library: add remote_metadata module structure and unified DTOs

- Add RemoteProvider enum (MusicBrainz, Discogs)
- Add RemoteAlbumSearchResult for lightweight search results
- Add RemoteAlbumMetadata for full metadata with tracks
- Add RemoteTrackMetadata for track info
- Add RemoteMetadataCache with TTL-based expiration
- Add adapter functions for MusicBrainz and Discogs
- Add Discogs position/duration parsing utilities
This commit is contained in:
vicrodh committed 2026-01-26 18:37:45 -06:00
1 parent 50d765509d
commit 70df504af3
4 files changed
+717

No files matched your search

+1
View File
@@ -9,6 +9,7 @@ pub mod database;
pub mod errors;
pub mod metadata;
pub mod models;
pub mod remote_metadata;
pub mod scanner;
pub mod tag_sidecar;
pub mod thumbnails;
@@ -0,0 +1,225 @@
//! In-memory cache for remote metadata requests
//!
//! Provides short-lived caching to avoid redundant API calls during
//! a tag editing session.
use std::collections::HashMap;
use std::sync::Arc;
use std::time::{Duration, Instant};
use tokio::sync::RwLock;
use super::models::{RemoteAlbumMetadata, RemoteAlbumSearchResult, RemoteProvider};
/// Cache entry with expiration
#[derive(Debug, Clone)]
struct CacheEntry<T> {
data: T,
inserted_at: Instant,
ttl: Duration,
}
impl<T> CacheEntry<T> {
fn new(data: T, ttl: Duration) -> Self {
Self {
data,
inserted_at: Instant::now(),
ttl,
}
}
fn is_expired(&self) -> bool {
self.inserted_at.elapsed() > self.ttl
}
}
/// Cache key for search results
#[derive(Debug, Clone, Hash, PartialEq, Eq)]
struct SearchCacheKey {
provider: String,
query: String,
catalog_id: Option<String>,
artist: Option<String>,
}
/// Cache key for full metadata
#[derive(Debug, Clone, Hash, PartialEq, Eq)]
struct MetadataCacheKey {
provider: String,
provider_id: String,
}
/// Remote metadata cache
pub struct RemoteMetadataCache {
/// Search results cache (short TTL: 10 minutes)
search_cache: Arc<RwLock<HashMap<SearchCacheKey, CacheEntry<Vec<RemoteAlbumSearchResult>>>>>,
/// Full metadata cache (longer TTL: 1 hour)
metadata_cache: Arc<RwLock<HashMap<MetadataCacheKey, CacheEntry<RemoteAlbumMetadata>>>>,
/// TTL for search results
search_ttl: Duration,
/// TTL for full metadata
metadata_ttl: Duration,
}
impl Default for RemoteMetadataCache {
fn default() -> Self {
Self::new()
}
}
impl RemoteMetadataCache {
/// Create a new cache with default TTLs
pub fn new() -> Self {
Self {
search_cache: Arc::new(RwLock::new(HashMap::new())),
metadata_cache: Arc::new(RwLock::new(HashMap::new())),
search_ttl: Duration::from_secs(10 * 60), // 10 minutes
metadata_ttl: Duration::from_secs(60 * 60), // 1 hour
}
}
/// Get cached search results
pub async fn get_search(
&self,
provider: RemoteProvider,
query: &str,
catalog_id: Option<&str>,
artist: Option<&str>,
) -> Option<Vec<RemoteAlbumSearchResult>> {
let key = SearchCacheKey {
provider: provider.to_string(),
query: query.to_lowercase(),
catalog_id: catalog_id.map(|s| s.to_lowercase()),
artist: artist.map(|s| s.to_lowercase()),
};
let cache = self.search_cache.read().await;
if let Some(entry) = cache.get(&key) {
if !entry.is_expired() {
log::debug!(
"Cache hit for search: {} {:?}",
provider,
query
);
return Some(entry.data.clone());
}
}
None
}
/// Store search results in cache
pub async fn set_search(
&self,
provider: RemoteProvider,
query: &str,
catalog_id: Option<&str>,
artist: Option<&str>,
results: Vec<RemoteAlbumSearchResult>,
) {
let key = SearchCacheKey {
provider: provider.to_string(),
query: query.to_lowercase(),
catalog_id: catalog_id.map(|s| s.to_lowercase()),
artist: artist.map(|s| s.to_lowercase()),
};
let mut cache = self.search_cache.write().await;
cache.insert(key, CacheEntry::new(results, self.search_ttl));
}
/// Get cached full metadata
pub async fn get_metadata(
&self,
provider: RemoteProvider,
provider_id: &str,
) -> Option<RemoteAlbumMetadata> {
let key = MetadataCacheKey {
provider: provider.to_string(),
provider_id: provider_id.to_string(),
};
let cache = self.metadata_cache.read().await;
if let Some(entry) = cache.get(&key) {
if !entry.is_expired() {
log::debug!(
"Cache hit for metadata: {} {}",
provider,
provider_id
);
return Some(entry.data.clone());
}
}
None
}
/// Store full metadata in cache
pub async fn set_metadata(
&self,
provider: RemoteProvider,
provider_id: &str,
metadata: RemoteAlbumMetadata,
) {
let key = MetadataCacheKey {
provider: provider.to_string(),
provider_id: provider_id.to_string(),
};
let mut cache = self.metadata_cache.write().await;
cache.insert(key, CacheEntry::new(metadata, self.metadata_ttl));
}
/// Clear expired entries from both caches
pub async fn cleanup_expired(&self) -> usize {
let mut removed = 0;
{
let mut cache = self.search_cache.write().await;
let before = cache.len();
cache.retain(|_, entry| !entry.is_expired());
removed += before - cache.len();
}
{
let mut cache = self.metadata_cache.write().await;
let before = cache.len();
cache.retain(|_, entry| !entry.is_expired());
removed += before - cache.len();
}
if removed > 0 {
log::debug!("Cleaned up {} expired cache entries", removed);
}
removed
}
/// Clear all cached data
pub async fn clear_all(&self) {
{
let mut cache = self.search_cache.write().await;
cache.clear();
}
{
let mut cache = self.metadata_cache.write().await;
cache.clear();
}
log::info!("Cleared remote metadata cache");
}
/// Get cache statistics
pub async fn stats(&self) -> CacheStats {
let search_count = self.search_cache.read().await.len();
let metadata_count = self.metadata_cache.read().await.len();
CacheStats {
search_entries: search_count,
metadata_entries: metadata_count,
}
}
}
/// Cache statistics
#[derive(Debug, Clone, serde::Serialize)]
pub struct CacheStats {
pub search_entries: usize,
pub metadata_entries: usize,
}
@@ -0,0 +1,254 @@
//! Remote metadata fetching for the Tag Editor
//!
//! This module provides a unified interface for fetching album metadata
//! from MusicBrainz and Discogs. The Tag Editor uses these services to
//! help users fill in metadata for their local library albums.
//!
//! # Providers
//!
//! - **MusicBrainz** (default): Free, community-maintained database.
//! No authentication required. Uses tags for genres.
//!
//! - **Discogs**: Commercial database with extensive catalog.
//! Requires API credentials (handled via proxy). Has genres and styles.
//!
//! # Usage
//!
//! 1. Search for albums using `search_albums()`
//! 2. Display results to user
//! 3. Fetch full metadata using `get_album_metadata()`
//! 4. Apply to editor form
mod cache;
mod models;
pub use cache::{CacheStats, RemoteMetadataCache};
pub use models::{
RemoteAlbumMetadata, RemoteAlbumSearchResult, RemoteMetadataError, RemoteProvider,
RemoteSearchRequest, RemoteSearchResponse, RemoteTrackMetadata,
};
use std::sync::Arc;
use tokio::sync::Mutex;
use crate::discogs::DiscogsClient;
use crate::musicbrainz::MusicBrainzSharedState;
/// Shared state for remote metadata operations
pub struct RemoteMetadataState {
/// In-memory cache
pub cache: RemoteMetadataCache,
/// MusicBrainz client (shared with existing integration)
pub musicbrainz: Option<Arc<MusicBrainzSharedState>>,
/// Discogs client
pub discogs: Arc<Mutex<DiscogsClient>>,
}
impl RemoteMetadataState {
pub fn new(musicbrainz: Option<Arc<MusicBrainzSharedState>>) -> Self {
Self {
cache: RemoteMetadataCache::new(),
musicbrainz,
discogs: Arc::new(Mutex::new(DiscogsClient::new())),
}
}
}
// ============ MusicBrainz Adapter ============
/// Convert MusicBrainz release search result to unified DTO
pub fn musicbrainz_release_to_search_result(
release: &crate::musicbrainz::models::ReleaseResult,
) -> RemoteAlbumSearchResult {
// Extract artist from artist-credit
let artist = release
.artist_credit
.as_ref()
.map(|credits| {
credits
.iter()
.map(|c| {
format!(
"{}{}",
c.name.as_deref().unwrap_or(&c.artist.name),
c.joinphrase.as_deref().unwrap_or("")
)
})
.collect::<Vec<_>>()
.join("")
})
.unwrap_or_default();
// Extract year from date (YYYY or YYYY-MM-DD)
let year = release.date.as_ref().and_then(|d| {
d.split('-').next().and_then(|y| y.parse::<u16>().ok())
});
// Extract label and catalog number
let (label, catalog_number) = release
.label_info
.as_ref()
.and_then(|info| info.first())
.map(|li| {
(
li.label.as_ref().map(|l| l.name.clone()),
li.catalog_number.clone(),
)
})
.unwrap_or((None, None));
RemoteAlbumSearchResult {
provider: RemoteProvider::MusicBrainz,
provider_id: release.id.clone(),
title: release.title.clone(),
artist,
year,
track_count: None, // Not available in search results
country: release.country.clone(),
label,
catalog_number,
confidence: release.score.map(|s| s.min(100) as u8),
format: None,
}
}
// ============ Discogs Adapter ============
/// Convert Discogs release search result to unified DTO
pub fn discogs_release_to_search_result(
result: &crate::discogs::SearchResult,
) -> RemoteAlbumSearchResult {
// Discogs title format is usually "Artist - Album"
let (artist, title) = if let Some(pos) = result.title.find(" - ") {
let (a, t) = result.title.split_at(pos);
(a.to_string(), t.trim_start_matches(" - ").to_string())
} else {
("Unknown Artist".to_string(), result.title.clone())
};
RemoteAlbumSearchResult {
provider: RemoteProvider::Discogs,
provider_id: result.id.to_string(),
title,
artist,
year: None, // Not in search result, need to fetch details
track_count: None,
country: None,
label: None,
catalog_number: None,
confidence: None,
format: None,
}
}
/// Parse Discogs track position to (disc_number, track_number)
/// Handles formats: "1", "A1", "1-1", "CD1-1", "1.1"
pub fn parse_discogs_position(position: &str) -> (u8, u8) {
let position = position.trim();
// Handle empty position
if position.is_empty() {
return (1, 1);
}
// Try "X-Y" format (e.g., "1-5", "CD1-3")
if let Some(pos) = position.find('-') {
let disc_part = &position[..pos];
let track_part = &position[pos + 1..];
// Extract number from disc part (handle "CD1", "1", etc.)
let disc = disc_part
.chars()
.filter(|c| c.is_ascii_digit())
.collect::<String>()
.parse::<u8>()
.unwrap_or(1);
let track = track_part.parse::<u8>().unwrap_or(1);
return (disc, track);
}
// Try "X.Y" format
if let Some(pos) = position.find('.') {
let disc_part = &position[..pos];
let track_part = &position[pos + 1..];
let disc = disc_part.parse::<u8>().unwrap_or(1);
let track = track_part.parse::<u8>().unwrap_or(1);
return (disc, track);
}
// Handle vinyl sides (A, B, C, D -> disc 1, 1, 2, 2)
if position.starts_with(|c: char| c.is_ascii_alphabetic()) {
let side = position.chars().next().unwrap().to_ascii_uppercase();
let track_str: String = position.chars().skip(1).collect();
let track = track_str.parse::<u8>().unwrap_or(1);
let disc = match side {
'A' | 'B' => 1,
'C' | 'D' => 2,
'E' | 'F' => 3,
_ => 1,
};
return (disc, track);
}
// Simple number
let track = position.parse::<u8>().unwrap_or(1);
(1, track)
}
/// Parse Discogs duration string to milliseconds
/// Handles format: "M:SS" or "MM:SS" or "H:MM:SS"
pub fn parse_discogs_duration(duration: &str) -> Option<u32> {
let parts: Vec<&str> = duration.split(':').collect();
match parts.len() {
2 => {
// M:SS or MM:SS
let minutes: u32 = parts[0].parse().ok()?;
let seconds: u32 = parts[1].parse().ok()?;
Some((minutes * 60 + seconds) * 1000)
}
3 => {
// H:MM:SS
let hours: u32 = parts[0].parse().ok()?;
let minutes: u32 = parts[1].parse().ok()?;
let seconds: u32 = parts[2].parse().ok()?;
Some((hours * 3600 + minutes * 60 + seconds) * 1000)
}
_ => None,
}
}
#[cfg(test)]
mod tests {
use super::*;
#[test]
fn test_parse_discogs_position() {
assert_eq!(parse_discogs_position("1"), (1, 1));
assert_eq!(parse_discogs_position("5"), (1, 5));
assert_eq!(parse_discogs_position("1-1"), (1, 1));
assert_eq!(parse_discogs_position("1-5"), (1, 5));
assert_eq!(parse_discogs_position("2-3"), (2, 3));
assert_eq!(parse_discogs_position("CD1-5"), (1, 5));
assert_eq!(parse_discogs_position("CD2-3"), (2, 3));
assert_eq!(parse_discogs_position("A1"), (1, 1));
assert_eq!(parse_discogs_position("B2"), (1, 2));
assert_eq!(parse_discogs_position("C1"), (2, 1));
assert_eq!(parse_discogs_position("D3"), (2, 3));
assert_eq!(parse_discogs_position("1.5"), (1, 5));
assert_eq!(parse_discogs_position("2.3"), (2, 3));
}
#[test]
fn test_parse_discogs_duration() {
assert_eq!(parse_discogs_duration("3:45"), Some(225000));
assert_eq!(parse_discogs_duration("0:30"), Some(30000));
assert_eq!(parse_discogs_duration("10:00"), Some(600000));
assert_eq!(parse_discogs_duration("1:00:00"), Some(3600000));
assert_eq!(parse_discogs_duration("invalid"), None);
}
}
@@ -0,0 +1,237 @@
//! Unified DTOs for remote metadata providers (MusicBrainz, Discogs)
//!
//! These structs provide a provider-neutral interface for the Tag Editor
//! to consume album metadata from different sources.
use serde::{Deserialize, Serialize};
/// Remote metadata provider
#[derive(Debug, Clone, Copy, PartialEq, Eq, Serialize, Deserialize)]
#[serde(rename_all = "lowercase")]
pub enum RemoteProvider {
MusicBrainz,
Discogs,
}
impl std::fmt::Display for RemoteProvider {
fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
match self {
Self::MusicBrainz => write!(f, "musicbrainz"),
Self::Discogs => write!(f, "discogs"),
}
}
}
impl std::str::FromStr for RemoteProvider {
type Err = String;
fn from_str(s: &str) -> Result<Self, Self::Err> {
match s.to_lowercase().as_str() {
"musicbrainz" | "mb" => Ok(Self::MusicBrainz),
"discogs" => Ok(Self::Discogs),
_ => Err(format!("Unknown provider: {}", s)),
}
}
}
/// Lightweight search result for displaying in results list
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct RemoteAlbumSearchResult {
/// Provider identifier
pub provider: RemoteProvider,
/// Provider-specific ID (MusicBrainz release ID or Discogs release ID)
pub provider_id: String,
/// Album title
pub title: String,
/// Album artist
pub artist: String,
/// Release year (extracted from date)
pub year: Option<u16>,
/// Number of tracks (if available from search)
pub track_count: Option<u16>,
/// Release country
pub country: Option<String>,
/// Record label
pub label: Option<String>,
/// Catalog number
pub catalog_number: Option<String>,
/// Match confidence (0-100, provider-specific)
pub confidence: Option<u8>,
/// Format info (e.g., "CD", "Vinyl", "Digital")
pub format: Option<String>,
}
/// Full album metadata with tracks (for applying to form)
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct RemoteAlbumMetadata {
/// Provider identifier
pub provider: RemoteProvider,
/// Provider-specific ID
pub provider_id: String,
/// Album title
pub title: String,
/// Album artist
pub artist: String,
/// Release year
pub year: Option<u16>,
/// Genres/styles/tags
pub genres: Vec<String>,
/// Record label
pub label: Option<String>,
/// Catalog number
pub catalog_number: Option<String>,
/// Release country
pub country: Option<String>,
/// Barcode/UPC (if available)
pub barcode: Option<String>,
/// Track list organized by disc
pub tracks: Vec<RemoteTrackMetadata>,
/// Total disc count
pub disc_count: u8,
/// URL to view on provider website
pub source_url: Option<String>,
}
/// Single track metadata
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct RemoteTrackMetadata {
/// Disc number (1-based)
pub disc_number: u8,
/// Track number within disc (1-based)
pub track_number: u8,
/// Track title
pub title: String,
/// Duration in milliseconds (if available)
pub duration_ms: Option<u32>,
}
/// Search request parameters
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct RemoteSearchRequest {
/// Provider to search
pub provider: RemoteProvider,
/// Search query (usually "artist album")
pub query: String,
/// Optional catalog number for more precise matching
pub catalog_id: Option<String>,
/// Optional artist name for filtering
pub artist: Option<String>,
/// Maximum results to return (default: 10)
pub limit: Option<usize>,
}
impl RemoteSearchRequest {
pub fn limit(&self) -> usize {
self.limit.unwrap_or(10).min(25)
}
}
/// Search response wrapper
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct RemoteSearchResponse {
/// Provider that was searched
pub provider: RemoteProvider,
/// Search results
pub results: Vec<RemoteAlbumSearchResult>,
/// Total results available (may be more than returned)
pub total_count: Option<usize>,
/// Whether rate limit was hit
pub rate_limited: bool,
}
impl RemoteSearchResponse {
pub fn empty(provider: RemoteProvider) -> Self {
Self {
provider,
results: Vec::new(),
total_count: Some(0),
rate_limited: false,
}
}
pub fn rate_limited(provider: RemoteProvider) -> Self {
Self {
provider,
results: Vec::new(),
total_count: None,
rate_limited: true,
}
}
}
/// Error types for remote metadata operations
#[derive(Debug, Clone, Serialize, Deserialize)]
#[serde(tag = "type", content = "message")]
pub enum RemoteMetadataError {
/// Network or connection error
NetworkError(String),
/// Rate limit exceeded
RateLimited(String),
/// Invalid response from provider
InvalidResponse(String),
/// No results found
NoResults,
/// Provider not available
ProviderUnavailable(String),
/// Invalid provider ID
InvalidProviderId(String),
}
impl std::fmt::Display for RemoteMetadataError {
fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
match self {
Self::NetworkError(msg) => write!(f, "Network error: {}", msg),
Self::RateLimited(msg) => write!(f, "Rate limited: {}", msg),
Self::InvalidResponse(msg) => write!(f, "Invalid response: {}", msg),
Self::NoResults => write!(f, "No results found"),
Self::ProviderUnavailable(msg) => write!(f, "Provider unavailable: {}", msg),
Self::InvalidProviderId(msg) => write!(f, "Invalid provider ID: {}", msg),
}
}
}
impl From<RemoteMetadataError> for String {
fn from(err: RemoteMetadataError) -> Self {
err.to_string()
}
}
#[cfg(test)]
mod tests {
use super::*;
#[test]
fn test_provider_parsing() {
assert_eq!(
"musicbrainz".parse::<RemoteProvider>().unwrap(),
RemoteProvider::MusicBrainz
);
assert_eq!(
"discogs".parse::<RemoteProvider>().unwrap(),
RemoteProvider::Discogs
);
assert_eq!(
"MB".parse::<RemoteProvider>().unwrap(),
RemoteProvider::MusicBrainz
);
assert!("unknown".parse::<RemoteProvider>().is_err());
}
#[test]
fn test_search_request_limit() {
let req = RemoteSearchRequest {
provider: RemoteProvider::MusicBrainz,
query: "test".to_string(),
catalog_id: None,
artist: None,
limit: None,
};
assert_eq!(req.limit(), 10);
let req_custom = RemoteSearchRequest {
limit: Some(50),
..req.clone()
};
assert_eq!(req_custom.limit(), 25); // Capped at 25
}
}