next.orly.dev

Commit Graph

Author	SHA1	Message	Date
woikos	34ab56815e	Add database IPC split mode and blob storage abstraction (v0.53.0) Major refactoring to support running the database as a separate gRPC service: - Add gRPC database server (cmd/orly-db/) and launcher (cmd/orly-launcher/) - Add proto definitions for all database operations (proto/orlydb/v1/) - Add gRPC client implementing database.Database interface - Abstract Blossom blob storage into database interface (9 methods) - Remove BBolt database backend (pkg/bbolt/ deleted) - Hide Blossom tab in UI when blob storage unavailable - Update documentation for new architecture Database backends: badger (default), neo4j, wasmdb, grpc Files modified: - cmd/orly-db/: New gRPC database server - cmd/orly-launcher/: Process supervisor for split mode - proto/orlydb/v1/: Protocol buffer definitions - pkg/database/grpc/: gRPC client implementation - pkg/database/blob.go: Badger blob storage implementation - pkg/database/interface.go: Added blob storage methods - pkg/database/types.go: Added BlobMetadata, BlobDescriptor types - pkg/database/factory.go: Removed BBolt, added gRPC backend - pkg/neo4j/blob.go: Blob storage stubs - pkg/wasmdb/blob.go: Blob storage stubs - pkg/blossom/: Refactored to use database.Database interface - app/config/config.go: Removed BBolt config, added gRPC settings - CLAUDE.md: Updated documentation Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>	4 months ago
woikos	9fed1261ad	Add BBolt database backend for HDD-optimized archival relays (v0.48.0) - BBolt B+tree backend with sequential access patterns for spinning disks - Write batching (5000 events / 128MB / 30s flush) to reduce disk thrashing - Adjacency list storage for graph data (one key per vertex, not per edge) - Bloom filter for fast negative edge existence checks (~12MB for 10M edges) - No query cache (saves RAM, B+tree reads are fast enough on HDD) - Migration tool: orly migrate --from badger --to bbolt - Configuration: ORLY_BBOLT_* environment variables Files modified: - app/config/config.go: Added BBolt configuration options - main.go: Added migrate subcommand and BBolt config wiring - pkg/database/factory.go: Added BBolt factory registration - pkg/bbolt/*: New BBolt database backend implementation 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>	4 months ago
mleku	f22bf3f388	Add Neo4j memory tuning config and query result limits (v0.43.0) - Add Neo4j driver config options for memory management: - ORLY_NEO4J_MAX_CONN_POOL (default: 25) - connection pool size - ORLY_NEO4J_FETCH_SIZE (default: 1000) - records per batch - ORLY_NEO4J_MAX_TX_RETRY_SEC (default: 30) - transaction retry timeout - ORLY_NEO4J_QUERY_RESULT_LIMIT (default: 10000) - max results per query - Apply driver settings when creating Neo4j connection (pool size, fetch size, retry time) - Enforce query result limit as safety cap on all Cypher queries - Fix QueryForSerials and QueryForIds to preserve LIMIT clauses - Add comprehensive memory tuning documentation with sizing guidelines - Add NIP-46 signer-based authentication for bunker connections - Update go.mod with new dependencies Files modified: - app/config/config.go: Add Neo4j driver tuning config vars - main.go: Pass new config values to database factory - pkg/database/factory.go: Add Neo4j tuning fields to DatabaseConfig - pkg/database/factory_wasm.go: Mirror factory.go changes for WASM - pkg/neo4j/neo4j.go: Apply driver config, add getter methods - pkg/neo4j/query-events.go: Enforce query result limit, fix LIMIT preservation - docs/NEO4J_BACKEND.md: Add Memory Tuning section, update Docker example - CLAUDE.md: Add Neo4j memory tuning quick reference - app/handle-req.go: NIP-46 signer authentication - app/publisher.go: HasActiveNIP46Signer check - pkg/protocol/publish/publisher.go: NIP46SignerChecker interface - go.mod: Add dependencies 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>	5 months ago
mleku	c1bd05fb04	Adjust ACL behavior for "none" mode and make query cache optional This commit allows skipping authentication, permission checks, and certain filters (e.g., deletions, expirations) when the ACL mode is set to "none" (open relay mode). It also introduces a configuration option to disable query caching to reduce memory usage. These changes improve operational flexibility for open relay setups and resource-constrained environments.	6 months ago
mleku	290fcbf8f0	remove outdated configuration items for obsolete tail packing optimization	6 months ago
mleku	54ead81791	merge authors/nostruser in neo4j, add compact pubkey/e/p serial refs	6 months ago
mleku	59247400dc	Remove Dgraph support from the codebase. Dgraph-related functionality, configuration, and benchmarks have been removed from the project. This streamlines the codebase to focus on supported backends, specifically eliminating Dgraph references in favor of Neo4j and other implementations. Version bumped to reflect the changes.	6 months ago
mleku	6bd56a30c9	remove dgraph completely	6 months ago
mleku	0a61f274d5	implement wasm/js specific database engine	6 months ago
mleku	016e97925a	Refactor database configuration to use centralized struct Replaced individual environment variable access with a unified `DatabaseConfig` struct for all database backends. This centralizes configuration management, reduces redundant code, and ensures all options are documented in `app/config/config.go`. Backward compatibility is maintained with default values and retained constructors.	6 months ago
mleku	86481a42e8	initial draft of neo4j database driver	6 months ago
mleku	95bcf85ad7	optimizing badger cache, won a 10-15% improvement in most benchmarks	6 months ago

12 Commits (34ab56815e32aeaf166bd69acbec52ec1b1d9170)