CSV

Read the files

read_csv(glob_paths)
read_csv(glob_paths, delimiter)
read_csv(glob_paths, delimiter, infer_records)

Beacon infers the schema from the file contents. The first row must be a header row.

• delimiter: the field separator, one character (default: ,)
• infer_records: the number of rows that Beacon samples for the column types (default: 128000)

SELECT * FROM read_csv('metadata/*.csv')

-- Tab-separated, sample 500 rows for type inference
SELECT * FROM read_csv(['data/*.tsv'], '\t', 500)

Inspect the schema

Check the columns and the types before you write a query:

SELECT * FROM read_csv('stations/*.csv') LIMIT 0;

Inspect a schema compares the _schema functions, SUMMARIZE, DESCRIBE and LIMIT 0, and says what each one costs.

Format details

• The first row must be a header row with the column names.
• A file must use UTF-8 encoding.
• Beacon infers the schema from the file contents.

As an external table

CREATE EXTERNAL TABLE station_metadata
STORED AS CSV
LOCATION 'metadata/stations/'

See Create External Tables for the full DDL. See Data Sources for the full read model.

OPTIONS

STORED AS CSV reads two keys. They are the two optional arguments of read_csv:

Option	Type	Default	Description
delimiter	One ASCII character, or an escape	,	The field separator. The escapes are \t, \n, \r, \0 and \\. A value of two or more characters is an error.
infer_records	Whole number	1000	The number of rows that Beacon samples for the column types. read_csv samples 128000 rows instead.

-- A tab-separated collection, with a deeper sample for the column types
CREATE EXTERNAL TABLE station_metadata
STORED AS CSV
LOCATION 'metadata/stations/'
OPTIONS ('delimiter' '\t', 'infer_records' '5000')

See OPTIONS for the rules that hold for every key.