Erq is an alternative query language for relational databases. It is designed to be easy to learn and use, while being as expressive and powerful as SQL.

I originally used SQLite to analyze my own kanji database. However, the verbose and cumbersome syntax of SQL prevented me from doing ad hoc analysis quickly. So I created Erq, which simplifies the syntax of SQL.

See Quickstart for installation instructions, CLI examples, and a minimal .erq script.

Erq CLI works like SQLite CLI. You can use it to query SQLite databases.

Basically, you can use Erq CLI like this:

erq your_database.db <your_query.erq >your_output.txt

Or you can use it interactively:

erq your_database.db

Erq writes debug information to stderr. You can opt into specific groups by setting the ERQ_DEBUG environment variable to a space- or comma-separated list of categories.

Available categories include:

• sql – SQL statements generated during script execution or imports
• script – script-specific helpers such as loop conditions
• import – SQL emitted when loading external data sources
• lifecycle – parent/child process lifecycle events
• ipc – IPC messages (e.g. completer results)
• stack – stack traces for caught exceptions
• preprocess – interpolation details while expanding ERQ expressions
• general – legacy bucket that enables all historical debug output

You can combine categories (e.g. ERQ_DEBUG=sql,stack) or use 1, true, or all to enable every debug message.

During an interactive session you can inspect or change the same setting with the .debug dot command, for example .debug sql stack to enable only SQL and stack trace logging or .debug off to disable it again.

Inside the interactive CLI you can enter dot-prefixed utility commands. Use .help to list them or .help COMMAND to see detailed usage. Tab completion now suggests dot command names as well as arguments such as .format modes or .debug categories.

See Syntax Comparison with SQL.