Read the tidied Parquet schema as a tibble: one row per leaf column with its Parquet physical type,
converted/logical types, repetition, mapped R type, and (for geometry columns) the encoding and geometry types
from the geo metadata. For a high-level summary instead of the full table, see gpq_schema_info().
Value
A tibble::tibble() with columns id, name, parquet_type, converted_type, logical_type, repetition,
r_type, and info.
Examples
f <- system.file("extdata/atlanta/atlanta.parquet", package = "gdalvector")
gpq_schema(f)
#> # A tibble: 6 × 8
#> id name parquet_type converted_type logical_type repetition r_type info
#> <int> <chr> <chr> <chr> <chr> <chr> <chr> <chr>
#> 1 1 OGC_FID INT64 INT_64 NA OPTIONAL double NA
#> 2 2 source… BYTE_ARRAY UTF8 NA OPTIONAL chara… NA
#> 3 3 geoid BYTE_ARRAY UTF8 NA OPTIONAL chara… NA
#> 4 4 state_… BYTE_ARRAY UTF8 NA OPTIONAL chara… NA
#> 5 5 county… BYTE_ARRAY UTF8 NA OPTIONAL chara… NA
#> 6 6 geom BYTE_ARRAY NA UNKNOWN() OPTIONAL blob WKB …