About the ClickHouse compiler
Column-oriented analytics SQL, translated to DuckDB and run in your browser. The WASM build has no timezone database, so now() is evaluated as a local timestamp. Run ClickHouse code in your browser with no install and no signup — the editor, the input and the output share one screen. ClickHouse compiles and runs on your own device through WebAssembly, so once the page has loaded it keeps working without a network connection.
Press Run to execute the program, edit it and press Run again, or paste your own code over it. Everything you type stays in the page; the editor caches your work locally so a reload does not lose it.
A ClickHouse example
This is the program the ClickHouse editor opens with — run it as it is, then change it.
-- ClickHouse SQL — translated to DuckDB and run in your browser.
-- toStartOfHour() and now() - INTERVAL are translated for the WASM build;
-- now() is a plain timestamp there because the build has no timezone data.
CREATE TABLE analytics (
timestamp DateTime,
user_id UInt32,
event String,
properties String
) ENGINE = MergeTree()
ORDER BY (timestamp, user_id);
INSERT INTO analytics VALUES
('2024-01-01 09:15:00', 1, 'page_view', '{"page": "/home"}'),
('2024-01-01 09:30:00', 2, 'click', '{"button": "signup"}'),
('2024-01-01 10:05:00', 1, 'page_view', '{"page": "/pricing"}');
-- Aggregation query
SELECT
toStartOfHour(timestamp) as hour,
event,
count() as events,
uniq(user_id) as unique_users
FROM analytics
GROUP BY hour, event
ORDER BY hour;
-- Window functions
The editor opens with the full program.
More Databases compilers
Browse all 109 languages — the same editor runs each of them, in the browser or on a shared runner.