fix(scraper): extract real artist slugs and parse current uploader markup

__mocks__/fixtures/video-detail.html

@@ -16,8 +16,28 @@
 <a class="item btn_link video_meta_pill" href="https://rule34video.com/members/99999/">
   <div class="wrap_image"><img src="https://rule34video.com/contents/avatars/0/99999.jpg" alt="TestUser"/></div>
   TestUser
+  <div class="verified-status"><svg class="custom-svg custom-verified"><use xlink:href="#custom-verified"></use></svg></div>
+</a>
+<span class="tag_vote_chip video_meta_vote_chip js-video-vote-chip" data-video-id="4406161" data-item-type="model" data-item-id="1" data-status="normal">
+  <span class="tag_vote_progress" aria-hidden="true"></span>
+  <a class="item btn_link video_meta_pill" href="https://rule34video.com/models/starfire/">
+    <div class="wrap_image"><img src="https://rule34video.com/contents/models/1/s1.jpg" alt="Starfire"/></div>
+    <span class="name">Starfire</span>
+  </a>
+  <span class="tag_vote_controls"></span>
+</span>
+<span class="tag_vote_chip video_meta_vote_chip js-video-vote-chip" data-video-id="4406161" data-item-type="model" data-item-id="2" data-status="normal">
+  <span class="tag_vote_progress" aria-hidden="true"></span>
+  <a class="item btn_link video_meta_pill" href="https://rule34video.com/models/opennsfw/">
+    <div class="wrap_image"><img src="https://rule34video.com/contents/models/2/s1.png" alt="OpenNSFW (VA)"/></div>
+    <span class="name">OpenNSFW (VA)</span>
+  </a>
+  <span class="tag_vote_controls"></span>
+</span>
+<a href="https://rule34video.com/models/jackerman/" class="item">
+  <div class="position">1</div>
+  <div class="name">Jackerman</div>
 </a>
-<a class="item btn_link" href="https://rule34video.com/models/starfire/"><span class="name">Starfire</span></a>
 <a class="tag_item" href="https://rule34video.com/tags/12345/">test tag</a>
 <a class="tag_item" href="https://rule34video.com/tags/67890/">animation</a>
 <a href="https://rule34video.com/categories/test-category/">Test Category</a>

src/api/common/r34.test.ts

@@ -65,7 +65,10 @@ describe('r34Client', () => {
 
       expect(result.uploader).toBe('TestUser');
       expect(result.uploaderMemberId).toBe('99999');
-      expect(result.artist).toBe('Starfire');
+      expect(result.artists).toEqual([
+        { name: 'Starfire', slug: 'starfire' },
+        { name: 'OpenNSFW (VA)', slug: 'opennsfw' },
+      ]);
       expect(result.formats.length).toBeGreaterThan(0);
     });
   });

src/app/post/[id].tsx

@@ -9,6 +9,7 @@ import { useVideoDetail } from '@/api/video-queries';
 import { ActivityIndicator, Button, FocusAwareStatusBar, Text } from '@/components/ui';
 import { useVideoDownload } from '@/lib/hooks';
 import { useTranslate } from '@/lib/i18n';
+import { artistChipLabel } from '@/lib/r34/artists';
 import { toOfflineDetail } from '@/lib/r34/offline-detail';
 import type { VideoDetail, VideoFormat } from '@/lib/r34/types';
 import { baseIdOf, useDownloadedStore } from '@/lib/stores/downloaded-store';
@@ -312,23 +313,27 @@ export default function Post(): React.ReactElement | null {
           </View>
 
           <View className="mb-6 flex-row flex-wrap items-center gap-4">
-            {data.artist ? (
+            {(data.artists ?? []).map((artist) => (
               <Link
+                key={artist.slug}
                 href={{
                   pathname: '/model/[slug]',
-                  params: { slug: data.artist.toLowerCase().replace(/\s+/g, '-') },
+                  // Navigate with the real site slug from the scraper —
+                  // deriving it from the display name 404s for names like
+                  // "OpenNSFW (VA)" (actual slug: "opennsfw").
+                  params: { slug: artist.slug },
                 }}
                 asChild
               >
                 <TouchableOpacity>
                   <View className="flex-row items-center rounded-full bg-primary-100 px-3 py-1.5 dark:bg-primary-900/30">
                     <Text className="text-sm font-semibold text-primary-900 dark:text-primary-100">
-                      Artist: {data.artist}
+                      {artistChipLabel(artist.name)}
                     </Text>
                   </View>
                 </TouchableOpacity>
               </Link>
-            ) : null}
+            ))}
 
             {data.uploader && data.uploaderMemberId ? (
               <Link

src/lib/r34/artists.test.ts

@@ -0,0 +1,26 @@
+import { artistChipLabel } from './artists';
+
+describe('artistChipLabel', () => {
+  it('uses the generic Artist prefix for plain names', () => {
+    expect(artistChipLabel('Jackerman')).toBe('Artist: Jackerman');
+    expect(artistChipLabel('General Butch')).toBe('Artist: General Butch');
+  });
+
+  it('promotes known role suffixes to the prefix and strips them', () => {
+    expect(artistChipLabel('OpenNSFW (VA)')).toBe('VA: OpenNSFW');
+    expect(artistChipLabel('MizzPeachy (VA)')).toBe('VA: MizzPeachy');
+    expect(artistChipLabel('HentAudio (Audio)')).toBe('Audio: HentAudio');
+    expect(artistChipLabel('Someone (Artist)')).toBe('Artist: Someone');
+  });
+
+  it('keeps unknown parentheticals as part of the name', () => {
+    // Only VA/Audio/Artist are known roles — anything else stays verbatim so
+    // meaningful suffixes aren't silently dropped.
+    expect(artistChipLabel('Studio X (official)')).toBe('Artist: Studio X (official)');
+  });
+
+  it('handles degenerate inputs without matching', () => {
+    expect(artistChipLabel('(VA)')).toBe('Artist: (VA)');
+    expect(artistChipLabel('')).toBe('Artist: ');
+  });
+});

src/lib/r34/artists.ts

@@ -0,0 +1,18 @@
+/**
+ * Chip label for a site model (artist). The site appends a role to some model
+ * display names in parentheses — "(VA)" for voice actors, "(Audio)" for audio
+ * work — e.g. "OpenNSFW (VA)". Promote a known role to the label prefix and
+ * strip it from the name: "VA: OpenNSFW". Models without a known role keep
+ * the generic "Artist: Name" (with their name untouched, parenthetical
+ * included — it may carry meaning we don't want to silently drop).
+ */
+const ROLE_PREFIXES = new Set(['VA', 'Audio', 'Artist']);
+
+export function artistChipLabel(name: string): string {
+  const match = name.match(/^(.+?)\s*\(([^()]+)\)$/);
+  const role = match?.[2].trim();
+  if (match && role && ROLE_PREFIXES.has(role)) {
+    return `${role}: ${match[1].trim()}`;
+  }
+  return `Artist: ${name}`;
+}

src/lib/r34/offline-detail.test.ts

@@ -37,7 +37,8 @@ describe('toOfflineDetail', () => {
 
     expect(detail.tags).toEqual([]);
     expect(detail.categories).toEqual([]);
-    expect(detail.artist).toBeUndefined();
+    expect(detail.artists).toEqual([]);
+    expect(detail.uploader).toBeUndefined();
     expect(detail.description).toBeUndefined();
   });
 });

src/lib/r34/offline-detail.ts

@@ -5,7 +5,7 @@ import type { VideoDetail } from './types';
 /**
  * Build a minimal VideoDetail from a downloaded item so the post page can
  * render (and play the local file) with no network. Rich metadata
- * (tags/categories/artist) is unavailable offline by design.
+ * (tags/categories/artists) is unavailable offline by design.
  */
 export function toOfflineDetail(meta: DownloadMetadata, id: string): VideoDetail {
   return {
@@ -19,5 +19,6 @@ export function toOfflineDetail(meta: DownloadMetadata, id: string): VideoDetail
     formats: [{ url: meta.uri, quality: meta.quality, ext: 'mp4' }],
     tags: [],
     categories: [],
+    artists: [],
   };
 }

src/lib/r34/scraper.test.ts

@@ -98,11 +98,46 @@ describe('scraper', () => {
       expect(detail.tags).toContainEqual({ id: '67890', name: 'animation' });
       expect(detail.uploader).toBe('TestUser');
       expect(detail.uploaderMemberId).toBe('99999');
-      expect(detail.artist).toBe('Starfire');
+      expect(detail.artists).toEqual([
+        { name: 'Starfire', slug: 'starfire' },
+        { name: 'OpenNSFW (VA)', slug: 'opennsfw' },
+      ]);
       expect(detail.categories).toContain('Test Category');
       expect(detail.description).toBe('A test video description for Starfire.');
     });
 
+    it('does not treat sidebar "Top Artists" links as video artists', () => {
+      const html = loadFixture('video-detail.html');
+      const detail = parseVideoDetail(html, '4406161', 'starfire-quick-hot-encounter');
+
+      // The fixture contains a sidebar-style link (class="item", no btn_link)
+      // to /models/jackerman/ that must not bleed into the artist list.
+      expect(detail.artists.some((a) => a.slug === 'jackerman')).toBe(false);
+    });
+
+    it('navigates artists by their real slug, not a name-derived one', () => {
+      const html = loadFixture('video-detail.html');
+      const detail = parseVideoDetail(html, '4406161', 'starfire-quick-hot-encounter');
+
+      // "OpenNSFW (VA)" used to become the slug "opennsfw-(va)" (a 404 on the
+      // site); the real slug from the href is "opennsfw".
+      const derived = detail.artists[1].name.toLowerCase().replace(/\s+/g, '-');
+      expect(derived).not.toBe(detail.artists[1].slug);
+      expect(detail.artists[1].slug).toBe('opennsfw');
+    });
+
+    it('falls back to the anchor text when the uploader has no avatar', () => {
+      const html = `
+        <html><body><title>Minimal - Rule34Video</title>
+        <div>Uploaded by</div>
+        <a class="item btn_link video_meta_pill" href="https://rule34video.com/members/4242/">NoAvatarUser</a>
+        </body></html>`;
+      const detail = parseVideoDetail(html, '1', 'x');
+
+      expect(detail.uploaderMemberId).toBe('4242');
+      expect(detail.uploader).toBe('NoAvatarUser');
+    });
+
     it('does not expose removed uploaderUrl field', () => {
       const html = loadFixture('video-detail.html');
       const detail = parseVideoDetail(html, '4406161', 'starfire-quick-hot-encounter') as Record<
@@ -122,6 +157,9 @@ describe('scraper', () => {
       expect(detail.title).toBe('Minimal');
       expect(detail.formats).toEqual([]);
       expect(detail.tags).toEqual([]);
+      expect(detail.artists).toEqual([]);
+      expect(detail.uploader).toBeUndefined();
+      expect(detail.uploaderMemberId).toBeUndefined();
     });
   });
 

src/lib/r34/scraper.ts

@@ -159,19 +159,43 @@ export function parseVideoDetail(html: string, videoId: string, slug: string): V
     categories.push(match[1].trim());
   }
 
-  // Extract uploader (a site member). Capture the member id from the profile
-  // link and the name from the avatar's alt attribute.
+  // Extract the uploader (a site member). Anchor on the "Uploaded by" label,
+  // then bind the name to the member anchor's own inner HTML — the avatar's
+  // alt text first, plain text (tags stripped) as a fallback for members
+  // without an avatar. This can't bleed into unrelated elements further down
+  // the page the way a lazy document-wide scan could.
   const uploaderMatch = html.match(
-    /Uploaded by<\/div>\s*<a[^>]+href="(?<url>https:\/\/rule34video\.com\/members\/(?<id>\d+)\/)"[\s\S]*?alt="(?<name>[^"]*)"/i
+    /Uploaded by<\/div>\s*<a[^>]*href="https:\/\/rule34video\.com\/members\/(\d+)\/"[^>]*>([\s\S]*?)<\/a>/i
   );
-  const uploader = uploaderMatch?.groups?.name?.trim();
-  const uploaderMemberId = uploaderMatch?.groups?.id;
-
-  // Extract artist
-  const artistMatch = html.match(
-    /<a[^>]+class="item btn_link"[^>]+href="(?<url>[^"]+)"[^>]*>\s*<span class="name">\s*(?<name>[^<]+)\s*<\/span>\s*<\/a>/i
-  );
-  const artist = artistMatch?.groups?.name?.trim();
+  const uploaderMemberId = uploaderMatch?.[1];
+  const uploaderInner = uploaderMatch?.[2] ?? '';
+  const uploader =
+    uploaderInner.match(/alt="([^"]*)"/)?.[1]?.trim() ||
+    uploaderInner
+      .replace(/<[^>]*>/g, ' ')
+      .replace(/\s+/g, ' ')
+      .trim() ||
+    undefined;
+
+  // Extract artists (site "models"). Each artist is a pill anchor (class
+  // contains "item btn_link") linking to /models/{slug}/ — the sidebar's "Top
+  // Artists" links (class="item", no btn_link) must not match. The slug comes
+  // from the href itself: deriving it from the display name breaks for names
+  // like "OpenNSFW (VA)", whose real site slug is "opennsfw".
+  // NOTE: positional capture groups (match[1..n]) — Hermes' matchAll does not
+  // populate `.groups` for named captures the way V8 does.
+  const artists: { name: string; slug: string }[] = [];
+  const artistRegex =
+    /<a[^>]*class="[^"]*\bitem btn_link\b[^"]*"[^>]*href="https:\/\/rule34video\.com\/models\/([^\/"]+)\/"[^>]*>([\s\S]*?)<\/a>/g;
+  for (const match of html.matchAll(artistRegex)) {
+    const slug = match[1];
+    if (artists.some((a) => a.slug === slug)) continue;
+    const name =
+      match[2].match(/<span class="name">\s*([^<]+?)\s*<\/span>/)?.[1]?.trim() ||
+      match[2].match(/alt="([^"]*)"/)?.[1]?.trim() ||
+      slug;
+    artists.push({ name, slug });
+  }
 
   // Extract description
   const descriptionMatch = html.match(/<script type="application\/ld\+json">([\s\S]*?)<\/script>/);
@@ -197,7 +221,7 @@ export function parseVideoDetail(html: string, videoId: string, slug: string): V
     formats,
     tags,
     categories,
-    artist,
+    artists,
     uploader,
     uploaderMemberId,
     description,

src/lib/r34/types.ts

@@ -20,11 +20,18 @@ export type VideoTag = {
   name: string;
 };
 
+/** A site "model" (artist) with its real URL slug — names don't map to slugs. */
+export type VideoArtist = {
+  name: string;
+  slug: string;
+};
+
 export type VideoDetail = VideoListItem & {
   formats: VideoFormat[];
   tags: VideoTag[];
   categories: string[];
-  artist?: string;
+  /** Artists (site models) — a video can have several. */
+  artists: VideoArtist[];
   uploader?: string;
   uploaderMemberId?: string;
   description?: string;