[R] [R-pkgs] RODBC 1.1-1
Prof Brian Ripley
ripley at stats.ox.ac.uk
Wed Sep 15 17:38:17 CEST 2004
The first non-maintenance update of RODBC since January 2003 is now on
CRAN and will soon propagate to mirrors. From the ChangeLog:
* Select the decimal point from Sys.localeconv.
* Add an external reference and finalizer so open channels get
closed at the end of the session or when there is no R object
referring to them.
* There is no longer a restriction to 16 channels.
* Add NAMESPACE.
* odbcConnect{Access,Dbase,Excel} allow a missing file name
(and will bring up a dialog box to search for it).
* odbcGetInfo returns more information in a 8-element character
vector (based on an idea of Matthew Dowle).
* The C code calls SQLExecuteDirect rather than SQLExecute and
does not call SQLCloseCursor, based on a problem report from
Matthew Dowle using MS SQLServer. Repeated calls to
sqlGetResults now work.
* New function sqlFetchMore.
* Table names in Access with embedded spaces are mapped to the
[name space] form which Access requires.
* Table creation no longer removes _ from column names.
* New functions get/setSqlTypeInfo and the typeInfo argument to
sqlSave allow users to specify the mapping from R types to DBMS
datatypes. sqlSave also allows the specification of DBMS
datatypes by column.
* It is now possible to write more than 255 chars to a field with
sqlSave and sqlUpdate.
* Dates and timestamps are now read as 'Date' and 'POSIXct'
columns by sqlGetResults (unless as.is = TRUE for the column).
I have been able to test SQL Server reasonably extensively this time
around, as well as MySQL, PostgreSQL, Access, Excel and SQLite.
--
Brian D. Ripley, ripley at stats.ox.ac.uk
Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/
University of Oxford, Tel: +44 1865 272861 (self)
1 South Parks Road, +44 1865 272866 (PA)
Oxford OX1 3TG, UK Fax: +44 1865 272595
_______________________________________________
R-packages mailing list
R-packages at stat.math.ethz.ch
https://stat.ethz.ch/mailman/listinfo/r-packages
More information about the R-help
mailing list