- Increase MySQL pool max from 1 to 3 to avoid contention across tabs
- Merge get_columns serial queries into a single LEFT JOIN
- Clone pool handles before health checks in refresh_connections to reduce lock contention
- Skip COUNT(*) when result rows are fewer than the page size
- Add 150ms debounce on async completion path to reduce redundant Tauri IPC calls during rapid typing
- Parallelize schema table fetching with max concurrency of 5
- Separate schema list cache from table completion cache for longer-lived reuse
- Return keyword completions synchronously when no backend data is needed
- Lower snippet base boost for label-only matches so keywords rank above snippets when user types past snippet prefix
Execute COUNT query asynchronously after main query to show total
matching rows. Strip user-supplied LIMIT/OFFSET from count SQL so
the total reflects all rows, not just the limited page.
Add a `connect_timeout_secs` field to ConnectionConfig that allows users
to configure the database connection timeout per connection (1-300s,
default 5s). Previously all connections used a hardcoded 5-second timeout.
- Add connect_timeout_secs field with serde default and clamp logic
- Add parse_connect_timeout_with_fallback to honor URL params over config
- Pass user-configured timeout to all drivers (MySQL, Postgres, Redis,
MongoDB, ClickHouse, SQL Server, Elasticsearch) and TCP probe
- Add UI input in connection dialog SSH tab
- Add i18n keys for en/zh-CN/es
- frontend: use e.message instead of String(e) in toErrorResult to avoid
"Error: " prefix for HTTP-mode errors, matching connectionErrorMessage
- frontend: apply same fix to MongoDocBrowser error display
- backend(postgres): use pg_error_to_string across all query execution
paths (execute, execute_batch, copy_out, copy_in) to extract DbError
details instead of generic transport-layer errors
Other databases (MySQL, SQL Server, SQLite, DuckDB) already include
server-level error details in their native Error::to_string().
Add queryTimeoutSecs to editor settings with UI control in the settings
dialog. 0 means no timeout. Applied to all query execution paths including
sidebar table data loading and EXPLAIN queries.
Closes#441.
Users can now create, edit, and delete SQL snippet templates from the settings dialog (new "Snippets" tab). Each template has a label, trigger prefix, and SQL body. Built-in defaults are modifiable and can be restored. Templates persist across sessions via the existing settings storage.
t8y2