I have the following problem. I have an automatically generated named
list with "stringified" names:
a <- list("A"=..., "B"=..., "C"=..., )
then I want to refer to the elements of the list, stored as an vector
of names:
nn <- c("A", "B", "C"), so that I could get
list elements like
a$nn[1], a$nn[2], etc. Obviously it doesn't work. Instead I do:
nn.Exp <- substitute(expression(a$b), list(b=nn[1]))
eval(nn.Exp)
in a result I get
expession(a$"A") but not the value stored in the list.
Meanwhile if I manually construct expression as expression(a$"A") and
then evaluate it, it works fine.
How do I solve that problem? Perhaps using such a list with
"stringified" names are not very good programming style, but this is
convenient to store and retrieve elements if list should be filled
automatically from numerous sources. In this way I'm trying to emulate
a hash-table behavior. Perhaps there is a better way?
Any help is highly appreciated
Thanks,
-Serge
Serge Boiko wrote:> I have the following problem. I have an automatically generated named > list with "stringified" names: > > a <- list("A"=..., "B"=..., "C"=..., ) > > then I want to refer to the elements of the list, stored as an vector > of names: > > nn <- c("A", "B", "C"), so that I could get list elements like > > a$nn[1], a$nn[2], etc. Obviously it doesn't work. Instead I do:a$nn is evaluated at first, which looks for a list element called "nn"! It's the convinient form for a[["nn"]].> nn.Exp <- substitute(expression(a$b), list(b=nn[1])) > eval(nn.Exp)That's an enormous *overhead*. To construct such an expression, I'd try (given here just for fun): nn.Exp <- substitute(a$b, list(b=nn[1])) eval(nn.Exp)> in a result I get > expession(a$"A") but not the value stored in the list. > Meanwhile if I manually construct expression as expression(a$"A") and > then evaluate it, it works fine. > > How do I solve that problem? Perhaps using such a list with > "stringified" names are not very good programming style, but this is > convenient to store and retrieve elements if list should be filled > automatically from numerous sources. In this way I'm trying to emulate > a hash-table behavior. Perhaps there is a better way? > > Any help is highly appreciatedUse the more general [[ ]] indexing mechanism for lists as in: a[[nn[1]]] a[[nn[2]]] See "An Introduction to R" or any book on the S language for more details regarding indexing. Uwe Ligges
On Wed, Jan 29, 2003 at 01:07:52PM +0100, Serge Boiko wrote:> > I have the following problem. I have an automatically generated named > list with "stringified" names: > > a <- list("A"=..., "B"=..., "C"=..., ) > > then I want to refer to the elements of the list, stored as an vector > of names: > > nn <- c("A", "B", "C"), so that I could get list elements like > > a$nn[1], a$nn[2], etc. Obviously it doesn't work. Instead I do: > > nn.Exp <- substitute(expression(a$b), list(b=nn[1])) > eval(nn.Exp) > > in a result I get > expession(a$"A") but not the value stored in the list. > Meanwhile if I manually construct expression as expression(a$"A") and > then evaluate it, it works fine. > > How do I solve that problem? Perhaps using such a list with > "stringified" names are not very good programming style, but this is > convenient to store and retrieve elements if list should be filled > automatically from numerous sources. In this way I'm trying to emulate > a hash-table behavior. Perhaps there is a better way?If you want hash table behavior you could try using environments (as that is really about all that they are). I have been spending some time thinking (and a little coding) about a hash table class but it is unlikely to appear before the summer. Robert> > Any help is highly appreciated > Thanks, > -Serge > > ______________________________________________ > R-help at stat.math.ethz.ch mailing list > http://www.stat.math.ethz.ch/mailman/listinfo/r-help-- +---------------------------------------------------------------------------+ | Robert Gentleman phone : (617) 632-5250 | | Associate Professor fax: (617) 632-2444 | | Department of Biostatistics office: M1B20 | Harvard School of Public Health email: rgentlem at jimmy.dfci.harvard.edu | +---------------------------------------------------------------------------+