[R-sig-DB] "easydb" package

[Julien Moeys]

Hi all,

I have recently created a new project/package on r-forge, "easydb".
http://easydb.r-forge.r-project.org/

In short:
- This package provides a simplified interface to SQLite, MS Access and MySQL (on top of RSQLite and RODBC);
- Identical interface (= same functions & arguments) for these 3 databases;
- Data subsetting style similar to data.frame, with single square brackets
  myDatabase[ "myTable", row_constrains, col_name ]
- Intended to be user friendly;
- Similar to what had been done by John Fox and Oswaldo Cruz in their package dfdb-rodbc;
- Database connection handled internally;
- Some SQL queries handled internally;
- Minimal operation log system, to keep track of operations done on a database;
- Internally handles SQL's "LAST_INSERT_ID()" for the 3 databases;

Limitations:
- Not intended to replace DBI and RODBC. Only the most common operations are covered.
- Not designed for speed (but rather usability)

All functions are documented + I made a vignette too that cover the most important operation.

I was considering to release the package on CRAN, to make it more visible to potential users. Nonetheless, I was wondering what could be the opinion of the R/Database community on this package:

Could it be beneficial to the R community, especially R beginners? Or will it add confusion to the confusion (Yet another database package, Yet another data subsetting style)? 

I was also wondering how to make sure that the code always works as it is supposed to work. As far as I know there are no possibilities for automatic tests of RODBC Access and MySQL during package compilation on r-forge and CRAN -- At the moment I run the tests (testthat, ...) on my local Windows / Linux machines.

all the best

Julien
PS: I must admit I am not an R guRu, and neither an SQL pro... But the package has been tested on rather data-intensive applications.