[R] subtract a list of vectors from a list of data.frames in an elegant way

[Gabor Grothendieck]

On Thu, Mar 29, 2012 at 4:28 AM, Eric Fail <eric.fail at gmx.us> wrote:
> Dear R experts,
>
> I've realized that it might not be possible to define a negative SELCET statement in a SQL call so now I'm looking for the smoothest way to generate a list of what I would like from my large database by first pulling all the names with a query like this "SELECT top 1 * FROM your_table" (thank you Bart Joosen for the idea) and then subtract the variables I am not allow to pull manually ending up with a 'positive' definition of what I want, something I can use in a SQL SELCT statement (see my email on this list from yesterday for more on that).
>
> When I query the database for the variable names I get something similar to 'DBquery' in my working example below, but considerable longer with over 2400 hundred variables. As I only need to remove two or three variables I would like to define a lookup table (like the list 'lookup' in my example) and subtract that from my data base query. Now to my question. Is there a way I can subtract one list from another? Like setoff or alike?
>
> I would like to end up with a list like the one shown in my example called 'result.' In short, I would like to subtract 'lookup' from 'DBquery' and end up with 'result,' please note that 'result' is a list fo vecktors and not a list of dataframes. In my real life example DBquery is considerable longer so defining that by hand would make a really really long syntax.
>

One can use  except  in sqlite and in a number of other database systems:

library(sqldf)

# first two rows of BOD
BOD12 <- BOD[1:2, ]

# all rows of BOD except those in BOD12
sqldf("select * from BOD except (select * from BOD12)")



-- 
Statistics & Software Consulting
GKX Group, GKX Associates Inc.
tel: 1-877-GKX-GROUP
email: ggrothendieck at gmail.com