Skip to contents

Save a Knowledge Graph to disk

Source: R/causal_kg_io.R
causal_kg_save.Rd

Writes an edaphos_causal_kg to an .rds file. The KG is serialised through its tidy edge list plus metadata (version, timestamp, R and package versions) rather than through the raw igraph object, so the file is:

Usage

causal_kg_save(kg, path)

Arguments

kg

An edaphos_causal_kg.

path

Path to the target .rds file. The parent directory is created if it does not exist.

Value

Invisibly returns path (for pipelining).

Details

  • Portable across igraph versions (the C-level pointer layout that saveRDS(igraph_object) would capture is not written).

  • Deterministic (no hash-randomised attributes), so two saves of the same KG produce byte-identical files.

  • Small — only the edge list and node names are written, not igraph's internal indices.

Load with causal_kg_load().

See also

causal_kg_load(), causal_kg_to_turtle() for the human-readable RDF variant.

Examples

# \donttest{
  kg <- causal_kg_new()
  kg <- causal_kg_add_edge(kg, "precipitation", "soc",
                            source = "Jenny 1941",
                            confidence = 0.9)
  f <- tempfile(fileext = ".rds")
  causal_kg_save(kg, f)
  kg2 <- causal_kg_load(f)
  identical(causal_kg_edges(kg), causal_kg_edges(kg2))
#> [1] TRUE
# }