Drop a CSV, get an inline data structure you can paste straight into a script. Pick pandas, Polars, R, PHP, or Ruby — numbers stay numeric, strings are escaped, and empty cells become the right null for each language.
Pick a language, paste or upload your CSV, and copy the generated literal. Everything runs in your browser — your data never touches a server.
Paste rows in directly or drop a file. The parser reads your headers and infers which columns are numeric, so the output uses the right literal types.
Switch between pandas, Polars, R, PHP, and Ruby. pandas and Polars emit a DataFrame from a dict of columns; R builds a data.frame; PHP and Ruby produce a list of rows.
Grab the generated code and drop it into your notebook or script. No CSV file to ship, no loader to write — the data is inlined and ready to run.
Inlining a small dataset as source removes a moving part. No file path to get wrong, no read_csv call to configure, no encoding surprises — just a literal that runs the same everywhere.
Paste a known dataset straight into a unit test or a notebook cell. No tmp files, no fixtures directory, no I/O that can fail in CI.
Plain numeric cells are emitted as bare numbers so pandas, Polars, and R see a numeric column — not text you have to cast after loading.
Quotes, backslashes, tabs, and newlines are escaped per language. PHP keys and values use single-quote rules; pandas, R, and Ruby use double-quote rules.
Generation happens in your browser with client-side JavaScript. Your rows are never uploaded, logged, or stored anywhere.