Reading dates in R using SQL and otherwise (and some interesting behavior by the data editor)

-----Original Message-----
From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-project.org]
On Behalf Of Paul Miller
Sent: April 8, 2010 11:14 AM
To: r-help at r-project.org
Subject: [R] Reading dates in R using SQL and otherwise (and
someinteresting behavior by the data editor)

Hello Everyone,
?
I am a newbie with about a month's worth of experience using R. I've
just spent a little time learning how to work with date values. Mostly,
this has involved converting text values into dates.
?
So far, I've managed to figure out how to do this in R proper using
as.Date. However, I'm still struggling with doing this using SQL and
RODBC.
?
In the process of learning to create date values in R proper, I noticed
some interesting behavior on the part of the data editor. At first, this
led me to believe that my efforts had been unsuccessful. The output from
my R console below illustrates this behavior.
?

test <- mydata
test$test_date <- as.Date(test$ae_datestarted, format='%m/%d/%Y')
class(test$test_date)

[1] "Date"

mode(test$test_date)

[1] "numeric"

fix(test)

?
(At this point, I clicked on the test_date column)

Warning: class discarded from column ???test_date???

class(test$test_date)

[1] "character"

mode(test$test_date)

[1] "character"

?
When I run my code, it works correctly. But when??I open the data frame
in the data editor and click on the test_date column, the editor says
that it is character. And beyond that, the editor discards the class for
test_date. Should the editor do this???Or is it my fault for trying to
look at test_date in the editor in the first place? In SAS, I'm used to
creating data and then opening the dataset to look at what I've done.
Maybe I shouldn't be doing this in R though.
?
Returning to the issue of converting text values to dates using SQL
(Server) and RODBC. Does anyone know how ot do this? I've been trying to
do this using things like Cast and Convert. Usually, these attempts
fail. When??SQL Server does seem to be sending something back, it
appears that R cannot accept it. Any help with this problem would be
greatly appreciated.
?
Thanks,
?
Paul?????????
?