Re-cluster table data

Rewrites table data files to match the current clustering order. Use this after setting clustering on a table that already contains data, or after many appends have fragmented the sort order.

DuckLake automatically sorts data during compaction based on the table's sort order (SET SORTED BY), so this function compacts files to re-sort.

Usage

db_recluster(schema = "main", table, max_files = NULL)

Arguments

schema

Schema name (default "main")

table

Table name

max_files

Maximum number of compaction operations per call. Lower values use less memory. Default NULL processes all files.

Value

Invisibly returns the qualified table name

See also

db_set_clustering() to configure clustering order, db_compact()

Examples

if (FALSE) { # \dontrun{
db_connect()

# Set clustering then recluster existing data
db_set_clustering(table = "events", columns = c("event_date"))
db_recluster(table = "events")

# Recluster with memory limit
db_recluster(table = "events", max_files = 100)
} # }