All together now:

?merge

:-)


-----Original Message-----
From:   [EMAIL PROTECTED] on behalf of Paul Sorenson
Sent:   Mon 1/24/2005 10:34 PM
To:     [email protected]
Cc:     
Subject:        [R] lookups and joins
I have some data coming from SQL sources that I wish to relate in various ways. 
 For reasons only known to our IT people, this can't be done in SQL at present.

I am looking for an R'ish technique for looking up new columns on a data frame. 
 As a simple, hardwired example I have tried the following:

# This gives me two columns, one the lookup value and the second one
# the result column, ie my lookup table.
stcl = read.csv("stockclass.csv")
stockclass = as.vector(stcl$stock_class)
# This gives me what appears to be a dictionary or map
names(stockclass) = as.vector(stcl$stock_group)

getstockclass = function(stock_group) {
        try(stockclass[[stock_group]], TRUE)
}
csg$stk_class=factor(sapply(csg$stock_group, getstockclass))

I need the try since if there is a missing value I get an exception.

I also tried something along the lines of (from memory):
getstockclass = function(stock_group) {
        stcl[which(stcl$stock_group == stock_group),]$stock_class
}

These work but I just wanted to check if there was an inbuilt way to do this 
kind of thing in R?  I searched on "join" without much luck.

Really what I would like is a generic function that:
        - Takes 2 data frames,
        - Some kind of specification on which column(s) to join
        - Outputs the joined frames, or perhaps a vector which is an index 
vector that I can use on the second data frame.

I don't really want to reinvent SQL and my data sets are not huge.

cheers

______________________________________________
[email protected] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html

______________________________________________
[email protected] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html

Reply via email to