please work please work ci please

.dockerignore

···
+# Rust build artifacts
+target/
+**/target/
+services/target/
+apps/*/target/
+
+# Node.js dependencies and build artifacts
+node_modules/
+**/node_modules/
+.turbo/
+**/.turbo/
+build/
+dist/
+.next/
+
+# Development and cache files
+.git/
+.gitignore
+**/.DS_Store
+*.log
+*.tmp
+*.temp
+
+# IDE and editor files
+.vscode/
+.idea/
+*.swp
+*.swo
+*~
+
+# Environment and config files
+.env
+.env.local
+.env.*.local
+
+# Database files
+*.db
+*.sqlite
+*.sqlite3
+
+# Test coverage
+coverage/
+**/coverage/
+
+# Temporary files
+tmp/
+temp/
+
+# SQLx offline query cache
+# Include workspace-level cache for monorepo builds
+# Uncomment the line below if you want to force online compilation
+# .sqlx/

.github/workflows/amethyst.yml

···
         run: pnpm lex:gen-server
 
       - name: Build web
-        run: |
-          cd apps/amethyst
-          pnpm build:web
+        run: pnpm turbo build:web --filter=@teal/amethyst
 
       - name: Upload web build artifacts
         uses: actions/upload-artifact@v4
···
         run: npm install -g @expo/cli
 
       - name: Build iOS
-        run: |
-          cd apps/amethyst
-          pnpm build:ios
+        run: pnpm turbo build:ios --filter=@teal/amethyst
 
       - name: Upload iOS build artifacts
         uses: actions/upload-artifact@v4
···
         run: pnpm lex:gen-server
 
       - name: Type check
-        run: |
-          cd apps/amethyst
-          npx tsc --noEmit
+        run: pnpm turbo check-types --filter=@teal/amethyst
 
       - name: Run tests
-        run: |
-          cd apps/amethyst
-          pnpm test --watchAll=false
+        run: pnpm turbo test --filter=@teal/amethyst

.github/workflows/aqua.yml

···
       - name: Build and push Docker image
         uses: docker/build-push-action@v5
         with:
-          context: ./apps/aqua
+          context: .
           file: ./apps/aqua/Dockerfile
           push: ${{ github.event_name != 'pull_request' }}
           tags: ${{ steps.meta.outputs.tags }}

.gitignore

···
 vendor/**/*.d.ts
 vendor/**/dist/
 vendor/**/node_modules/
-
-# lexicons directory structure
-!lexicons/
-# Track our custom lexicons
-!lexicons/fm.teal.alpha/
-!lexicons/fm.teal.alpha/**/*.json
-# Ignore symlinks to atproto lexicons (created during setup)
-lexicons/app
-lexicons/chat
-lexicons/com
-lexicons/tools
-# But ignore any generated files within lexicons
-lexicons/**/*.js
-lexicons/**/*.d.ts
-lexicons/**/dist/
-lexicons/**/node_modules/

Cargo.lock

···
  "chrono",
  "clap",
  "dotenvy",
- "iroh-car 0.4.0",
+ "iroh-car",
  "redis",
  "reqwest",
  "serde",
···
  "dashmap",
  "futures",
  "ipld-core",
- "iroh-car 0.5.1",
+ "iroh-car",
  "log",
  "multihash 0.19.3",
  "serde",
···
 ]
 
 [[package]]
+name = "backon"
+version = "1.5.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "592277618714fbcecda9a02ba7a8781f319d26532a88553bbacc77ba5d2b3a8d"
+dependencies = [
+ "fastrand",
+]
+
+[[package]]
 name = "backtrace"
 version = "0.3.75"
 source = "registry+https://github.com/rust-lang/crates.io-index"
···
  "dotenvy",
  "flume",
  "futures",
- "iroh-car 0.4.0",
+ "iroh-car",
  "libipld",
  "metrics 0.23.1",
  "metrics-exporter-prometheus",
···
 
 [[package]]
 name = "iroh-car"
-version = "0.4.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "475a6f0ebd64c87ea011021c67f10b57930f6c286e0163807066bfb83553b1b6"
-dependencies = [
- "anyhow",
- "cid 0.10.1",
- "futures",
- "libipld",
- "thiserror 1.0.69",
- "tokio",
- "unsigned-varint 0.7.2",
-]
-
-[[package]]
-name = "iroh-car"
 version = "0.5.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "cb7f8cd4cb9aa083fba8b52e921764252d0b4dcb1cd6d120b809dbfe1106e81a"
···
 ]
 
 [[package]]
+name = "num-bigint"
+version = "0.4.6"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "a5e44f723f1133c9deac646763579fdb3ac745e418f2a7af9cd0c431da1f20b9"
+dependencies = [
+ "num-integer",
+ "num-traits",
+]
+
+[[package]]
 name = "num-bigint-dig"
 version = "0.8.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
···
 checksum = "e3148f5046208a5d56bcfc03053e3ca6334e51da8dfb19b6cdc8b306fae3283e"
 
 [[package]]
-name = "pin-project"
-version = "1.1.10"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "677f1add503faace112b9f1373e43e9e054bfdd22ff1a63c1bc485eaec6a6a8a"
-dependencies = [
- "pin-project-internal",
-]
-
-[[package]]
-name = "pin-project-internal"
-version = "1.1.10"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "6e918e4ff8c4549eb882f14b3a4bc8c8bc93de829416eacf579f1207a8fbf861"
-dependencies = [
- "proc-macro2",
- "quote",
- "syn 2.0.104",
-]
-
-[[package]]
 name = "pin-project-lite"
 version = "0.2.16"
 source = "registry+https://github.com/rust-lang/crates.io-index"
···
  "once_cell",
  "socket2 0.5.10",
  "tracing",
- "windows-sys 0.52.0",
+ "windows-sys 0.59.0",
 ]
 
 [[package]]
···
 
 [[package]]
 name = "redis"
-version = "0.24.0"
+version = "0.32.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "c580d9cbbe1d1b479e8d67cf9daf6a62c957e6846048408b80b43ac3f6af84cd"
+checksum = "e1f66bf4cac9733a23bcdf1e0e01effbaaad208567beba68be8f67e5f4af3ee1"
 dependencies = [
  "arc-swap",
- "async-trait",
+ "backon",
  "bytes",
+ "cfg-if",
  "combine",
- "futures",
+ "futures-channel",
  "futures-util",
  "itoa",
+ "num-bigint",
  "percent-encoding",
  "pin-project-lite",
  "ryu",
  "sha1_smol",
- "socket2 0.4.10",
+ "socket2 0.6.0",
  "tokio",
- "tokio-retry",
  "tokio-util",
  "url",
 ]
···
 
 [[package]]
 name = "socket2"
-version = "0.4.10"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "9f7916fc008ca5542385b89a3d3ce689953c143e9304a9bf8beec1de48994c0d"
-dependencies = [
- "libc",
- "winapi",
-]
-
-[[package]]
-name = "socket2"
 version = "0.5.10"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "e22376abed350d73dd1cd119b57ffccad95b4e585a7cda43e286245ce23c0678"
···
 checksum = "bbae76ab933c85776efabc971569dd6119c580d8f5d448769dec1764bf796ef2"
 dependencies = [
  "native-tls",
- "tokio",
-]
-
-[[package]]
-name = "tokio-retry"
-version = "0.3.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "7f57eb36ecbe0fc510036adff84824dd3c24bb781e21bfa67b69d556aa85214f"
-dependencies = [
- "pin-project",
- "rand 0.8.5",
  "tokio",
 ]
 

Cargo.toml

···
 rocketman = { path = "services/rocketman" }
 
 # CAR and IPLD dependencies
-iroh-car = "0.4"
+iroh-car = "0.5"
 libipld = { version = "0.16", features = ["dag-cbor", "dag-json"] }
 cid = "0.11"
 base64 = "0.22"
 atmst = "0.0.1"
 
 # Redis for job queues and caching
-redis = { version = "0.24", features = ["tokio-comp", "connection-manager"] }
+redis = { version = "0.32", features = ["tokio-comp", "connection-manager"] }

apps/aqua/Dockerfile

···
 # Set up cross-compilation environment
 ENV CARGO_TARGET_AARCH64_UNKNOWN_LINUX_GNU_LINKER=aarch64-linux-gnu-gcc
 
+
 # Debug platform detection and run build
-RUN echo "DEBUG Before target.sh: TARGETPLATFORM=$TARGETPLATFORM TARGETARCH=$TARGETARCH" && \
-    . ./target.sh && \
-    touch src/main.rs && \
+RUN . ./target.sh && \
+    touch apps/aqua/src/main.rs && \
     echo "Building for $TARGET_ARCH" && \
-    cargo build --release --target $RUST_TARGET && \
+    cargo build --release --target $RUST_TARGET --package aqua && \
     cp target/$RUST_TARGET/release/aqua target/aqua
 
 FROM --platform=${TARGETARCH:-$BUILDPLATFORM} gcr.io/distroless/cc

lexicons/app

···
+../vendor/atproto/lexicons/app

lexicons/chat

···
+../vendor/atproto/lexicons/chat

lexicons/com

···
+../vendor/atproto/lexicons/com

lexicons/tools

···
+../vendor/atproto/lexicons/tools

services/cadet/Dockerfile

···
 # Set up cross-compilation environment
 ENV CARGO_TARGET_AARCH64_UNKNOWN_LINUX_GNU_LINKER=aarch64-linux-gnu-gcc
 
+# Force SQLx to use offline mode with workspace cache
+ENV SQLX_OFFLINE=true
+
+# copy sqlx in
+COPY .sqlx ./services/cadet/.sqlx
+
 # Debug platform detection and run build
 RUN echo "DEBUG Before target.sh: TARGETPLATFORM=$TARGETPLATFORM TARGETARCH=$TARGETARCH" && \
     . ./target.sh && \
-    touch src/main.rs && \
+    touch services/cadet/src/main.rs && \
     echo "Building for $TARGET_ARCH" && \
-    cargo build --release --target $RUST_TARGET && \
+    cargo build --release --target $RUST_TARGET --package cadet && \
     cp target/$RUST_TARGET/release/cadet target/cadet
 
 FROM --platform=${TARGETARCH:-$BUILDPLATFORM} gcr.io/distroless/cc

services/migrations/20241220000001_initial_schema.sql

···
--- Initial comprehensive schema for Teal music platform
--- Based on services/cadet/sql/base.sql
-
-CREATE TABLE artists (
-    mbid UUID PRIMARY KEY,
-    name TEXT NOT NULL,
-    play_count INTEGER DEFAULT 0
-);
-
--- releases are synologous to 'albums'
-CREATE TABLE releases (
-    mbid UUID PRIMARY KEY,
-    name TEXT NOT NULL,
-    play_count INTEGER DEFAULT 0
-);
-
--- recordings are synologous to 'tracks' BUT tracks can be in multiple releases!
-CREATE TABLE recordings (
-    mbid UUID PRIMARY KEY,
-    name TEXT NOT NULL,
-    play_count INTEGER DEFAULT 0
-);
-
-CREATE TABLE plays (
-    uri TEXT PRIMARY KEY,
-    did TEXT NOT NULL,
-    rkey TEXT NOT NULL,
-    cid TEXT NOT NULL,
-    isrc TEXT,
-    duration INTEGER,
-    track_name TEXT NOT NULL,
-    played_time TIMESTAMP WITH TIME ZONE,
-    processed_time TIMESTAMP WITH TIME ZONE DEFAULT NOW(),
-    release_mbid UUID,
-    release_name TEXT,
-    recording_mbid UUID,
-    submission_client_agent TEXT,
-    music_service_base_domain TEXT,
-    origin_url TEXT,
-    FOREIGN KEY (release_mbid) REFERENCES releases (mbid),
-    FOREIGN KEY (recording_mbid) REFERENCES recordings (mbid)
-);
-
-CREATE INDEX idx_plays_release_mbid ON plays (release_mbid);
-CREATE INDEX idx_plays_recording_mbid ON plays (recording_mbid);
-CREATE INDEX idx_plays_played_time ON plays (played_time);
-CREATE INDEX idx_plays_did ON plays (did);
-
-CREATE TABLE play_to_artists (
-    play_uri TEXT, -- references plays(uri)
-    artist_mbid UUID REFERENCES artists (mbid),
-    artist_name TEXT, -- storing here for ease of use when joining
-    PRIMARY KEY (play_uri, artist_mbid),
-    FOREIGN KEY (play_uri) REFERENCES plays (uri)
-);
-
-CREATE INDEX idx_play_to_artists_artist ON play_to_artists (artist_mbid);
-
--- Profiles table
-CREATE TABLE profiles (
-    did TEXT PRIMARY KEY,
-    handle TEXT,
-    display_name TEXT,
-    description TEXT,
-    description_facets JSONB,
-    avatar TEXT, -- IPLD of the image, bafy...
-    banner TEXT,
-    created_at TIMESTAMP WITH TIME ZONE
-);
-
--- User featured items table
-CREATE TABLE featured_items (
-    did TEXT PRIMARY KEY,
-    mbid TEXT NOT NULL,
-    type TEXT NOT NULL
-);
-
--- Statii table (status records)
-CREATE TABLE statii (
-    uri TEXT PRIMARY KEY,
-    did TEXT NOT NULL,
-    rkey TEXT NOT NULL,
-    cid TEXT NOT NULL,
-    record JSONB NOT NULL,
-    indexed_at TIMESTAMP WITH TIME ZONE DEFAULT NOW()
-);
-
-CREATE INDEX idx_statii_did_rkey ON statii (did, rkey);
-
--- Materialized view for artists' play counts
-CREATE MATERIALIZED VIEW mv_artist_play_counts AS
-SELECT
-    a.mbid AS artist_mbid,
-    a.name AS artist_name,
-    COUNT(p.uri) AS play_count
-FROM
-    artists a
-    LEFT JOIN play_to_artists pta ON a.mbid = pta.artist_mbid
-    LEFT JOIN plays p ON p.uri = pta.play_uri
-GROUP BY
-    a.mbid,
-    a.name;
-
-CREATE UNIQUE INDEX idx_mv_artist_play_counts ON mv_artist_play_counts (artist_mbid);
-
--- Materialized view for releases' play counts
-CREATE MATERIALIZED VIEW mv_release_play_counts AS
-SELECT
-    r.mbid AS release_mbid,
-    r.name AS release_name,
-    COUNT(p.uri) AS play_count
-FROM
-    releases r
-    LEFT JOIN plays p ON p.release_mbid = r.mbid
-GROUP BY
-    r.mbid,
-    r.name;
-
-CREATE UNIQUE INDEX idx_mv_release_play_counts ON mv_release_play_counts (release_mbid);
-
--- Materialized view for recordings' play counts
-CREATE MATERIALIZED VIEW mv_recording_play_counts AS
-SELECT
-    rec.mbid AS recording_mbid,
-    rec.name AS recording_name,
-    COUNT(p.uri) AS play_count
-FROM
-    recordings rec
-    LEFT JOIN plays p ON p.recording_mbid = rec.mbid
-GROUP BY
-    rec.mbid,
-    rec.name;
-
-CREATE UNIQUE INDEX idx_mv_recording_play_counts ON mv_recording_play_counts (recording_mbid);
-
--- Global play count materialized view
-CREATE MATERIALIZED VIEW mv_global_play_count AS
-SELECT
-    COUNT(uri) AS total_plays,
-    COUNT(DISTINCT did) AS unique_listeners
-FROM plays;
-
-CREATE UNIQUE INDEX idx_mv_global_play_count ON mv_global_play_count(total_plays);
-
--- Top artists in the last 30 days
-CREATE MATERIALIZED VIEW mv_top_artists_30days AS
-SELECT
-    a.mbid AS artist_mbid,
-    a.name AS artist_name,
-    COUNT(p.uri) AS play_count
-FROM artists a
-INNER JOIN play_to_artists pta ON a.mbid = pta.artist_mbid
-INNER JOIN plays p ON p.uri = pta.play_uri
-WHERE p.played_time >= NOW() - INTERVAL '30 days'
-GROUP BY a.mbid, a.name
-ORDER BY COUNT(p.uri) DESC;
-
--- Top releases in the last 30 days
-CREATE MATERIALIZED VIEW mv_top_releases_30days AS
-SELECT
-    r.mbid AS release_mbid,
-    r.name AS release_name,
-    COUNT(p.uri) AS play_count
-FROM releases r
-INNER JOIN plays p ON p.release_mbid = r.mbid
-WHERE p.played_time >= NOW() - INTERVAL '30 days'
-GROUP BY r.mbid, r.name
-ORDER BY COUNT(p.uri) DESC;
-
--- Top artists for user in the last 30 days
-CREATE MATERIALIZED VIEW mv_top_artists_for_user_30days AS
-SELECT
-    prof.did,
-    a.mbid AS artist_mbid,
-    a.name AS artist_name,
-    COUNT(p.uri) AS play_count
-FROM artists a
-INNER JOIN play_to_artists pta ON a.mbid = pta.artist_mbid
-INNER JOIN plays p ON p.uri = pta.play_uri
-INNER JOIN profiles prof ON prof.did = p.did
-WHERE p.played_time >= NOW() - INTERVAL '30 days'
-GROUP BY prof.did, a.mbid, a.name
-ORDER BY COUNT(p.uri) DESC;
-
--- Top artists for user in the last 7 days
-CREATE MATERIALIZED VIEW mv_top_artists_for_user_7days AS
-SELECT
-    prof.did,
-    a.mbid AS artist_mbid,
-    a.name AS artist_name,
-    COUNT(p.uri) AS play_count
-FROM artists a
-INNER JOIN play_to_artists pta ON a.mbid = pta.artist_mbid
-INNER JOIN plays p ON p.uri = pta.play_uri
-INNER JOIN profiles prof ON prof.did = p.did
-WHERE p.played_time >= NOW() - INTERVAL '7 days'
-GROUP BY prof.did, a.mbid, a.name
-ORDER BY COUNT(p.uri) DESC;
-
--- Top releases for user in the last 30 days
-CREATE MATERIALIZED VIEW mv_top_releases_for_user_30days AS
-SELECT
-    prof.did,
-    r.mbid AS release_mbid,
-    r.name AS release_name,
-    COUNT(p.uri) AS play_count
-FROM releases r
-INNER JOIN plays p ON p.release_mbid = r.mbid
-INNER JOIN profiles prof ON prof.did = p.did
-WHERE p.played_time >= NOW() - INTERVAL '30 days'
-GROUP BY prof.did, r.mbid, r.name
-ORDER BY COUNT(p.uri) DESC;
-
--- Top releases for user in the last 7 days
-CREATE MATERIALIZED VIEW mv_top_releases_for_user_7days AS
-SELECT
-    prof.did,
-    r.mbid AS release_mbid,
-    r.name AS release_name,
-    COUNT(p.uri) AS play_count
-FROM releases r
-INNER JOIN plays p ON p.release_mbid = r.mbid
-INNER JOIN profiles prof ON prof.did = p.did
-WHERE p.played_time >= NOW() - INTERVAL '7 days'
-GROUP BY prof.did, r.mbid, r.name
-ORDER BY COUNT(p.uri) DESC;

services/migrations/20241220000002_car_import_tables.sql

···
--- CAR import functionality tables
--- For handling AT Protocol CAR file imports and processing
-
--- Tracks uploaded CAR files that are queued for processing
-CREATE TABLE IF NOT EXISTS car_import_requests (
-    import_id TEXT PRIMARY KEY,
-    car_data_base64 TEXT NOT NULL,
-    status TEXT NOT NULL DEFAULT 'pending', -- pending, processing, completed, failed
-    created_at TIMESTAMP WITH TIME ZONE DEFAULT NOW(),
-    processed_at TIMESTAMP WITH TIME ZONE,
-    error_message TEXT,
-    file_size_bytes INTEGER,
-    block_count INTEGER,
-    extracted_records_count INTEGER DEFAULT 0
-);
-
-CREATE INDEX idx_car_import_requests_status ON car_import_requests (status);
-CREATE INDEX idx_car_import_requests_created_at ON car_import_requests (created_at);
-
--- Tracks raw IPLD blocks extracted from CAR files
-CREATE TABLE IF NOT EXISTS car_blocks (
-    cid TEXT PRIMARY KEY,
-    import_id TEXT NOT NULL REFERENCES car_import_requests(import_id),
-    block_data BYTEA NOT NULL,
-    decoded_successfully BOOLEAN DEFAULT FALSE,
-    collection_type TEXT, -- e.g., 'fm.teal.alpha.feed.play', 'commit', etc.
-    created_at TIMESTAMP WITH TIME ZONE DEFAULT NOW()
-);
-
-CREATE INDEX idx_car_blocks_import_id ON car_blocks (import_id);
-CREATE INDEX idx_car_blocks_collection_type ON car_blocks (collection_type);
-
--- Tracks records extracted from CAR imports that were successfully processed
-CREATE TABLE IF NOT EXISTS car_extracted_records (
-    id SERIAL PRIMARY KEY,
-    import_id TEXT NOT NULL REFERENCES car_import_requests(import_id),
-    cid TEXT NOT NULL REFERENCES car_blocks(cid),
-    collection_type TEXT NOT NULL,
-    record_uri TEXT, -- AT URI if applicable (e.g., for play records)
-    synthetic_did TEXT, -- DID assigned for CAR imports (e.g., 'car-import:123')
-    rkey TEXT,
-    extracted_at TIMESTAMP WITH TIME ZONE DEFAULT NOW(),
-    processing_notes TEXT
-);
-
-CREATE INDEX idx_car_extracted_records_import_id ON car_extracted_records (import_id);
-CREATE INDEX idx_car_extracted_records_collection_type ON car_extracted_records (collection_type);
-CREATE INDEX idx_car_extracted_records_record_uri ON car_extracted_records (record_uri);
-
--- Tracks import metadata and commit information
-CREATE TABLE IF NOT EXISTS car_import_metadata (
-    import_id TEXT NOT NULL REFERENCES car_import_requests(import_id),
-    metadata_key TEXT NOT NULL,
-    metadata_value JSONB NOT NULL,
-    created_at TIMESTAMP WITH TIME ZONE DEFAULT NOW(),
-    PRIMARY KEY (import_id, metadata_key)
-);
-
-CREATE INDEX idx_car_import_metadata_key ON car_import_metadata (metadata_key);

services/migrations/20241220000003_artists_without_mbids.sql

···
--- Migration to support artists without MusicBrainz IDs
--- This allows the system to comply with the Teal lexicon where only trackName is required
-
--- Add a field to plays table to store raw artist names for records without MBIDs
-ALTER TABLE plays ADD COLUMN artist_names_raw JSONB;
-
--- Create a new artists table that doesn't require MBID as primary key
-CREATE TABLE artists_extended (
-    id SERIAL PRIMARY KEY,
-    mbid UUID UNIQUE, -- Optional MusicBrainz ID
-    name TEXT NOT NULL,
-    name_normalized TEXT GENERATED ALWAYS AS (LOWER(TRIM(name))) STORED,
-    play_count INTEGER DEFAULT 0,
-    created_at TIMESTAMP WITH TIME ZONE DEFAULT NOW(),
-    updated_at TIMESTAMP WITH TIME ZONE DEFAULT NOW()
-);
-
--- Create index for efficient lookups
-CREATE INDEX idx_artists_extended_mbid ON artists_extended (mbid) WHERE mbid IS NOT NULL;
-CREATE INDEX idx_artists_extended_name_normalized ON artists_extended (name_normalized);
-CREATE UNIQUE INDEX idx_artists_extended_name_unique ON artists_extended (name_normalized) WHERE mbid IS NULL;
-
--- Create a new junction table that can handle both MBID and non-MBID artists
-CREATE TABLE play_to_artists_extended (
-    play_uri TEXT NOT NULL REFERENCES plays(uri),
-    artist_id INTEGER NOT NULL REFERENCES artists_extended(id),
-    artist_name TEXT NOT NULL, -- Denormalized for performance
-    PRIMARY KEY (play_uri, artist_id)
-);
-
-CREATE INDEX idx_play_to_artists_extended_artist ON play_to_artists_extended (artist_id);
-
--- Migrate existing data from old tables to new structure
-INSERT INTO artists_extended (mbid, name, play_count)
-SELECT mbid, name, play_count FROM artists;
-
-INSERT INTO play_to_artists_extended (play_uri, artist_id, artist_name)
-SELECT
-    pta.play_uri,
-    ae.id,
-    pta.artist_name
-FROM play_to_artists pta
-JOIN artists_extended ae ON ae.mbid = pta.artist_mbid;
-
--- Update materialized views to use new structure
-DROP MATERIALIZED VIEW IF EXISTS mv_artist_play_counts;
-CREATE MATERIALIZED VIEW mv_artist_play_counts AS
-SELECT
-    ae.id AS artist_id,
-    ae.mbid AS artist_mbid,
-    ae.name AS artist_name,
-    COUNT(p.uri) AS play_count
-FROM
-    artists_extended ae
-    LEFT JOIN play_to_artists_extended ptae ON ae.id = ptae.artist_id
-    LEFT JOIN plays p ON p.uri = ptae.play_uri
-GROUP BY
-    ae.id, ae.mbid, ae.name;
-
-CREATE UNIQUE INDEX idx_mv_artist_play_counts_new ON mv_artist_play_counts (artist_id);
-
--- Update other materialized views that reference artists
-DROP MATERIALIZED VIEW IF EXISTS mv_top_artists_30days;
-CREATE MATERIALIZED VIEW mv_top_artists_30days AS
-SELECT
-    ae.id AS artist_id,
-    ae.mbid AS artist_mbid,
-    ae.name AS artist_name,
-    COUNT(p.uri) AS play_count
-FROM artists_extended ae
-INNER JOIN play_to_artists_extended ptae ON ae.id = ptae.artist_id
-INNER JOIN plays p ON p.uri = ptae.play_uri
-WHERE p.played_time >= NOW() - INTERVAL '30 days'
-GROUP BY ae.id, ae.mbid, ae.name
-ORDER BY COUNT(p.uri) DESC;
-
-DROP MATERIALIZED VIEW IF EXISTS mv_top_artists_for_user_30days;
-CREATE MATERIALIZED VIEW mv_top_artists_for_user_30days AS
-SELECT
-    prof.did,
-    ae.id AS artist_id,
-    ae.mbid AS artist_mbid,
-    ae.name AS artist_name,
-    COUNT(p.uri) AS play_count
-FROM artists_extended ae
-INNER JOIN play_to_artists_extended ptae ON ae.id = ptae.artist_id
-INNER JOIN plays p ON p.uri = ptae.play_uri
-INNER JOIN profiles prof ON prof.did = p.did
-WHERE p.played_time >= NOW() - INTERVAL '30 days'
-GROUP BY prof.did, ae.id, ae.mbid, ae.name
-ORDER BY COUNT(p.uri) DESC;
-
-DROP MATERIALIZED VIEW IF EXISTS mv_top_artists_for_user_7days;
-CREATE MATERIALIZED VIEW mv_top_artists_for_user_7days AS
-SELECT
-    prof.did,
-    ae.id AS artist_id,
-    ae.mbid AS artist_mbid,
-    ae.name AS artist_name,
-    COUNT(p.uri) AS play_count
-FROM artists_extended ae
-INNER JOIN play_to_artists_extended ptae ON ae.id = ptae.artist_id
-INNER JOIN plays p ON p.uri = ptae.play_uri
-INNER JOIN profiles prof ON prof.did = p.did
-WHERE p.played_time >= NOW() - INTERVAL '7 days'
-GROUP BY prof.did, ae.id, ae.mbid, ae.name
-ORDER BY COUNT(p.uri) DESC;
-
--- Comment explaining the migration strategy
-COMMENT ON TABLE artists_extended IS 'Extended artists table that supports both MusicBrainz and non-MusicBrainz artists. Uses serial ID as primary key with optional MBID.';
-COMMENT ON TABLE play_to_artists_extended IS 'Junction table linking plays to artists using the new artists_extended table structure.';
-COMMENT ON COLUMN plays.artist_names_raw IS 'Raw artist names as JSON array for plays without MusicBrainz data, used as fallback when artist relationships cannot be established.';

services/migrations/20241220000004_synthetic_mbids.sql

···
--- Migration to support synthetic MBIDs for artists without MusicBrainz data
--- This ensures all artists have some form of ID while maintaining uniqueness
-
--- Enable UUID extension for v5 UUID generation
-CREATE EXTENSION IF NOT EXISTS "uuid-ossp";
-
--- Add a column to track MBID type (musicbrainz, synthetic, unknown)
-ALTER TABLE artists_extended ADD COLUMN mbid_type TEXT DEFAULT 'unknown' NOT NULL;
-
--- Add check constraint for valid MBID types
-ALTER TABLE artists_extended ADD CONSTRAINT chk_mbid_type
-    CHECK (mbid_type IN ('musicbrainz', 'synthetic', 'unknown'));
-
--- Update existing records to set proper MBID type
-UPDATE artists_extended SET mbid_type = 'musicbrainz' WHERE mbid IS NOT NULL;
-
--- Drop the unique constraint on name_normalized for null MBIDs since we'll handle duplicates differently
-DROP INDEX IF EXISTS idx_artists_extended_name_unique;
-
--- Add index for efficient querying by MBID type
-CREATE INDEX idx_artists_extended_mbid_type ON artists_extended (mbid_type);
-
--- Create a view to easily work with different artist types
-CREATE VIEW artists_with_type AS
-SELECT
-    id,
-    mbid,
-    name,
-    mbid_type,
-    play_count,
-    created_at,
-    updated_at,
-    -- For synthetic MBIDs, we can show the source name used for generation
-    CASE
-        WHEN mbid_type = 'synthetic' THEN 'Generated from: ' || name
-        WHEN mbid_type = 'musicbrainz' THEN 'MusicBrainz: ' || mbid::text
-        ELSE 'No MBID available'
-    END as mbid_info
-FROM artists_extended;
-
--- Update materialized views to include MBID type information
-DROP MATERIALIZED VIEW IF EXISTS mv_artist_play_counts;
-CREATE MATERIALIZED VIEW mv_artist_play_counts AS
-SELECT
-    ae.id AS artist_id,
-    ae.mbid AS artist_mbid,
-    ae.name AS artist_name,
-    ae.mbid_type,
-    COUNT(p.uri) AS play_count
-FROM
-    artists_extended ae
-    LEFT JOIN play_to_artists_extended ptae ON ae.id = ptae.artist_id
-    LEFT JOIN plays p ON p.uri = ptae.play_uri
-GROUP BY
-    ae.id, ae.mbid, ae.name, ae.mbid_type;
-
-CREATE UNIQUE INDEX idx_mv_artist_play_counts_with_type ON mv_artist_play_counts (artist_id);
-
--- Add comments explaining the synthetic MBID system
-COMMENT ON COLUMN artists_extended.mbid_type IS 'Type of MBID: musicbrainz (real), synthetic (generated), or unknown (legacy data)';
-COMMENT ON COLUMN artists_extended.mbid IS 'MusicBrainz ID (for musicbrainz type) or synthetic UUID (for synthetic type)';
-COMMENT ON VIEW artists_with_type IS 'View that provides human-readable information about artist MBID sources';
-
--- Add a function to generate synthetic MBIDs
-CREATE OR REPLACE FUNCTION generate_synthetic_mbid(artist_name TEXT) RETURNS UUID AS $$
-DECLARE
-    namespace_uuid UUID := '6ba7b810-9dad-11d1-80b4-00c04fd430c8'; -- DNS namespace
-    result_uuid UUID;
-BEGIN
-    -- Generate deterministic UUID v5 based on artist name
-    SELECT uuid_generate_v5(namespace_uuid, artist_name) INTO result_uuid;
-    RETURN result_uuid;
-END;
-$$ LANGUAGE plpgsql IMMUTABLE;
-
-COMMENT ON FUNCTION generate_synthetic_mbid IS 'Generates a deterministic UUID v5 for artist names without MusicBrainz IDs';

services/migrations/20241220000005_fuzzy_matching.sql

···
--- Migration to add fuzzy text matching capabilities
--- This enables better artist name matching using trigram similarity
-
--- Enable pg_trgm extension for trigram similarity matching
-CREATE EXTENSION IF NOT EXISTS pg_trgm;
-
--- Create indexes for efficient trigram matching on artist names
-CREATE INDEX idx_artists_extended_name_trgm ON artists_extended USING gin (name gin_trgm_ops);
-CREATE INDEX idx_artists_extended_name_normalized_trgm ON artists_extended USING gin (name_normalized gin_trgm_ops);
-
--- Create a function to calculate comprehensive artist similarity
-CREATE OR REPLACE FUNCTION calculate_artist_similarity(
-    input_name TEXT,
-    existing_name TEXT,
-    input_album TEXT DEFAULT NULL,
-    existing_album TEXT DEFAULT NULL
-) RETURNS FLOAT AS $$
-DECLARE
-    name_similarity FLOAT;
-    album_similarity FLOAT := 0.0;
-    final_score FLOAT;
-BEGIN
-    -- Calculate trigram similarity for artist names
-    name_similarity := similarity(LOWER(TRIM(input_name)), LOWER(TRIM(existing_name)));
-
-    -- Boost for exact matches after normalization
-    IF LOWER(TRIM(regexp_replace(input_name, '[^a-zA-Z0-9\s]', '', 'g'))) =
-       LOWER(TRIM(regexp_replace(existing_name, '[^a-zA-Z0-9\s]', '', 'g'))) THEN
-        name_similarity := GREATEST(name_similarity, 0.95);
-    END IF;
-
-    -- Factor in album similarity if both are provided
-    IF input_album IS NOT NULL AND existing_album IS NOT NULL THEN
-        album_similarity := similarity(LOWER(TRIM(input_album)), LOWER(TRIM(existing_album)));
-        -- Weight: 80% name, 20% album
-        final_score := (name_similarity * 0.8) + (album_similarity * 0.2);
-    ELSE
-        final_score := name_similarity;
-    END IF;
-
-    RETURN final_score;
-END;
-$$ LANGUAGE plpgsql IMMUTABLE;
-
--- Create a view for fuzzy artist matching with confidence scores
-CREATE VIEW fuzzy_artist_matches AS
-SELECT DISTINCT
-    ae1.id as query_artist_id,
-    ae1.name as query_artist_name,
-    ae1.mbid_type as query_mbid_type,
-    ae2.id as match_artist_id,
-    ae2.name as match_artist_name,
-    ae2.mbid as match_mbid,
-    ae2.mbid_type as match_mbid_type,
-    similarity(LOWER(TRIM(ae1.name)), LOWER(TRIM(ae2.name))) as name_similarity,
-    CASE
-        WHEN ae2.mbid_type = 'musicbrainz' THEN 'upgrade_to_mb'
-        WHEN ae1.mbid_type = 'musicbrainz' AND ae2.mbid_type = 'synthetic' THEN 'consolidate_to_mb'
-        ELSE 'merge_synthetic'
-    END as match_action
-FROM artists_extended ae1
-CROSS JOIN artists_extended ae2
-WHERE ae1.id != ae2.id
-AND similarity(LOWER(TRIM(ae1.name)), LOWER(TRIM(ae2.name))) > 0.8
-AND (
-    ae1.mbid_type = 'synthetic' OR ae2.mbid_type = 'musicbrainz'
-);
-
--- Add comments
-COMMENT ON EXTENSION pg_trgm IS 'Trigram extension for fuzzy text matching';
-COMMENT ON INDEX idx_artists_extended_name_trgm IS 'GIN index for trigram similarity on artist names';
-COMMENT ON FUNCTION calculate_artist_similarity IS 'Calculates similarity score between artists considering name and optional album context';
-COMMENT ON VIEW fuzzy_artist_matches IS 'Shows potential artist matches with confidence scores and recommended actions';
-
--- Create a function to suggest artist consolidations
-CREATE OR REPLACE FUNCTION suggest_artist_consolidations(min_similarity FLOAT DEFAULT 0.9)
-RETURNS TABLE(
-    action TEXT,
-    synthetic_artist TEXT,
-    target_artist TEXT,
-    similarity_score FLOAT,
-    synthetic_plays INTEGER,
-    target_plays INTEGER
-) AS $$
-BEGIN
-    RETURN QUERY
-    SELECT
-        fam.match_action as action,
-        fam.query_artist_name as synthetic_artist,
-        fam.match_artist_name as target_artist,
-        fam.name_similarity as similarity_score,
-        (SELECT COUNT(*)::INTEGER FROM play_to_artists_extended WHERE artist_id = fam.query_artist_id) as synthetic_plays,
-        (SELECT COUNT(*)::INTEGER FROM play_to_artists_extended WHERE artist_id = fam.match_artist_id) as target_plays
-    FROM fuzzy_artist_matches fam
-    WHERE fam.name_similarity >= min_similarity
-    AND fam.match_action = 'upgrade_to_mb'
-    ORDER BY fam.name_similarity DESC, synthetic_plays DESC;
-END;
-$$ LANGUAGE plpgsql;
-
-COMMENT ON FUNCTION suggest_artist_consolidations IS 'Returns suggestions for consolidating synthetic artists with MusicBrainz artists based on similarity';

services/migrations/20241220000006_discriminant_fields.sql

···
--- Migration to add discriminant fields for track and release variants
--- This enables proper handling of different versions while maintaining grouping capabilities
-
--- Add discriminant fields to plays table
-ALTER TABLE plays ADD COLUMN track_discriminant TEXT;
-ALTER TABLE plays ADD COLUMN release_discriminant TEXT;
-
--- Add discriminant field to releases table
-ALTER TABLE releases ADD COLUMN discriminant TEXT;
-
--- Add discriminant field to recordings table
-ALTER TABLE recordings ADD COLUMN discriminant TEXT;
-
--- Create indexes for efficient searching and filtering
-CREATE INDEX idx_plays_track_discriminant ON plays (track_discriminant);
-CREATE INDEX idx_plays_release_discriminant ON plays (release_discriminant);
-CREATE INDEX idx_releases_discriminant ON releases (discriminant);
-CREATE INDEX idx_recordings_discriminant ON recordings (discriminant);
-
--- Create composite indexes for grouping by base name + discriminant
-CREATE INDEX idx_plays_track_name_discriminant ON plays (track_name, track_discriminant);
-CREATE INDEX idx_plays_release_name_discriminant ON plays (release_name, release_discriminant);
-
--- Update materialized views to include discriminant information
-DROP MATERIALIZED VIEW IF EXISTS mv_release_play_counts;
-CREATE MATERIALIZED VIEW mv_release_play_counts AS
-SELECT
-    r.mbid AS release_mbid,
-    r.name AS release_name,
-    r.discriminant AS release_discriminant,
-    COUNT(p.uri) AS play_count
-FROM
-    releases r
-    LEFT JOIN plays p ON p.release_mbid = r.mbid
-GROUP BY
-    r.mbid, r.name, r.discriminant;
-
-CREATE UNIQUE INDEX idx_mv_release_play_counts_discriminant ON mv_release_play_counts (release_mbid);
-
-DROP MATERIALIZED VIEW IF EXISTS mv_recording_play_counts;
-CREATE MATERIALIZED VIEW mv_recording_play_counts AS
-SELECT
-    rec.mbid AS recording_mbid,
-    rec.name AS recording_name,
-    rec.discriminant AS recording_discriminant,
-    COUNT(p.uri) AS play_count
-FROM
-    recordings rec
-    LEFT JOIN plays p ON p.recording_mbid = rec.mbid
-GROUP BY
-    rec.mbid, rec.name, rec.discriminant;
-
-CREATE UNIQUE INDEX idx_mv_recording_play_counts_discriminant ON mv_recording_play_counts (recording_mbid);
-
--- Create views for analyzing track/release variants
-CREATE VIEW track_variants AS
-SELECT
-    track_name,
-    track_discriminant,
-    COUNT(*) AS play_count,
-    COUNT(DISTINCT did) AS unique_listeners,
-    COUNT(DISTINCT recording_mbid) AS unique_recordings
-FROM plays
-WHERE track_name IS NOT NULL
-GROUP BY track_name, track_discriminant
-ORDER BY track_name, play_count DESC;
-
-CREATE VIEW release_variants AS
-SELECT
-    release_name,
-    release_discriminant,
-    COUNT(*) AS play_count,
-    COUNT(DISTINCT did) AS unique_listeners,
-    COUNT(DISTINCT release_mbid) AS unique_releases
-FROM plays
-WHERE release_name IS NOT NULL
-GROUP BY release_name, release_discriminant
-ORDER BY release_name, play_count DESC;
-
--- Create function to extract potential discriminants from existing names
-CREATE OR REPLACE FUNCTION extract_discriminant(name_text TEXT) RETURNS TEXT AS $$
-DECLARE
-    discriminant_patterns TEXT[] := ARRAY[
-        '\(([^)]*(?:deluxe|remaster|remastered|extended|acoustic|live|radio|edit|version|remix|demo|instrumental|explicit|clean|bonus).*?)\)',
-        '\[([^]]*(?:deluxe|remaster|remastered|extended|acoustic|live|radio|edit|version|remix|demo|instrumental|explicit|clean|bonus).*?)\]',
-        '\{([^}]*(?:deluxe|remaster|remastered|extended|acoustic|live|radio|edit|version|remix|demo|instrumental|explicit|clean|bonus).*?)\}'
-    ];
-    pattern TEXT;
-    match_result TEXT;
-BEGIN
-    -- Try each pattern to find discriminant information
-    FOREACH pattern IN ARRAY discriminant_patterns
-    LOOP
-        SELECT substring(name_text FROM pattern) INTO match_result;
-        IF match_result IS NOT NULL AND length(trim(match_result)) > 0 THEN
-            RETURN trim(match_result);
-        END IF;
-    END LOOP;
-
-    RETURN NULL;
-END;
-$$ LANGUAGE plpgsql IMMUTABLE;
-
--- Create function to get base name without discriminant
-CREATE OR REPLACE FUNCTION get_base_name(name_text TEXT) RETURNS TEXT AS $$
-DECLARE
-    cleanup_patterns TEXT[] := ARRAY[
-        '\s*\([^)]*(?:deluxe|remaster|remastered|extended|acoustic|live|radio|edit|version|remix|demo|instrumental|explicit|clean|bonus).*?\)\s*',
-        '\s*\[[^]]*(?:deluxe|remaster|remastered|extended|acoustic|live|radio|edit|version|remix|demo|instrumental|explicit|clean|bonus).*?\]\s*',
-        '\s*\{[^}]*(?:deluxe|remaster|remastered|extended|acoustic|live|radio|edit|version|remix|demo|instrumental|explicit|clean|bonus).*?\}\s*'
-    ];
-    pattern TEXT;
-    result_text TEXT := name_text;
-BEGIN
-    -- Remove discriminant patterns to get base name
-    FOREACH pattern IN ARRAY cleanup_patterns
-    LOOP
-        result_text := regexp_replace(result_text, pattern, ' ', 'gi');
-    END LOOP;
-
-    -- Clean up extra whitespace
-    result_text := regexp_replace(trim(result_text), '\s+', ' ', 'g');
-
-    RETURN result_text;
-END;
-$$ LANGUAGE plpgsql IMMUTABLE;
-
--- Add comments explaining the discriminant system
-COMMENT ON COLUMN plays.track_discriminant IS 'Distinguishing information for track variants (e.g., "Acoustic Version", "Live at Wembley", "Radio Edit")';
-COMMENT ON COLUMN plays.release_discriminant IS 'Distinguishing information for release variants (e.g., "Deluxe Edition", "Remastered", "2023 Remaster")';
-COMMENT ON COLUMN releases.discriminant IS 'Distinguishing information for release variants to enable proper grouping';
-COMMENT ON COLUMN recordings.discriminant IS 'Distinguishing information for recording variants to enable proper grouping';
-
-COMMENT ON VIEW track_variants IS 'Shows all variants of tracks with their play counts and unique listeners';
-COMMENT ON VIEW release_variants IS 'Shows all variants of releases with their play counts and unique listeners';
-
-COMMENT ON FUNCTION extract_discriminant IS 'Extracts discriminant information from track/release names for migration purposes';
-COMMENT ON FUNCTION get_base_name IS 'Returns the base name without discriminant information for grouping purposes';

services/migrations/20241220000007_enhanced_discriminant_extraction.sql

···
--- Enhanced discriminant extraction with comprehensive edition/version patterns
--- This migration improves the auto-population of discriminants for better metadata handling
-
--- Drop existing functions to replace them with enhanced versions
-DROP FUNCTION IF EXISTS extract_discriminant(TEXT);
-DROP FUNCTION IF EXISTS get_base_name(TEXT);
-
--- Enhanced function to extract discriminants with comprehensive patterns
-CREATE OR REPLACE FUNCTION extract_discriminant(name_text TEXT) RETURNS TEXT AS $$
-DECLARE
-    -- Comprehensive patterns for discriminant extraction
-    discriminant_patterns TEXT[] := ARRAY[
-        -- Parentheses patterns
-        '\(([^)]*(?:deluxe|remaster|remastered|extended|acoustic|live|radio|edit|version|remix|demo|instrumental|explicit|clean|bonus|edition|special|limited|expanded|director''s|uncut|final|ultimate|platinum|gold|anniversary|collector''s|standard|enhanced|super|mega|ultra|plus|pro|premium|complete|definitive|classic|original|alternate|alternative|unreleased|rare|exclusive|digital|vinyl|cd|dvd|blu-ray|hdtv|web|retail|promo|single|ep|lp|maxi|mini|radio|club|dance|house|techno|trance|ambient|classical|jazz|folk|country|rock|pop|metal|punk|indie|alternative).*?)\)',
-        '\(([^)]*(?:\d{4}|\d{2})\s*(?:remaster|edition|version|mix|cut|release).*?)\)',
-        '\(([^)]*(?:vol\.|volume|pt\.|part|disc|disk|cd)\s*\d+.*?)\)',
-        '\(([^)]*(?:feat\.|featuring|ft\.|with|vs\.|versus|&|and)\s+.*?)\)',
-        '\(([^)]*(?:from|soundtrack|ost|score|theme).*?)\)',
-
-        -- Brackets patterns
-        '\[([^]]*(?:deluxe|remaster|remastered|extended|acoustic|live|radio|edit|version|remix|demo|instrumental|explicit|clean|bonus|edition|special|limited|expanded|director''s|uncut|final|ultimate|platinum|gold|anniversary|collector''s|standard|enhanced|super|mega|ultra|plus|pro|premium|complete|definitive|classic|original|alternate|alternative|unreleased|rare|exclusive|digital|vinyl|cd|dvd|blu-ray|hdtv|web|retail|promo|single|ep|lp|maxi|mini|radio|club|dance|house|techno|trance|ambient|classical|jazz|folk|country|rock|pop|metal|punk|indie|alternative).*?)\]',
-        '\[([^]]*(?:\d{4}|\d{2})\s*(?:remaster|edition|version|mix|cut|release).*?)\]',
-        '\[([^]]*(?:vol\.|volume|pt\.|part|disc|disk|cd)\s*\d+.*?)\]',
-        '\[([^]]*(?:feat\.|featuring|ft\.|with|vs\.|versus|&|and)\s+.*?)\]',
-        '\[([^]]*(?:from|soundtrack|ost|score|theme).*?)\]',
-
-        -- Braces patterns
-        '\{([^}]*(?:deluxe|remaster|remastered|extended|acoustic|live|radio|edit|version|remix|demo|instrumental|explicit|clean|bonus|edition|special|limited|expanded|director''s|uncut|final|ultimate|platinum|gold|anniversary|collector''s|standard|enhanced|super|mega|ultra|plus|pro|premium|complete|definitive|classic|original|alternate|alternative|unreleased|rare|exclusive|digital|vinyl|cd|dvd|blu-ray|hdtv|web|retail|promo|single|ep|lp|maxi|mini|radio|club|dance|house|techno|trance|ambient|classical|jazz|folk|country|rock|pop|metal|punk|indie|alternative).*?)\}',
-        '\{([^}]*(?:\d{4}|\d{2})\s*(?:remaster|edition|version|mix|cut|release).*?)\}',
-        '\{([^}]*(?:vol\.|volume|pt\.|part|disc|disk|cd)\s*\d+.*?)\}',
-        '\{([^}]*(?:feat\.|featuring|ft\.|with|vs\.|versus|&|and)\s+.*?)\}',
-        '\{([^}]*(?:from|soundtrack|ost|score|theme).*?)\}',
-
-        -- Dash/hyphen patterns (common for editions)
-        '[-–—]\s*([^-–—]*(?:deluxe|remaster|remastered|extended|acoustic|live|radio|edit|version|remix|demo|instrumental|explicit|clean|bonus|edition|special|limited|expanded|director''s|uncut|final|ultimate|platinum|gold|anniversary|collector''s|standard|enhanced|super|mega|ultra|plus|pro|premium|complete|definitive|classic|original|alternate|alternative|unreleased|rare|exclusive|digital|vinyl|cd|dvd|blu-ray).*?)$',
-        '[-–—]\s*(\d{4}|\d{2})\s*(?:remaster|edition|version|mix|cut|release).*?$',
-
-        -- Colon patterns (common for subtitles and versions)
-        ':\s*([^:]*(?:deluxe|remaster|remastered|extended|acoustic|live|radio|edit|version|remix|demo|instrumental|explicit|clean|bonus|edition|special|limited|expanded|director''s|uncut|final|ultimate|platinum|gold|anniversary|collector''s|standard|enhanced|super|mega|ultra|plus|pro|premium|complete|definitive|classic|original|alternate|alternative|unreleased|rare|exclusive).*?)$',
-        ':\s*(\d{4}|\d{2})\s*(?:remaster|edition|version|mix|cut|release).*?$'
-    ];
-
-    pattern TEXT;
-    match_result TEXT;
-BEGIN
-    -- Return early if input is null or empty
-    IF name_text IS NULL OR trim(name_text) = '' THEN
-        RETURN NULL;
-    END IF;
-
-    -- Try each pattern to find discriminant information
-    FOREACH pattern IN ARRAY discriminant_patterns
-    LOOP
-        SELECT substring(name_text FROM pattern COLLATE "C") INTO match_result;
-        IF match_result IS NOT NULL AND length(trim(match_result)) > 0 THEN
-            -- Clean up the match result
-            match_result := trim(match_result);
-            -- Remove leading/trailing punctuation
-            match_result := regexp_replace(match_result, '^[^\w]+|[^\w]+$', '', 'g');
-            -- Ensure it's not just whitespace or empty after cleanup
-            IF length(trim(match_result)) > 0 THEN
-                RETURN match_result;
-            END IF;
-        END IF;
-    END LOOP;
-
-    RETURN NULL;
-END;
-$$ LANGUAGE plpgsql IMMUTABLE;
-
--- Enhanced function to get base name without discriminant
-CREATE OR REPLACE FUNCTION get_base_name(name_text TEXT) RETURNS TEXT AS $$
-DECLARE
-    -- Comprehensive cleanup patterns matching the extraction patterns
-    cleanup_patterns TEXT[] := ARRAY[
-        -- Remove parentheses content
-        '\s*\([^)]*(?:deluxe|remaster|remastered|extended|acoustic|live|radio|edit|version|remix|demo|instrumental|explicit|clean|bonus|edition|special|limited|expanded|director''s|uncut|final|ultimate|platinum|gold|anniversary|collector''s|standard|enhanced|super|mega|ultra|plus|pro|premium|complete|definitive|classic|original|alternate|alternative|unreleased|rare|exclusive|digital|vinyl|cd|dvd|blu-ray|hdtv|web|retail|promo|single|ep|lp|maxi|mini|radio|club|dance|house|techno|trance|ambient|classical|jazz|folk|country|rock|pop|metal|punk|indie|alternative).*?\)\s*',
-        '\s*\([^)]*(?:\d{4}|\d{2})\s*(?:remaster|edition|version|mix|cut|release).*?\)\s*',
-        '\s*\([^)]*(?:vol\.|volume|pt\.|part|disc|disk|cd)\s*\d+.*?\)\s*',
-        '\s*\([^)]*(?:feat\.|featuring|ft\.|with|vs\.|versus|&|and)\s+.*?\)\s*',
-        '\s*\([^)]*(?:from|soundtrack|ost|score|theme).*?\)\s*',
-
-        -- Remove brackets content
-        '\s*\[[^]]*(?:deluxe|remaster|remastered|extended|acoustic|live|radio|edit|version|remix|demo|instrumental|explicit|clean|bonus|edition|special|limited|expanded|director''s|uncut|final|ultimate|platinum|gold|anniversary|collector''s|standard|enhanced|super|mega|ultra|plus|pro|premium|complete|definitive|classic|original|alternate|alternative|unreleased|rare|exclusive|digital|vinyl|cd|dvd|blu-ray|hdtv|web|retail|promo|single|ep|lp|maxi|mini|radio|club|dance|house|techno|trance|ambient|classical|jazz|folk|country|rock|pop|metal|punk|indie|alternative).*?\]\s*',
-        '\s*\[[^]]*(?:\d{4}|\d{2})\s*(?:remaster|edition|version|mix|cut|release).*?\]\s*',
-        '\s*\[[^]]*(?:vol\.|volume|pt\.|part|disc|disk|cd)\s*\d+.*?\]\s*',
-        '\s*\[[^]]*(?:feat\.|featuring|ft\.|with|vs\.|versus|&|and)\s+.*?\]\s*',
-        '\s*\[[^]]*(?:from|soundtrack|ost|score|theme).*?\]\s*',
-
-        -- Remove braces content
-        '\s*\{[^}]*(?:deluxe|remaster|remastered|extended|acoustic|live|radio|edit|version|remix|demo|instrumental|explicit|clean|bonus|edition|special|limited|expanded|director''s|uncut|final|ultimate|platinum|gold|anniversary|collector''s|standard|enhanced|super|mega|ultra|plus|pro|premium|complete|definitive|classic|original|alternate|alternative|unreleased|rare|exclusive|digital|vinyl|cd|dvd|blu-ray|hdtv|web|retail|promo|single|ep|lp|maxi|mini|radio|club|dance|house|techno|trance|ambient|classical|jazz|folk|country|rock|pop|metal|punk|indie|alternative).*?\}\s*',
-        '\s*\{[^}]*(?:\d{4}|\d{2})\s*(?:remaster|edition|version|mix|cut|release).*?\}\s*',
-        '\s*\{[^}]*(?:vol\.|volume|pt\.|part|disc|disk|cd)\s*\d+.*?\}\s*',
-        '\s*\{[^}]*(?:feat\.|featuring|ft\.|with|vs\.|versus|&|and)\s+.*?\}\s*',
-        '\s*\{[^}]*(?:from|soundtrack|ost|score|theme).*?\}\s*',
-
-        -- Remove dash/hyphen patterns
-        '\s*[-–—]\s*[^-–—]*(?:deluxe|remaster|remastered|extended|acoustic|live|radio|edit|version|remix|demo|instrumental|explicit|clean|bonus|edition|special|limited|expanded|director''s|uncut|final|ultimate|platinum|gold|anniversary|collector''s|standard|enhanced|super|mega|ultra|plus|pro|premium|complete|definitive|classic|original|alternate|alternative|unreleased|rare|exclusive|digital|vinyl|cd|dvd|blu-ray).*?$',
-        '\s*[-–—]\s*(?:\d{4}|\d{2})\s*(?:remaster|edition|version|mix|cut|release).*?$',
-
-        -- Remove colon patterns
-        '\s*:\s*[^:]*(?:deluxe|remaster|remastered|extended|acoustic|live|radio|edit|version|remix|demo|instrumental|explicit|clean|bonus|edition|special|limited|expanded|director''s|uncut|final|ultimate|platinum|gold|anniversary|collector''s|standard|enhanced|super|mega|ultra|plus|pro|premium|complete|definitive|classic|original|alternate|alternative|unreleased|rare|exclusive).*?$',
-        '\s*:\s*(?:\d{4}|\d{2})\s*(?:remaster|edition|version|mix|cut|release).*?$'
-    ];
-
-    pattern TEXT;
-    result_text TEXT := name_text;
-BEGIN
-    -- Return early if input is null or empty
-    IF name_text IS NULL OR trim(name_text) = '' THEN
-        RETURN name_text;
-    END IF;
-
-    -- Remove discriminant patterns to get base name
-    FOREACH pattern IN ARRAY cleanup_patterns
-    LOOP
-        result_text := regexp_replace(result_text, pattern, ' ', 'gi');
-    END LOOP;
-
-    -- Clean up extra whitespace and normalize
-    result_text := regexp_replace(trim(result_text), '\s+', ' ', 'g');
-
-    -- Remove trailing punctuation that might be left after removal
-    result_text := regexp_replace(result_text, '[,;:\-–—]\s*$', '', 'g');
-    result_text := trim(result_text);
-
-    -- Ensure we don't return an empty string
-    IF length(result_text) = 0 THEN
-        RETURN name_text;
-    END IF;
-
-    RETURN result_text;
-END;
-$$ LANGUAGE plpgsql IMMUTABLE;
-
--- Create function to extract discriminant specifically for editions and versions
-CREATE OR REPLACE FUNCTION extract_edition_discriminant(name_text TEXT) RETURNS TEXT AS $$
-DECLARE
-    -- Focused patterns for edition/version extraction
-    edition_patterns TEXT[] := ARRAY[
-        -- Edition patterns
-        '\(([^)]*edition[^)]*)\)',
-        '\[([^]]*edition[^]]*)\]',
-        '\{([^}]*edition[^}]*)\}',
-        '[-–—]\s*([^-–—]*edition[^-–—]*)$',
-        ':\s*([^:]*edition[^:]*)$',
-
-        -- Version patterns
-        '\(([^)]*version[^)]*)\)',
-        '\[([^]]*version[^]]*)\]',
-        '\{([^}]*version[^}]*)\}',
-        '[-–—]\s*([^-–—]*version[^-–—]*)$',
-        ':\s*([^:]*version[^:]*)$',
-
-        -- Remaster patterns
-        '\(([^)]*remaster[^)]*)\)',
-        '\[([^]]*remaster[^]]*)\]',
-        '\{([^}]*remaster[^}]*)\}',
-        '[-–—]\s*([^-–—]*remaster[^-–—]*)$',
-        ':\s*([^:]*remaster[^:]*)$',
-
-        -- Year-based patterns
-        '\(([^)]*(?:\d{4}|\d{2})\s*(?:remaster|edition|version|mix|cut|release)[^)]*)\)',
-        '\[([^]]*(?:\d{4}|\d{2})\s*(?:remaster|edition|version|mix|cut|release)[^]]*)\]',
-        '\{([^}]*(?:\d{4}|\d{2})\s*(?:remaster|edition|version|mix|cut|release)[^}]*)\}'
-    ];
-
-    pattern TEXT;
-    match_result TEXT;
-BEGIN
-    -- Return early if input is null or empty
-    IF name_text IS NULL OR trim(name_text) = '' THEN
-        RETURN NULL;
-    END IF;
-
-    -- Try edition-specific patterns first
-    FOREACH pattern IN ARRAY edition_patterns
-    LOOP
-        SELECT substring(name_text FROM pattern COLLATE "C") INTO match_result;
-        IF match_result IS NOT NULL AND length(trim(match_result)) > 0 THEN
-            match_result := trim(match_result);
-            match_result := regexp_replace(match_result, '^[^\w]+|[^\w]+$', '', 'g');
-            IF length(trim(match_result)) > 0 THEN
-                RETURN match_result;
-            END IF;
-        END IF;
-    END LOOP;
-
-    RETURN NULL;
-END;
-$$ LANGUAGE plpgsql IMMUTABLE;
-
--- Update recordings table to populate discriminants from existing names
-UPDATE recordings
-SET discriminant = extract_discriminant(name)
-WHERE discriminant IS NULL
-  AND extract_discriminant(name) IS NOT NULL;
-
--- Update releases table to populate discriminants from existing names
-UPDATE releases
-SET discriminant = extract_discriminant(name)
-WHERE discriminant IS NULL
-  AND extract_discriminant(name) IS NOT NULL;
-
--- Update plays table to populate discriminants from existing names where not already set
-UPDATE plays
-SET track_discriminant = extract_discriminant(track_name)
-WHERE track_discriminant IS NULL
-  AND extract_discriminant(track_name) IS NOT NULL;
-
-UPDATE plays
-SET release_discriminant = extract_discriminant(release_name)
-WHERE release_discriminant IS NULL
-  AND release_name IS NOT NULL
-  AND extract_discriminant(release_name) IS NOT NULL;
-
--- Create indexes for efficient discriminant queries
-CREATE INDEX IF NOT EXISTS idx_recordings_name_discriminant ON recordings (name, discriminant);
-CREATE INDEX IF NOT EXISTS idx_releases_name_discriminant ON releases (name, discriminant);
-
--- Add comments for the new function
-COMMENT ON FUNCTION extract_discriminant IS 'Enhanced discriminant extraction supporting comprehensive edition/version patterns including parentheses, brackets, braces, dashes, and colons';
-COMMENT ON FUNCTION get_base_name IS 'Enhanced base name extraction removing comprehensive discriminant patterns to enable proper grouping';
-COMMENT ON FUNCTION extract_edition_discriminant IS 'Specialized function for extracting edition and version discriminants with focused patterns';
-
--- Create a view to show discriminant extraction results for analysis
-CREATE OR REPLACE VIEW discriminant_analysis AS
-SELECT
-    'recordings' as table_name,
-    name as original_name,
-    discriminant,
-    get_base_name(name) as base_name,
-    extract_discriminant(name) as extracted_discriminant,
-    extract_edition_discriminant(name) as edition_discriminant
-FROM recordings
-WHERE name IS NOT NULL
-UNION ALL
-SELECT
-    'releases' as table_name,
-    name as original_name,
-    discriminant,
-    get_base_name(name) as base_name,
-    extract_discriminant(name) as extracted_discriminant,
-    extract_edition_discriminant(name) as edition_discriminant
-FROM releases
-WHERE name IS NOT NULL;
-
-COMMENT ON VIEW discriminant_analysis IS 'Analysis view showing discriminant extraction results for quality assessment and debugging';
-
--- Refresh materialized views to include discriminant information
-REFRESH MATERIALIZED VIEW mv_release_play_counts;
-REFRESH MATERIALIZED VIEW mv_recording_play_counts;
-
--- Create summary statistics for discriminant usage
-CREATE OR REPLACE VIEW discriminant_stats AS
-SELECT
-    'recordings' as entity_type,
-    COUNT(*) as total_count,
-    COUNT(CASE WHEN discriminant IS NOT NULL THEN 1 END) as with_discriminant,
-    COUNT(CASE WHEN discriminant IS NULL AND extract_discriminant(name) IS NOT NULL THEN 1 END) as extractable_discriminant,
-    ROUND(
-        COUNT(CASE WHEN discriminant IS NOT NULL THEN 1 END) * 100.0 / COUNT(*), 2
-    ) as discriminant_percentage
-FROM recordings
-UNION ALL
-SELECT
-    'releases' as entity_type,
-    COUNT(*) as total_count,
-    COUNT(CASE WHEN discriminant IS NOT NULL THEN 1 END) as with_discriminant,
-    COUNT(CASE WHEN discriminant IS NULL AND extract_discriminant(name) IS NOT NULL THEN 1 END) as extractable_discriminant,
-    ROUND(
-        COUNT(CASE WHEN discriminant IS NOT NULL THEN 1 END) * 100.0 / COUNT(*), 2
-    ) as discriminant_percentage
-FROM releases;
-
-COMMENT ON VIEW discriminant_stats IS 'Statistics showing discriminant usage and extraction potential across entity types';

services/migrations/20241220000008_fix_discriminant_case_sensitivity.sql

···
--- Fix case sensitivity in discriminant extraction patterns
--- This migration updates the discriminant extraction functions to properly handle case-insensitive matching
-
--- Drop dependent views first, then functions, then recreate everything
-DROP VIEW IF EXISTS discriminant_analysis CASCADE;
-DROP VIEW IF EXISTS discriminant_stats CASCADE;
-
--- Drop existing functions to replace with case-insensitive versions
-DROP FUNCTION IF EXISTS extract_discriminant(TEXT) CASCADE;
-DROP FUNCTION IF EXISTS get_base_name(TEXT) CASCADE;
-DROP FUNCTION IF EXISTS extract_edition_discriminant(TEXT) CASCADE;
-
--- Enhanced function to extract discriminants with case-insensitive matching
-CREATE OR REPLACE FUNCTION extract_discriminant(name_text TEXT) RETURNS TEXT AS $$
-DECLARE
-    -- Comprehensive patterns for discriminant extraction with case-insensitive flags
-    discriminant_patterns TEXT[] := ARRAY[
-        -- Parentheses patterns
-        '(?i)\(([^)]*(?:deluxe|remaster|remastered|extended|acoustic|live|radio|edit|version|remix|demo|instrumental|explicit|clean|bonus|edition|special|limited|expanded|director''s|uncut|final|ultimate|platinum|gold|anniversary|collector''s|standard|enhanced|super|mega|ultra|plus|pro|premium|complete|definitive|classic|original|alternate|alternative|unreleased|rare|exclusive|digital|vinyl|cd|dvd|blu-ray|hdtv|web|retail|promo|single|ep|lp|maxi|mini|radio|club|dance|house|techno|trance|ambient|classical|jazz|folk|country|rock|pop|metal|punk|indie|alternative).*?)\)',
-        '(?i)\(([^)]*(?:\d{4}|\d{2})\s*(?:remaster|edition|version|mix|cut|release).*?)\)',
-        '(?i)\(([^)]*(?:vol\.|volume|pt\.|part|disc|disk|cd)\s*\d+.*?)\)',
-        '(?i)\(([^)]*(?:feat\.|featuring|ft\.|with|vs\.|versus|&|and)\s+.*?)\)',
-        '(?i)\(([^)]*(?:from|soundtrack|ost|score|theme).*?)\)',
-
-        -- Brackets patterns
-        '(?i)\[([^]]*(?:deluxe|remaster|remastered|extended|acoustic|live|radio|edit|version|remix|demo|instrumental|explicit|clean|bonus|edition|special|limited|expanded|director''s|uncut|final|ultimate|platinum|gold|anniversary|collector''s|standard|enhanced|super|mega|ultra|plus|pro|premium|complete|definitive|classic|original|alternate|alternative|unreleased|rare|exclusive|digital|vinyl|cd|dvd|blu-ray|hdtv|web|retail|promo|single|ep|lp|maxi|mini|radio|club|dance|house|techno|trance|ambient|classical|jazz|folk|country|rock|pop|metal|punk|indie|alternative).*?)\]',
-        '(?i)\[([^]]*(?:\d{4}|\d{2})\s*(?:remaster|edition|version|mix|cut|release).*?)\]',
-        '(?i)\[([^]]*(?:vol\.|volume|pt\.|part|disc|disk|cd)\s*\d+.*?)\]',
-        '(?i)\[([^]]*(?:feat\.|featuring|ft\.|with|vs\.|versus|&|and)\s+.*?)\]',
-        '(?i)\[([^]]*(?:from|soundtrack|ost|score|theme).*?)\]',
-
-        -- Braces patterns
-        '(?i)\{([^}]*(?:deluxe|remaster|remastered|extended|acoustic|live|radio|edit|version|remix|demo|instrumental|explicit|clean|bonus|edition|special|limited|expanded|director''s|uncut|final|ultimate|platinum|gold|anniversary|collector''s|standard|enhanced|super|mega|ultra|plus|pro|premium|complete|definitive|classic|original|alternate|alternative|unreleased|rare|exclusive|digital|vinyl|cd|dvd|blu-ray|hdtv|web|retail|promo|single|ep|lp|maxi|mini|radio|club|dance|house|techno|trance|ambient|classical|jazz|folk|country|rock|pop|metal|punk|indie|alternative).*?)\}',
-        '(?i)\{([^}]*(?:\d{4}|\d{2})\s*(?:remaster|edition|version|mix|cut|release).*?)\}',
-        '(?i)\{([^}]*(?:vol\.|volume|pt\.|part|disc|disk|cd)\s*\d+.*?)\}',
-        '(?i)\{([^}]*(?:feat\.|featuring|ft\.|with|vs\.|versus|&|and)\s+.*?)\}',
-        '(?i)\{([^}]*(?:from|soundtrack|ost|score|theme).*?)\}',
-
-        -- Dash/hyphen patterns (common for editions)
-        '(?i)[-–—]\s*([^-–—]*(?:deluxe|remaster|remastered|extended|acoustic|live|radio|edit|version|remix|demo|instrumental|explicit|clean|bonus|edition|special|limited|expanded|director''s|uncut|final|ultimate|platinum|gold|anniversary|collector''s|standard|enhanced|super|mega|ultra|plus|pro|premium|complete|definitive|classic|original|alternate|alternative|unreleased|rare|exclusive|digital|vinyl|cd|dvd|blu-ray).*?)$',
-        '(?i)[-–—]\s*(\d{4}|\d{2})\s*(?:remaster|edition|version|mix|cut|release).*?$',
-
-        -- Colon patterns (common for subtitles and versions)
-        '(?i):\s*([^:]*(?:deluxe|remaster|remastered|extended|acoustic|live|radio|edit|version|remix|demo|instrumental|explicit|clean|bonus|edition|special|limited|expanded|director''s|uncut|final|ultimate|platinum|gold|anniversary|collector''s|standard|enhanced|super|mega|ultra|plus|pro|premium|complete|definitive|classic|original|alternate|alternative|unreleased|rare|exclusive).*?)$',
-        '(?i):\s*(\d{4}|\d{2})\s*(?:remaster|edition|version|mix|cut|release).*?$'
-    ];
-
-    pattern TEXT;
-    match_result TEXT;
-BEGIN
-    -- Return early if input is null or empty
-    IF name_text IS NULL OR trim(name_text) = '' THEN
-        RETURN NULL;
-    END IF;
-
-    -- Try each pattern to find discriminant information
-    FOREACH pattern IN ARRAY discriminant_patterns
-    LOOP
-        SELECT substring(name_text FROM pattern) INTO match_result;
-        IF match_result IS NOT NULL AND length(trim(match_result)) > 0 THEN
-            -- Clean up the match result
-            match_result := trim(match_result);
-            -- Remove leading/trailing punctuation
-            match_result := regexp_replace(match_result, '^[^\w]+|[^\w]+$', '', 'g');
-            -- Ensure it's not just whitespace or empty after cleanup
-            IF length(trim(match_result)) > 0 THEN
-                RETURN match_result;
-            END IF;
-        END IF;
-    END LOOP;
-
-    RETURN NULL;
-END;
-$$ LANGUAGE plpgsql IMMUTABLE;
-
--- Enhanced function to get base name without discriminant with case-insensitive matching
-CREATE OR REPLACE FUNCTION get_base_name(name_text TEXT) RETURNS TEXT AS $$
-DECLARE
-    -- Comprehensive cleanup patterns matching the extraction patterns
-    cleanup_patterns TEXT[] := ARRAY[
-        -- Remove parentheses content
-        '(?i)\s*\([^)]*(?:deluxe|remaster|remastered|extended|acoustic|live|radio|edit|version|remix|demo|instrumental|explicit|clean|bonus|edition|special|limited|expanded|director''s|uncut|final|ultimate|platinum|gold|anniversary|collector''s|standard|enhanced|super|mega|ultra|plus|pro|premium|complete|definitive|classic|original|alternate|alternative|unreleased|rare|exclusive|digital|vinyl|cd|dvd|blu-ray|hdtv|web|retail|promo|single|ep|lp|maxi|mini|radio|club|dance|house|techno|trance|ambient|classical|jazz|folk|country|rock|pop|metal|punk|indie|alternative).*?\)\s*',
-        '(?i)\s*\([^)]*(?:\d{4}|\d{2})\s*(?:remaster|edition|version|mix|cut|release).*?\)\s*',
-        '(?i)\s*\([^)]*(?:vol\.|volume|pt\.|part|disc|disk|cd)\s*\d+.*?\)\s*',
-        '(?i)\s*\([^)]*(?:feat\.|featuring|ft\.|with|vs\.|versus|&|and)\s+.*?\)\s*',
-        '(?i)\s*\([^)]*(?:from|soundtrack|ost|score|theme).*?\)\s*',
-
-        -- Remove brackets content
-        '(?i)\s*\[[^]]*(?:deluxe|remaster|remastered|extended|acoustic|live|radio|edit|version|remix|demo|instrumental|explicit|clean|bonus|edition|special|limited|expanded|director''s|uncut|final|ultimate|platinum|gold|anniversary|collector''s|standard|enhanced|super|mega|ultra|plus|pro|premium|complete|definitive|classic|original|alternate|alternative|unreleased|rare|exclusive|digital|vinyl|cd|dvd|blu-ray|hdtv|web|retail|promo|single|ep|lp|maxi|mini|radio|club|dance|house|techno|trance|ambient|classical|jazz|folk|country|rock|pop|metal|punk|indie|alternative).*?\]\s*',
-        '(?i)\s*\[[^]]*(?:\d{4}|\d{2})\s*(?:remaster|edition|version|mix|cut|release).*?\]\s*',
-        '(?i)\s*\[[^]]*(?:vol\.|volume|pt\.|part|disc|disk|cd)\s*\d+.*?\]\s*',
-        '(?i)\s*\[[^]]*(?:feat\.|featuring|ft\.|with|vs\.|versus|&|and)\s+.*?\]\s*',
-        '(?i)\s*\[[^]]*(?:from|soundtrack|ost|score|theme).*?\]\s*',
-
-        -- Remove braces content
-        '(?i)\s*\{[^}]*(?:deluxe|remaster|remastered|extended|acoustic|live|radio|edit|version|remix|demo|instrumental|explicit|clean|bonus|edition|special|limited|expanded|director''s|uncut|final|ultimate|platinum|gold|anniversary|collector''s|standard|enhanced|super|mega|ultra|plus|pro|premium|complete|definitive|classic|original|alternate|alternative|unreleased|rare|exclusive|digital|vinyl|cd|dvd|blu-ray|hdtv|web|retail|promo|single|ep|lp|maxi|mini|radio|club|dance|house|techno|trance|ambient|classical|jazz|folk|country|rock|pop|metal|punk|indie|alternative).*?\}\s*',
-        '(?i)\s*\{[^}]*(?:\d{4}|\d{2})\s*(?:remaster|edition|version|mix|cut|release).*?\}\s*',
-        '(?i)\s*\{[^}]*(?:vol\.|volume|pt\.|part|disc|disk|cd)\s*\d+.*?\}\s*',
-        '(?i)\s*\{[^}]*(?:feat\.|featuring|ft\.|with|vs\.|versus|&|and)\s+.*?\}\s*',
-        '(?i)\s*\{[^}]*(?:from|soundtrack|ost|score|theme).*?\}\s*',
-
-        -- Remove dash/hyphen patterns
-        '(?i)\s*[-–—]\s*[^-–—]*(?:deluxe|remaster|remastered|extended|acoustic|live|radio|edit|version|remix|demo|instrumental|explicit|clean|bonus|edition|special|limited|expanded|director''s|uncut|final|ultimate|platinum|gold|anniversary|collector''s|standard|enhanced|super|mega|ultra|plus|pro|premium|complete|definitive|classic|original|alternate|alternative|unreleased|rare|exclusive|digital|vinyl|cd|dvd|blu-ray).*?$',
-        '(?i)\s*[-–—]\s*(?:\d{4}|\d{2})\s*(?:remaster|edition|version|mix|cut|release).*?$',
-
-        -- Remove colon patterns
-        '(?i)\s*:\s*[^:]*(?:deluxe|remaster|remastered|extended|acoustic|live|radio|edit|version|remix|demo|instrumental|explicit|clean|bonus|edition|special|limited|expanded|director''s|uncut|final|ultimate|platinum|gold|anniversary|collector''s|standard|enhanced|super|mega|ultra|plus|pro|premium|complete|definitive|classic|original|alternate|alternative|unreleased|rare|exclusive).*?$',
-        '(?i)\s*:\s*(?:\d{4}|\d{2})\s*(?:remaster|edition|version|mix|cut|release).*?$'
-    ];
-
-    pattern TEXT;
-    result_text TEXT := name_text;
-BEGIN
-    -- Return early if input is null or empty
-    IF name_text IS NULL OR trim(name_text) = '' THEN
-        RETURN name_text;
-    END IF;
-
-    -- Remove discriminant patterns to get base name
-    FOREACH pattern IN ARRAY cleanup_patterns
-    LOOP
-        result_text := regexp_replace(result_text, pattern, ' ', 'g');
-    END LOOP;
-
-    -- Clean up extra whitespace and normalize
-    result_text := regexp_replace(trim(result_text), '\s+', ' ', 'g');
-
-    -- Remove trailing punctuation that might be left after removal
-    result_text := regexp_replace(result_text, '[,;:\-–—]\s*$', '', 'g');
-    result_text := trim(result_text);
-
-    -- Ensure we don't return an empty string
-    IF length(result_text) = 0 THEN
-        RETURN name_text;
-    END IF;
-
-    RETURN result_text;
-END;
-$$ LANGUAGE plpgsql IMMUTABLE;
-
--- Enhanced function to extract discriminant specifically for editions and versions with case-insensitive matching
-CREATE OR REPLACE FUNCTION extract_edition_discriminant(name_text TEXT) RETURNS TEXT AS $$
-DECLARE
-    -- Focused patterns for edition/version extraction with case-insensitive flags
-    edition_patterns TEXT[] := ARRAY[
-        -- Edition patterns
-        '(?i)\(([^)]*edition[^)]*)\)',
-        '(?i)\[([^]]*edition[^]]*)\]',
-        '(?i)\{([^}]*edition[^}]*)\}',
-        '(?i)[-–—]\s*([^-–—]*edition[^-–—]*)$',
-        '(?i):\s*([^:]*edition[^:]*)$',
-
-        -- Version patterns
-        '(?i)\(([^)]*version[^)]*)\)',
-        '(?i)\[([^]]*version[^]]*)\]',
-        '(?i)\{([^}]*version[^}]*)\}',
-        '(?i)[-–—]\s*([^-–—]*version[^-–—]*)$',
-        '(?i):\s*([^:]*version[^:]*)$',
-
-        -- Remaster patterns
-        '(?i)\(([^)]*remaster[^)]*)\)',
-        '(?i)\[([^]]*remaster[^]]*)\]',
-        '(?i)\{([^}]*remaster[^}]*)\}',
-        '(?i)[-–—]\s*([^-–—]*remaster[^-–—]*)$',
-        '(?i):\s*([^:]*remaster[^:]*)$',
-
-        -- Year-based patterns
-        '(?i)\(([^)]*(?:\d{4}|\d{2})\s*(?:remaster|edition|version|mix|cut|release)[^)]*)\)',
-        '(?i)\[([^]]*(?:\d{4}|\d{2})\s*(?:remaster|edition|version|mix|cut|release)[^]]*)\]',
-        '(?i)\{([^}]*(?:\d{4}|\d{2})\s*(?:remaster|edition|version|mix|cut|release)[^}]*)\}'
-    ];
-
-    pattern TEXT;
-    match_result TEXT;
-BEGIN
-    -- Return early if input is null or empty
-    IF name_text IS NULL OR trim(name_text) = '' THEN
-        RETURN NULL;
-    END IF;
-
-    -- Try edition-specific patterns first
-    FOREACH pattern IN ARRAY edition_patterns
-    LOOP
-        SELECT substring(name_text FROM pattern) INTO match_result;
-        IF match_result IS NOT NULL AND length(trim(match_result)) > 0 THEN
-            match_result := trim(match_result);
-            match_result := regexp_replace(match_result, '^[^\w]+|[^\w]+$', '', 'g');
-            IF length(trim(match_result)) > 0 THEN
-                RETURN match_result;
-            END IF;
-        END IF;
-    END LOOP;
-
-    RETURN NULL;
-END;
-$$ LANGUAGE plpgsql IMMUTABLE;
-
--- Update existing records with newly extracted discriminants (case-insensitive)
-UPDATE recordings
-SET discriminant = extract_discriminant(name)
-WHERE discriminant IS NULL
-  AND extract_discriminant(name) IS NOT NULL;
-
-UPDATE releases
-SET discriminant = extract_discriminant(name)
-WHERE discriminant IS NULL
-  AND extract_discriminant(name) IS NOT NULL;
-
-UPDATE plays
-SET track_discriminant = extract_discriminant(track_name)
-WHERE track_discriminant IS NULL
-  AND extract_discriminant(track_name) IS NOT NULL;
-
-UPDATE plays
-SET release_discriminant = extract_discriminant(release_name)
-WHERE release_discriminant IS NULL
-  AND release_name IS NOT NULL
-  AND extract_discriminant(release_name) IS NOT NULL;
-
--- Update comments for the enhanced functions
-COMMENT ON FUNCTION extract_discriminant IS 'Enhanced case-insensitive discriminant extraction supporting comprehensive edition/version patterns including parentheses, brackets, braces, dashes, and colons';
-COMMENT ON FUNCTION get_base_name IS 'Enhanced case-insensitive base name extraction removing comprehensive discriminant patterns to enable proper grouping';
-COMMENT ON FUNCTION extract_edition_discriminant IS 'Specialized case-insensitive function for extracting edition and version discriminants with focused patterns';
-
--- Refresh materialized views to reflect the case-insensitive improvements
-REFRESH MATERIALIZED VIEW mv_release_play_counts;
-REFRESH MATERIALIZED VIEW mv_recording_play_counts;
-
--- Update discriminant analysis view to include case-insensitive results
-DROP VIEW IF EXISTS discriminant_analysis;
-CREATE OR REPLACE VIEW discriminant_analysis AS
-SELECT
-    'recordings' as table_name,
-    name as original_name,
-    discriminant,
-    get_base_name(name) as base_name,
-    extract_discriminant(name) as extracted_discriminant,
-    extract_edition_discriminant(name) as edition_discriminant
-FROM recordings
-WHERE name IS NOT NULL
-UNION ALL
-SELECT
-    'releases' as table_name,
-    name as original_name,
-    discriminant,
-    get_base_name(name) as base_name,
-    extract_discriminant(name) as extracted_discriminant,
-    extract_edition_discriminant(name) as edition_discriminant
-FROM releases
-WHERE name IS NOT NULL;
-
-COMMENT ON VIEW discriminant_analysis IS 'Analysis view showing case-insensitive discriminant extraction results for quality assessment and debugging';