Documentation
Everything needed to install UNSeQueL, write `.unsequel` queries, run them from the CLI or VS Code, and understand the semantics behind the results.
Getting started
Install the compiler, run your first self-contained query, format it, and open the REPL.
Language guide
Every stage — from, joins, filter, grouping, windows, projections, set operators — with syntax and examples.
Expressions and functions
Operators, literals, quoted identifiers, case, aggregates, window functions, and scalar functions.
CLI reference
check, compile, preview, run, fmt, and repl with every flag and directive.
Semantics and data rules
NULLs, CSV type inference, sort order, ambiguous columns, engine boundaries, and error behavior.
Worked examples
The five analyst queries that ship with the repository, with their data declarations and results.
VS Code
Syntax highlighting, Run Selection/File, and the interactive REPL from the editor.
Python prototype
The early .usql ordered surface and .pusql pipeline surface, for comparison and learning.
Comparison and scope
How UNSeQueL relates to DuckDB FROM-first SQL, PRQL, Malloy, Kusto, and pipe syntax — and what it does not do.
Roadmap
Prioritised language and tooling gaps, including window frames and the deferred second SQL target.