playlists on `music.163.com` get an `outchain/player` iframe, and the author is read from the description. URLs with a `#/` route are requested without it. MV pages (`/mv?id=` and `/#/mv?id=`) get metadata only, because NetEase has no official embeddable MV player [#243]
+- `Extractor::normalizeUri()`, used by adapters to rewrite a URL before it is requested
 
 ## [4.5.4] - 2026-10-30
 ### Security
@@ -350,6 +354,7 @@ Full library refactoring.
 [#572]: https://github.com/oscarotero/Embed/issues/572
 [#573]: https://github.com/oscarotero/Embed/issues/573
 [#584]: https://github.com/php-embed/Embed/issues/584
+[#243]: https://github.com/php-embed/Embed/issues/243
 [#575]: https://github.com/php-embed/Embed/issues/575
 [#574]: https://github.com/oscarotero/Embed/issues/574
 [#577]: https://github.com/oscarotero/Embed/issues/577

README.md

@@ -263,6 +263,8 @@ The `Extractor` class has many `Detectors`. Each detector is responsible to dete
 
 So, an adapter is basically an extractor created specifically for a site. It can contains also custom detectors or apis. If you see the `src/Adapters` folder you can see all adapters.
 
+An adapter can also rewrite the URL before it is requested by overriding the static `normalizeUri()` method of its extractor. For example, the NetEase Music adapter turns `music.163.com/#/song?id=1` into `music.163.com/song?id=1`, because the part after `#` is never sent to the server.
+
 If you create an adapter, you need also register to Embed, so it knows in which website needs to use. To do that, there's the `ExtractorFactory` object, that is responsible for instantiate the right extractor for each site.
 
 ```php

src/Adapters/Bilibili/Detectors/Code.php

@@ -0,0 +1,33 @@
+<?php
+declare(strict_types = 1);
+
+namespace Embed\Adapters\Bilibili\Detectors;
+
+use Embed\Adapters\Bilibili\Extractor;
+use Embed\Detectors\Code as Detector;
+use Embed\EmbedCode;
+use function Embed\html;
+
+/**
+ * @extends Detector<\Embed\Adapters\Bilibili\Extractor>
+ */
+class Code extends Detector
+{
+    public function detect(): ?EmbedCode
+    {
+        $params = Extractor::parseVideoPath($this->extractor->getUri()->getPath());
+
+        if ($params === null) {
+            return parent::detect();
+        }
+
+        return new EmbedCode(html('iframe', [
+            'src' => 'https://player.bilibili.com/player.html?'.http_build_query($params + ['autoplay' => 0]),
+            'frameborder' => 0,
+            'allowfullscreen' => 'true',
+            'scrolling' => 'no',
+            'width' => 640,
+            'height' => 360,
+        ]), 640, 360);
+    }
+}

src/Adapters/Bilibili/Detectors/Description.php

@@ -0,0 +1,28 @@
+<?php
+declare(strict_types = 1);
+
+namespace Embed\Adapters\Bilibili\Detectors;
+
+use Embed\Adapters\Bilibili\Extractor;
+use Embed\Detectors\Description as Detector;
+
+/**
+ * @extends Detector<\Embed\Adapters\Bilibili\Extractor>
+ */
+class Description extends Detector
+{
+    public function detect(): ?string
+    {
+        if ($this->extractor->isErrorPage()) {
+            return null;
+        }
+
+        $description = parent::detect();
+
+        if ($description !== null && Extractor::isTemplate($description)) {
+            return null;
+        }
+
+        return $description;
+    }
+}

src/Adapters/Bilibili/Detectors/Image.php

@@ -0,0 +1,22 @@
+<?php
+declare(strict_types = 1);
+
+namespace Embed\Adapters\Bilibili\Detectors;
+
+use Embed\Detectors\Image as Detector;
+use Psr\Http\Message\UriInterface;
+
+/**
+ * @extends Detector<\Embed\Adapters\Bilibili\Extractor>
+ */
+class Image extends Detector
+{
+    public function detect(): ?UriInterface
+    {
+        if ($this->extractor->isErrorPage()) {
+            return null;
+        }
+
+        return parent::detect();
+    }
+}

src/Adapters/Bilibili/Detectors/Title.php

@@ -0,0 +1,21 @@
+<?php
+declare(strict_types = 1);
+
+namespace Embed\Adapters\Bilibili\Detectors;
+
+use Embed\Detectors\Title as Detector;
+
+/**
+ * @extends Detector<\Embed\Adapters\Bilibili\Extractor>
+ */
+class Title extends Detector
+{
+    public function detect(): ?string
+    {
+        if ($this->extractor->isErrorPage()) {
+            return null;
+        }
+
+        return parent::detect();
+    }
+}

src/Adapters/Bilibili/Extractor.php

@@ -0,0 +1,72 @@
+<?php
+declare(strict_types = 1);
+
+namespace Embed\Adapters\Bilibili;
+
+use Embed\Extractor as Base;
+
+/**
+ * @template-extends Base<\Embed\Detectors\Detector<self>>
+ */
+class Extractor extends Base
+{
+    /**
+     * Titles of the pages Bilibili serves when the video is not available,
+     * for example when the request comes from a restricted region.
+     */
+    private const ERROR_TITLES = ['出错啦', '视频去哪了呢'];
+
+    public function createCustomDetectors(): array
+    {
+        return [
+            'code' => new Detectors\Code($this),
+            'title' => new Detectors\Title($this),
+            'description' => new Detectors\Description($this),
+            'image' => new Detectors\Image($this),
+        ];
+    }
+
+    public function isErrorPage(): bool
+    {
+        $title = $this->getDocument()->select('.//head/title')->str();
+
+        if ($title === null) {
+            return false;
+        }
+
+        foreach (self::ERROR_TITLES as $errorTitle) {
+            if (strpos($title, $errorTitle) !== false) {
+                return true;
+            }
+        }
+
+        return false;
+    }
+
+    /**
+     * Error pages may also carry a description template that was never
+     * filled, with placeholders like {$0} or undefined.
+     */
+    public static function isTemplate(string $text): bool
+    {
+        return preg_match('/\{\$\d+\}|\bundefined\b/', $text) === 1;
+    }
+
+    /**
+     * Returns the player query (bvid or aid) for a video path.
+     *
+     * @return array<string, string>|null
+     */
+    public static function parseVideoPath(string $path): ?array
+    {
+        if (preg_match('#^/video/(BV[0-9A-Za-z]{10})(?:/|$)#', $path, $matches) === 1) {
+            return ['bvid' => $matches[1]];
+        }
+
+        if (preg_match('#^/video/av(\d+)(?:/|$)#i', $path, $matches) === 1) {
+            return ['aid' => $matches[1]];
+        }
+
+        return null;
+    }
+}

src/Adapters/NetEase/Detectors/AuthorName.php

@@ -0,0 +1,29 @@
+<?php
+declare(strict_types = 1);
+
+namespace Embed\Adapters\NetEase\Detectors;
+
+use Embed\Detectors\AuthorName as Detector;
+
+/**
+ * @extends Detector<\Embed\Adapters\NetEase\Extractor>
+ */
+class AuthorName extends Detector
+{
+    public function detect(): ?string
+    {
+        $result = parent::detect();
+
+        if ($result !== null) {
+            return $result;
+        }
+
+        $description = $this->extractor->getMetas()->str('description', 'og:description');
+
+        if ($description !== null && preg_match('/由\s*(.+?)\s*演唱/u', $description, $matches) === 1) {
+            return $matches[1];
+        }
+
+        return null;
+    }
+}

src/Adapters/NetEase/Detectors/Code.php

@@ -0,0 +1,39 @@
+<?php
+declare(strict_types = 1);
+
+namespace Embed\Adapters\NetEase\Detectors;
+
+use Embed\Adapters\NetEase\Extractor;
+use Embed\Detectors\Code as Detector;
+use Embed\EmbedCode;
+use function Embed\html;
+
+/**
+ * @extends Detector<\Embed\Adapters\NetEase\Extractor>
+ */
+class Code extends Detector
+{
+    public function detect(): ?EmbedCode
+    {
+        $item = Extractor::parseUri($this->extractor->getUri());
+
+        if ($item === null) {
+            return parent::detect();
+        }
+
+        $query = http_build_query([
+            'type' => $item[0],
+            'id' => $item[1],
+            'auto' => 0,
+            'height' => 66,
+        ]);
+
+        return new EmbedCode(html('iframe', [
+            'src' => "https://music.163.com/outchain/player?{$query}",
+            'frameborder' => 0,
+            'scrolling' => 'no',
+            'width' => 330,
+            'height' => 86,
+        ]), 330, 86);
+    }
+}

src/Adapters/NetEase/Extractor.php

@@ -0,0 +1,69 @@
+<?php
+declare(strict_types = 1);
+
+namespace Embed\Adapters\NetEase;
+
+use Embed\Extractor as Base;
+use Psr\Http\Message\UriInterface;
+
+/**
+ * @template-extends Base<\Embed\Detectors\Detector<self>>
+ */
+class Extractor extends Base
+{
+    /** Player types used by music.163.com/outchain/player */
+    private const TYPES = [
+        'song' => 2,
+        'album' => 1,
+        'playlist' => 0,
+    ];
+
+    public function createCustomDetectors(): array
+    {
+        return [
+            'code' => new Detectors\Code($this),
+            'authorName' => new Detectors\AuthorName($this),
+        ];
+    }
+
+    /**
+     * NetEase Music URLs like /#/song?id=1 keep the route in the fragment,
+     * which is never sent to the server. Move it to the path and query.
+     * MV pages (/mv?id=1) are normalized too, but NetEase has no official
+     * embeddable MV player, so they only get metadata.
+     */
+    public static function normalizeUri(UriInterface $uri): UriInterface
+    {
+        $fragment = $uri->getFragment();
+
+        if (preg_match('#^/(?:m/)?(song|album|playlist|mv)\?(.*)$#', $fragment, $matches) !== 1) {
+            return $uri;
+        }
+
+        return $uri->withPath('/'.$matches[1])->withQuery($matches[2])->withFragment('');
+    }
+
+    /**
+     * Returns the player type and the id of a song, album or playlist.
+     *
+     * @return array{0: int, 1: string}|null
+     */
+    public static function parseUri(UriInterface $uri): ?array
+    {
+        $path = $uri->getPath();
+        $type = self::TYPES[trim($path, '/')] ?? null;
+
+        if ($type === null) {
+            return null;
+        }
+
+        parse_str($uri->getQuery(), $query);
+        $id = $query['id'] ?? null;
+
+        if (!is_string($id) || preg_match('/^\d+$/', $id) !== 1) {
+            return null;
+        }
+
+        return [$type, $id];
+    }
+}

src/Adapters/Youku/Detectors/Code.php

@@ -0,0 +1,33 @@
+<?php
+declare(strict_types = 1);
+
+namespace Embed\Adapters\Youku\Detectors;
+
+use Embed\Adapters\Youku\Extractor;
+use Embed\Detectors\Code as Detector;
+use Embed\EmbedCode;
+use function Embed\html;
+
+/**
+ * @extends Detector<\Embed\Adapters\Youku\Extractor>
+ */
+class Code extends Detector
+{
+    public function detect(): ?EmbedCode
+    {
+        $id = Extractor::parseVideoPath($this->extractor->getUri()->getPath());
+
+        if ($id === null) {
+            return parent::detect();
+        }
+
+        return new EmbedCode(html('iframe', [
+            'src' => 'https://player.youku.com/embed/'.rawurlencode($id),
+            'frameborder' => 0,
+            'allowfullscreen' => 'true',
+            'scrolling' => 'no',
+            'width' => 640,
+            'height' => 360,
+        ]), 640, 360);
+    }
+}