The local file system is the simplest and most straightforward way for GlareDB to access data—it reads files directly from the machine’s file system, without any additional setup or configuration.
The local file system is only available when using the CLI or Python bindings, as the WebAssembly bindings do not have access to the local file system.
You can query files using either relative or absolute paths.
SELECT * FROM read_csv('./cities.csv');
This reads the cities.csv
file relative to the current working directory where
GlareDB is running.
SELECT * FROM read_csv('/home/user/data/cities.csv');
This reads the file directly from the specified location in the file system.
You can also query files directly by specifying the file path in the FROM clause:
SELECT * FROM './cities.csv';
SELECT * FROM '/home/user/data/cities.parquet';
All supported file formats, such as CSV and Parquet, can be read from the local
file system using their respective functions (e.g. read_csv
).