Re: [Swarm-Support] Subject: alternaties to inFile and outFile for ascii text?

[Marcus G. Daniels]

Paul Johnson wrote:

> Can you point me at the syntax on setting variables & arrays, possibly 
> just an explanation of the attr settings here.

Well, I don't want to get into a tutorial on R.   There are books and a 
user support list for R is good.

Arrays are created in R with

ary <- array(dims=c(1,2,3))

that would make a multidimensional array that is 1*2*3 and assign it to 
the variable "ary". Access can be complete ary[1,2,3], or partial 
ary[1,2,].  The latter returns a vector for the third dimension.

Vectors in R are simply:

v <- c(1,2,3)

Vectors are accessed with v[1], etc.

Finally the `attr' settings:  These are hints to Swarm to tell it what 
data structure to use to load it.  The rows of the table should get put 
into an (Objective C) objects of type MyClass and collected into a Swarm 
List.  It works the same in Java.

> Is there a way to just set a vector of doubles and have it read in 
> directly?  

You can save about any kind of variable using the R hdf5 module.  It's 
more a question of how to load it in Swarm. Although one could use the 
HDF5 interface in Swarm to do raw loads of arrays or vectors, I'd 
instead suggest making minimal data-related classes that contain related 
arrays or whatever and then just use object serialization.  You'll have 
more self-documenting metadata and structure in your program that way.

> Or, suppose I want to insert an array of integers valued 0 and 1 and 
> want them to come into Objective-C as BOOL.

Use "as.logical(v)" where v is your integer vector.  The result when 
loaded into Swarm will work as an Objective C BOOL.