On Fri, Sep 3, 2010 at 3:53 PM, Jaroslav Hajek
<address@hidden> wrote:
I think you didn't quite understand my view of this. I don't think of
df.as.cell is a virtual object that holds the dataframe data (without
headers) converted as a cell, which then can normally be indexed.
I basically agree that as_cell (df (I,J)) is more standard, I merely
pointed out that the former has a potential performance advantage.
Probably something like as_cell(df){I,J} would have the same performance advantages and also avoid hand-written dispatch code.
--judd