Show Parquet physical type mappings

Lists every Parquet physical type and the R storage type qio uses for it when the column carries no logical annotation. Missing values indicate unsupported mappings.

Usage

parquet_type_mapping()

Value

A data frame with one row per Parquet physical type and the columns:

physical_type

Parquet physical type, spelled as schema() and read_plan() report it.

r_type

R type produced with no logical annotation, spelled as read_plan() reports it. NA when the type cannot be read.

written_from

R input that infers this physical type, or NA when write_parquet() cannot produce it.

Details

These are physical fallbacks, and most real columns are annotated, so this table is not a prediction of what a given file will produce. Use read_plan() for that: it resolves the annotation, the int64, time, and tz options, and reports the R type each column will actually materialize as.

Two entries are worth reading carefully. BYTE_ARRAY and FIXED_LEN_BYTE_ARRAY are bytes here, returned as a list of raw vectors, because bytes are only text when the file says so; a STRING, ENUM, or JSON annotation is what makes a BYTE_ARRAY character. INT96 is a deprecated physical type used only for timestamps, so it is read as POSIXct with no annotation involved.

See also

read_parquet(), write_parquet(), schema()

Examples

parquet_type_mapping()
#>          physical_type  r_type              written_from
#> 1              BOOLEAN logical                   logical
#> 2                INT32 integer                   integer
#> 3                INT64  double numeric (explicit schema)
#> 4                INT96 POSIXct                      <NA>
#> 5                FLOAT  double numeric (explicit schema)
#> 6               DOUBLE  double                    double
#> 7           BYTE_ARRAY    list       character or factor
#> 8 FIXED_LEN_BYTE_ARRAY    list                      <NA>