James Wettenhall <[EMAIL PROTECTED]> writes: > Currently there is no official interface between R arrays and > Tcl arrays, so you have to write your own. > > I've sketched some ideas on : > http://bioinf.wehi.edu.au/~wettenhall/RTclTkExamples/tktable.html > > Here's a summary.
[implementation skipped] > tt <- tktoplevel() > tclArray <- tclArrayVar() > tclArrayName <- ls(tclArray$env) > table1 <- tkwidget(tt,"table",variable=tclArrayName) > tkpack(table1) > > # Set/Modify the values of some cells. > tclArray[0,0] <- "Zero" > tclArray[0,1] <- "One" > tclArray[1,0] <- "Two" > tclArray[1,1] <- "Three" > > # Now get the value of a cell. > tclArray[0,1] I've been working on some array stuff for 1.8.0. It doesn't quite pan out like the above because Tktable really uses arrays in a rather special way. A general Tcl array is more like a Perl hash, or an R environment (or list with non-fixed length), so you want to be able to do A[["foo"]] <- "bar", i.e. the index can be an arbitrary string. Tktable uses the special indices "0,0" "0,1" ... to simulate a matrix. Since it is probably useful to interface to the full capabilities of Tcl arrays (every time I skip something, someone comes up with a case where it is needed...), the thing that I have done works like A <- tclArray() A[["1,1"]] <- "blah" # autoconverts to tclObj A[["1,1"]] # a tclObj A$"1,1" # also works The matrix-like items could be implememnted on top of this, but I'm a little unsure precisely how, whether it would require a special class or not, and whether one should implement the various forms of smart indexing. -- O__ ---- Peter Dalgaard Blegdamsvej 3 c/ /'_ --- Dept. of Biostatistics 2200 Cph. N (*) \(*) -- University of Copenhagen Denmark Ph: (+45) 35327918 ~~~~~~~~~~ - ([EMAIL PROTECTED]) FAX: (+45) 35327907 ______________________________________________ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
