Load an existing sealed snapshot

Reopens a snapshot that already exists in a DuckDB file. This is the durable research workflow counterpart to ledgr_snapshot_from_df(): create and seal a snapshot once, then load it by path and snapshot id in later R sessions.

Usage

ledgr_snapshot_load(db_path, snapshot_id = NULL, verify = FALSE)

Arguments

db_path

Path to an existing DuckDB file.

snapshot_id

Snapshot id to load. The snapshot must exist and be SEALED.

verify

Logical scalar. If TRUE, recompute the snapshot hash and compare it with the stored hash before returning.

Value

A lazy ledgr_snapshot object.

Details

ledgr_snapshot_load() never creates or overwrites snapshots. It only returns a handle to an existing sealed snapshot. Closing the returned object releases its DuckDB connection; it does not delete the database file.

Articles

Durable experiment stores: vignette("experiment-store", package = "ledgr") system.file("doc", "experiment-store.html", package = "ledgr")

Examples

db_path <- tempfile(fileext = ".duckdb")
bars <- data.frame(
  ts_utc = as.POSIXct("2020-01-01", tz = "UTC") + 86400 * 0:2,
  instrument_id = "AAA",
  open = c(100, 101, 102),
  high = c(101, 102, 103),
  low = c(99, 100, 101),
  close = c(100, 101, 102),
  volume = 1000
)
snapshot <- ledgr_snapshot_from_df(bars, db_path = db_path)
snapshot_id <- snapshot$snapshot_id
ledgr_snapshot_close(snapshot)

snapshot <- ledgr_snapshot_load(db_path, snapshot_id, verify = TRUE)
ledgr_snapshot_info(snapshot)
#> # A tibble: 1 × 11
#>   snapshot_id        status created_at_utc sealed_at_utc snapshot_hash bar_count
#>   <chr>              <chr>  <chr>          <chr>         <chr>             <int>
#> 1 snapshot_20260515… SEALED 2026-05-15T16… 2026-05-15T1… 4d7d3b7dda80…         3
#> # ℹ 5 more variables: instrument_count <int>, start_date <chr>, end_date <chr>,
#> #   meta_json <chr>, error_msg <chr>
ledgr_snapshot_close(snapshot)