splat: Bin Point Samples in Sparse Grids

#' #' @importFrom raster cellFromXY
#' splatbin <- function(grid, xy, weight = NULL, previters = 0) {
#'   summ <- data_frame(cell = cellFromXY(grid, xy), bin = 1) %>%
#'     group_by(cell) %>%
#'     summarize(bin = sum(bin))
#' }
#' 
#' 
#' splatchain <- function(grid, chain) {
#'   l <- vector("list", nrow(chain))
#'   for (i in seq_along(l)) l[[i]] <- splatbin(grid, t(chain[i,,]))
#'   do.call(bind_rows, l)
#' }
#' 
#' cellnumber.pimg <- function (x) 
#' {
#'   xbnd <- x$xbound
#'   ybnd <- x$ybound
#'   offs <- x$p[[1]]$offset
#'   img <- x$p[[1]]$image
#'   tl <- ((ybnd[3L] - (offs[2L] + ncol(img))) + 1L) * xbnd[3L] + 
#'     offs[1L]
#'   sort(rep(seq(tl, by = xbnd[3], length = ncol(img)), each = nrow(img)) + 
#'          rep(seq_len(nrow(img)) - 1, ncol(img)))
#' }
#' 
#' 
#' #' Title
#' #'
#' #' @param x Pimage
#' #' @param proj PROJ.4 string, optional
#' #'
#' #' @return list of data_frames
#' #' @export
#' #' @importFrom dplyr data_frame row_number %>%  filter_
#' pimageTosplat <- function(x, proj = NULL) {
#'   allbins <- allcells <- vector("list", length(x$p))
#' 
#' 
#'   for (i in seq_along(allcells)) {
#'     allcells[[i]] <- cellnumber.pimg(x[[i]])
#'     allbins[[i]] <- as.vector(x[[i]]$p[[1]]$image)
#'   }
#'   index <- lapply(seq_along(allcells), function(xa) rep(xa, length(allcells[[xa]])))
#'   pix <- dplyr::data_frame(cell = as.integer(unlist(allcells)), index = as.integer(unlist(index)), bin = as.integer(unlist(allbins)))
#'   time <- dplyr::data_frame(time = SGAT:::as.POSIXct.Pimage(x), index = as.integer(seq(length(x))))
#'   ## filter out zeroes and the performance via dplyr is pretty close
#'   list(pixel = pix %>% dplyr::filter_("bin > 0"), time = time)
#' }
#'