Hi, Several years ago I was looking for ways to embed R graphics into Tk widgets, and abandoned the idea after finding out it was way too difficult, at least for my purposes. More recently, I looked again and found Luke Tierney's trkplot, which allows one to do just this. However, there seem to be limitations like inability to resize the graphs along with the widgets. That's not a big encumbrance in my case, but the difficulty in accessing the R plot's coordinates is. The only guidance I could find was that in James Wettenhall:
http://bioinf.wehi.edu.au/~wettenhall/RTclTkExamples/InteractiveTkrPlot.html but it issues an enigmatic message for the calculated coordinates: "Label Point Closest to These Approximate Plot Coordinates" Is that the best approach available, or are there better alternatives? My idea is to implement an integrated interface for a plotting utility I wrote for package diveMove, which gathers pairs of coordinates from mouse clicks (in standard unembedded graphs for now): ---<---------------cut here---------------start-------------->--- install.packages("diveMove") # from any CRAN data(divesTDR) depth.zoc <- zoc(getTime(divesTDR), getDepth(divesTDR), offset=3) coords <- plotTD(getTime(divesTDR), depth.zoc) ---<---------------cut here---------------end---------------->--- Clicking on the "Zero-Offset Correct a Range" button, followed by a pair of clicks in the graph, gathers coordinates using locator(). Because users may need to zoom and pan across different scales, it is important to allow for resizing of the plot. Any recommendations on how to embed R graphics on Tk widgets, and gathering coordinates as above, would be much appreciated. Thanks. Cheers, -- Seb _______________________________________________ R-SIG-GUI mailing list R-SIG-GUI@stat.math.ethz.ch https://stat.ethz.ch/mailman/listinfo/r-sig-gui
