Inspect Parquet geospatial statistics

Reports the GeospatialStatistics that writers record for GEOMETRY and GEOGRAPHY columns: each column chunk's bounding box and the geometry types it holds. One row per row group and geospatial column, whether or not the chunk carries them, so a reader can choose row groups by box before reading any data. qio's own writer records them for every geospatial column.

Usage

geospatial_statistics(x, ...)

# S3 method for class 'qio_parquet_file'
geospatial_statistics(x, ...)

Arguments

x

A qio_parquet_file object.

...

Reserved for future use.

Value

A data frame with one row per row group and geospatial column, ordered by row group, then column, with the columns:

row_group

1-based row-group ID, as row_groups() reports it.

column

1-based physical column index, as schema() reports it.

path

Complete dotted column path; see schema().

xmin, xmax, ymin, ymax

The bounding box, or NA when none was recorded.

zmin, zmax, mmin, mmax

The Z and M ranges, or NA when the geometries have no such dimension or none was recorded.

geometry_types

A list column of the ISO WKB geometry type codes present (1 for a point, 1001 for a point with Z, ...), or NULL when none were recorded, which Parquet reads as unknown.

Details

Like column_statistics(), the bounds are claims made by whoever wrote the file. For GEOGRAPHY a box can be smaller than the geometries' extent on the sphere, and an xmin greater than xmax means the box wraps the antimeridian.

See also

column_statistics(), parquet_schema() for writing geospatial columns.

Examples

path <- tempfile(fileext = ".parquet")
point <- function(x, y) {
  c(as.raw(1), writeBin(1L, raw(), size = 4, endian = "little"),
    writeBin(c(x, y), raw(), size = 8, endian = "little"))
}
df <- data.frame(id = 1:2)
df$geom <- list(point(1, 2), point(3, 4))
write_parquet(df, path, schema = parquet_schema(geom = "GEOMETRY"))
pf <- open_parquet(path)
geospatial_statistics(pf)
#>   row_group column path xmin xmax ymin ymax zmin zmax mmin mmax geometry_types
#> 1         1      2 geom    1    3    2    4   NA   NA   NA   NA              1
close_parquet(pf)