dbplyr 1.2.0 Now on CRAN

We are very excited to announce that dbplyr 1.2.0 is now available on CRAN! The full announcement was posted in tidyverse.org. Here are the highlights:

  • New custom translation for Microsoft Access and Teradata

  • Amazon Redshift connections now point to the PostgreSQL translation.

  • Adds support for two new database R packages. These new packages are fully DBI-compliant and tested with DBItest. We recommend to use these instead of older packages:
    1. RMariaDB, use in favor of RMySQL
    2. RPostgres, use in favor of RPostgreSQL
  • ROracle connections now point to the Oracle translation.
  • Cache query results using copy_to. This feature is very useful because after cached, the results of a long running query can be iteratively analyzed without having to wait for the same base query to run over and over.

  • stringr functions: str_length(), str_to_upper(), str_to_lower(), str_replace_all(), str_detect(), and str_trim() are now supported.

  • in_schema() should now work in more places, particularly in copy_to()

  • For those who need to extract the resulting SQL query from dplyr code, there is now a new function called remote_query(). Unlike show_query(), remote_query() returns a character object, and unlike sql_render(), the new function automatically detects the type of database attached to the given dplyr code and translates accordingly. There are four additional functions that provide a standard API to get metadata about a remote tbl, these are: remote_name(), remote_con(), remote_src(), and remote_query_plan().

  • Adds support for temporary tables in Microsoft SQL Server. Additionally, the issue of certain operators working only in mutate() or only in filter() has been resolved.

  • Two new functions for developers: sql_aggregate() and win_aggregate(). They are for generating SQL and windowed SQL functions for aggregates.