Skip to contents
rladybugdb 0.2.0
Breaking changes
- Python/reticulate backend removed.
ladybugdb_install() is now deprecated and does nothing. Remove any calls to it from your scripts.
New features
-
Native Rcpp C++ bindings to the LadybugDB C library (
liblbug v0.15.2). No Python runtime is required — the C library ships with the package.
-
ladybugdb_version() returns the version of the bundled C library.
- Parameterised queries via the
parameters list argument in lb_execute() and lb_query(). Use $name placeholders in Cypher to avoid injection bugs.
-
as_arrow_table() generic for zero-copy Arrow output from an lb_result.
-
as_tbl_graph() converts query results directly to a tidygraph::tbl_graph.
Improvements
-
LIST / ARRAY columns are now returned as proper R list columns (one R list element per row) rather than collapsed strings.
-
DATE columns map to Date; TIMESTAMP columns map to POSIXct; INT64 values map to double (exact up to 2^53, covers all graph IDs).
-
lb_copy_from_df() and lb_copy_from_csv() use LadybugDB’s native bulk loader, which is orders of magnitude faster than row-by-row CREATE.
