Added better artist bio/pics/lyrics

This commit is contained in:
Connor Johnstone
2026-03-19 14:53:39 -04:00
parent f6b363c40f
commit 51bcf26482
9 changed files with 580 additions and 23 deletions

View File

@@ -310,17 +310,17 @@ pub async fn enrich_artist(
let local_id = a.id;
(a, Some(local_id), mbid)
} else {
// Look up artist name from MusicBrainz by MBID — don't create a local record
let (name, _disambiguation) = state
// Look up artist info from MusicBrainz by MBID — don't create a local record
let info = state
.mb_client
.get_artist_by_mbid(&mbid)
.get_artist_info(&mbid)
.await
.map_err(|e| ApiError::NotFound(format!("artist MBID {mbid} not found: {e}")))?;
// Create a synthetic artist object for display only (not saved to DB)
let synthetic = shanty_db::entities::artist::Model {
id: 0,
name,
name: info.name.clone(),
musicbrainz_id: Some(mbid.clone()),
added_at: chrono::Utc::now().naive_utc(),
top_songs: "[]".to_string(),
@@ -330,6 +330,27 @@ pub async fn enrich_artist(
}
};
// Fetch detailed artist info (country, type, URLs) — best-effort
let artist_info = match state.mb_client.get_artist_info(&mbid).await {
Ok(info) => {
tracing::debug!(
mbid = %mbid,
urls = info.urls.len(),
country = ?info.country,
"fetched artist info"
);
Some(info)
}
Err(e) => {
tracing::warn!(mbid = %mbid, error = %e, "failed to fetch artist info");
None
}
};
// Fetch Wikipedia photo + bio (cached)
let (artist_photo, artist_bio) = fetch_wikipedia_data(state, &mbid, &artist_info).await;
tracing::debug!(mbid = %mbid, has_photo = artist_photo.is_some(), has_bio = artist_bio.is_some(), "wikipedia data");
// Fetch release groups and filter by allowed secondary types
let all_release_groups = state
.search
@@ -568,6 +589,9 @@ pub async fn enrich_artist(
"total_watched_tracks": total_artist_watched,
"total_owned_tracks": total_artist_owned,
"enriched": !skip_track_fetch,
"artist_info": artist_info,
"artist_photo": artist_photo,
"artist_bio": artist_bio,
}))
}
@@ -591,6 +615,140 @@ pub async fn enrich_all_watched_artists(state: &AppState) -> Result<u32, ApiErro
Ok(count)
}
/// Fetch artist photo and bio from Wikipedia, with caching.
async fn fetch_wikipedia_data(
state: &AppState,
mbid: &str,
artist_info: &Option<shanty_tag::provider::ArtistInfo>,
) -> (Option<String>, Option<String>) {
let cache_key = format!("artist_wiki:{mbid}");
// Check cache first
if let Ok(Some(json)) = queries::cache::get(state.db.conn(), &cache_key).await {
if let Ok(cached) = serde_json::from_str::<serde_json::Value>(&json) {
return (
cached.get("photo_url").and_then(|v| v.as_str()).map(String::from),
cached.get("bio").and_then(|v| v.as_str()).map(String::from),
);
}
}
// Find Wikipedia URL from artist info — try direct link first, then resolve via Wikidata
let wiki_url = if let Some(info) = artist_info.as_ref() {
if let Some(u) = info.urls.iter().find(|u| u.link_type == "wikipedia") {
Some(u.url.clone())
} else if let Some(wd) = info.urls.iter().find(|u| u.link_type == "wikidata") {
// Extract Wikidata entity ID and resolve to Wikipedia URL
let entity_id = wd.url.split('/').last().unwrap_or("");
resolve_wikidata_to_wikipedia(entity_id).await
} else {
None
}
} else {
None
};
let Some(wiki_url) = wiki_url else {
tracing::debug!(mbid = mbid, "no wikipedia URL found");
return (None, None);
};
tracing::debug!(mbid = mbid, wiki_url = %wiki_url, "found wikipedia URL");
// Parse article title from URL (e.g., https://en.wikipedia.org/wiki/Pink_Floyd → Pink_Floyd)
let title = wiki_url
.split("/wiki/")
.nth(1)
.unwrap_or("")
.to_string();
if title.is_empty() {
return (None, None);
}
// Detect language from URL (e.g., en.wikipedia.org → en)
let lang = wiki_url
.split("://")
.nth(1)
.and_then(|s| s.split('.').next())
.unwrap_or("en");
// Call Wikipedia REST API
let api_url = format!("https://{lang}.wikipedia.org/api/rest_v1/page/summary/{title}");
let client = reqwest::Client::builder()
.user_agent("Shanty/0.1.0 (shanty-music-app)")
.build()
.ok();
let Some(client) = client else {
return (None, None);
};
let resp = match client.get(&api_url).send().await {
Ok(r) if r.status().is_success() => r,
_ => return (None, None),
};
let body: serde_json::Value = match resp.json().await {
Ok(v) => v,
Err(_) => return (None, None),
};
let photo_url = body
.get("thumbnail")
.and_then(|t| t.get("source"))
.and_then(|s| s.as_str())
.map(String::from);
let bio = body
.get("extract")
.and_then(|e| e.as_str())
.map(String::from);
// Cache for 30 days
let cache_val = serde_json::json!({ "photo_url": photo_url, "bio": bio });
let _ = queries::cache::set(
state.db.conn(),
&cache_key,
"wikipedia",
&cache_val.to_string(),
30 * 86400,
)
.await;
(photo_url, bio)
}
/// Resolve a Wikidata entity ID to an English Wikipedia URL.
async fn resolve_wikidata_to_wikipedia(entity_id: &str) -> Option<String> {
if entity_id.is_empty() {
return None;
}
let url = format!(
"https://www.wikidata.org/w/api.php?action=wbgetentities&ids={entity_id}&props=sitelinks&sitefilter=enwiki&format=json"
);
let client = reqwest::Client::builder()
.user_agent("Shanty/0.1.0 (shanty-music-app)")
.build()
.ok()?;
let resp: serde_json::Value = client.get(&url).send().await.ok()?.json().await.ok()?;
let title = resp
.get("entities")
.and_then(|e| e.get(entity_id))
.and_then(|e| e.get("sitelinks"))
.and_then(|s| s.get("enwiki"))
.and_then(|w| w.get("title"))
.and_then(|t| t.as_str())?;
Some(format!(
"https://en.wikipedia.org/wiki/{}",
title.replace(' ', "_")
))
}
async fn add_artist(
state: web::Data<AppState>,
session: Session,