If you are viewing this file on CRAN, please check the latest news on GitHub where the formatting is also better.
Fixed a bug (Issue #6) where x[y] did not return the
reference result when x contains a WHERE
clause.
Evaluation of special functions (e.g., %in%) now
RDBMS specific.
.(name = symbol) to
the on arguement in the extract ([) method.
Thanks to @SugarRayLua for reporting the
issue.as_cte was putting the CTEs in the wrong order
when the input already included a CTE. This caused DuckDB >= 1.4.0 to
bork.RDBMS-specific functions are now generic and existing code has been moved to methods. Methods are provided for RMariaDB, RPostgres, RSQLite, duckdb, and Microsoft SQL Server. The default methods rely on DBI alone and should provide basic functionality where DBI is fully implemented.
data.table-style keys are partially implemented. A
key can be specified when creating a dbi.table.
dbi.tables created by dbi.catalog (and
dbi.attach) use the table’s primary key as the
default.
The column naming scheme for relational merge (i.e.,
merge(x)) has been changed.
dbi.tables in schemas are now implemented as active
bindings so that data.table’s setcolorder does
not change their column order.
In reference.test, when the output of
all.equal is not TRUE and
verbose = TRUE, the format of the message now
matches print(all.equal(...)). Thanks to @MichaelChirico for
the report and fix.
Need a high-speed mirror for your open-source project?
Contact our mirror admin team at info@clientvps.com.
This archive is provided as a free public service to the community.
Proudly supported by infrastructure from VPSPulse , RxServers , BuyNumber , UnitVPS , OffshoreName and secure payment technology by ArionPay.