- Windows: update to mariadb-connector-c 3.2.5. This new version still uses OpenSSL for R <= 4.1 but R 4.2 (ucrt) it switches to the new native Windows schannel TLS.
- Windows: add support for UCRT builds
- Windows: update to libmariadbclient 3.1.11
- Fixed improper escaping with dbQuoteString and dbQuoteLiteral, which can lead to SQL injection (Found by Nikolay Simakov and patched by Andrew Kane, CVE-2020-10380).
- Fix new warning for CRAN
- Small tweaks for configure script
- MacOS: CRAN binaries have been updated to mariadb-connector-c to 3.1.6
- Fix rcheck problems requested by CRAN
- Windows: Update mariadb-connector-c to v2.3.7 + OpenSSL 1.1.1
- MacOS: Update mariadb-connector-c to 3.0.8
- Fix for MySQL 8 (issue #223)
- Windows: Update mariadbclient to v 2.3.5
- Windows: Update mariadbclient to v 2.3.4
- Fix headers and linker flags for mariadb-connector-c 3.0 compatibility
- Remove some redundant include headers
- Add cleanup script (requested by CRAN)
- Fix PROTECT() bugs from rchk
- Windows: update libmariadbclient to v2.3.3 (fixes a timeout bug)
- Fix for OSX Mavericks
-
Window: update libmariadbclient to v2.3.2
-
OSX: extract autobrew script
-
OSX binary packages for OSX now use mariadb-connector-c instead of mysql-connector-c
-
CMD check fix for DBI 0.4
- Replace #include S.h with R.h as required by BDR
- Fix configure script for OSX
-
Refactor configure script
-
Fix for setInternet2 in R-devel on Windows
-
Fix dbWriteTable bug with MySQL-based Infobright database
-
Use mariadb_config if available
-
Properly import used utils functions
-
Fix the RPostgreSQL conflict
-
Remove the workaround for the bug in TSMySQL
-
Update URL to MariaDB client in DESCRIPTION
-
Backport fixes from Rcpp branch.
-
Configure script now uses
mysql_config
if available. -
MariaDB windows libraries have been updated to 2.1.0
-
Add SSL support to MariaDB windows libraries.
-
Fix configure script for OSX 10.6 Snow Leopard
-
Issue in
dbWriteTable()
with temporary files on Windows fixed.
-
New maintainer: Jeroen Ooms
-
Internal changes to support static linking on Windows; this means that windows a binary is now available on CRAN.
-
The following internal functions are no longer exported: please use the corresponding DBI generic instead:
mysqlInitDriver
,mysqlCloseDriver
,mysqlDescribeDriver
,mysqlDriverInfo
,mysqlNewConnection
,mysqlCloneConnection
,mysqlDescribeConnection
,mysqlConnectionInfo
,mysqlCloseConnection
,mysqlExecStatement
,mysqlQuickSQL
,mysqlDBApply
,mysqlFetch
,mysqlResultInfo
,mysqlDescribeResult
,mysqlDescribeFields
,mysqlCloseResult
,mysqlImportFile
,mysqlReadTable
,mysqlWriteTable
,mysqlEscapeStrings
,mysqlDataType
,safe.write
. -
RMySQL gains transaction support with
dbBegin()
,dbCommit()
, anddbRollback()
, but note that MySQL does not allow data definition language statements to be rolled back. -
The MySQLObject base class has been removed - there is no real shared behaviour for MySQLDriver, MySQLConnection and MySQLResult so this simplifies the code
-
Implemented methods for
dbIsValid()
; please use these instead ofisIdCurrent()
. -
Implement
dbFetch()
method; please use this in preference tofetch()
.dbFetch()
now returns a 0-row data frame (instead of an 0-col data frame) if there are no results. -
Methods no longer automatically close open result sets. This was implemented inconsistently in a handful of places.
-
dbBuildTableDefinition()
has been renamed tomysqlBuildTableDefinition()
. -
dbWriteTable()
has been rewritten:-
It quotes field names using
dbQuoteIdentifier()
, rather than use a flawed black-list based approach with name munging. -
It now throws errors on failure, rather than returning FALSE.
-
It will automatically add row names only if they are character, not integer.
-
When loading a file from disk,
dbWriteTable()
will no longer attempt to guess the correct values forrow.names
andheader
- instead supply them explicitly if the defaults are incorrect. -
When given a zero-row data frame it will just creates the table definition.
-
-
Assorted fixes accumulated since last release 3 years ago.
-
MySQL()
no longer hasforce.reload
argument - it's not obvious that this ever worked.