Updated documentation

R/classifyQA.R

@@ -5,7 +5,7 @@
 #' @param img RasterLayer or SpatRaster. Landsat 8 OLI QA band.
 #' @param type Character. Classes which should be returned. One or more of c("background", "cloud", "cirrus","snow", "water").
 #' @param confLayers Logical. Return one layer per class classified by confidence levels, i.e. cloud:low, cloud:med, cloud:high.
-#' @param ... further arguments passed to \link[raster]{writeRaster}
+#' @param ... further arguments passed to \link[terra]{writeRaster}
 #' @inheritParams encodeQA
 #' 
 #' @seealso \link{encodeQA} \link{decodeQA}

R/coregisterImages.R

@@ -12,7 +12,7 @@
 #' @param nBins Integer. Number of bins to calculate joint histogram.
 #' @param reportStats Logical. If \code{FALSE} it will return only the shifted images. Otherwise it will return the shifted image in a list containing stats such as mutual information per shift and joint histograms.
 #' @param verbose Logical. Print status messages. Overrides global RStoolbox.verbose option.
-#' @param ... further arguments passed to \code{\link[raster]{writeRaster}}.
+#' @param ... further arguments passed to \code{\link[terra]{writeRaster}}.
 #' @param slave DEPRECATED! Argument was renamed. Please use \code{img} from now on.
 #' @param master DEPRECATED! Argument was renamed. Please use \code{ref} from now on.
 #' @details 

R/ggR.R

@@ -29,7 +29,7 @@
 #' When \code{img} contains factor values and \code{annotation=TRUE}, the raster values will automatically be converted
 #' to numeric in order to proceed with the brightness calculation.
 #' 
-#' The raster package provides a class lookup-table for categorical rasters (e.g. what you get if you run superClass in classification mode). If your raster has a lookup-table ggR will automatically treat it as categorical (see \link[raster]{factor}). 
+#' The raster package provides a class lookup-table for categorical rasters (e.g. what you get if you run superClass in classification mode). If your raster has a lookup-table ggR will automatically treat it as categorical (see \link[terra]{factor}).
 #' However, the factor status of Raster objects is easily lost and the values are interpreted as numeric. In such cases you should make use of the \code{forceCat = TRUE} argument, which makes sure
 #' that ggplot2 uses a discrete scale, not a continuous one.
 #' 

R/ggRGB.R

@@ -2,7 +2,7 @@
 #' 
 #' Calculates RGB color composite raster for plotting with ggplot2. Optional values for clipping and and stretching can be used to enhance the imagery.
 #' 
-#' Functionality is based on \code{\link[raster]{plotRGB}} from the raster package.
+#' Functionality is based on \code{\link[terra]{plotRGB}} from the raster package.
 #' 
 #' @param img RasterStack or RasterBrick
 #' @param r Integer or character. Red layer in x. Can be set to \code{NULL}, in which case the red channel will be set to zero.

R/histMatch.R

@@ -11,7 +11,7 @@
 #' @param intersectOnly Logical. If \code{TRUE} sampling will only take place in the overlap extent of the two rasters. Otherwise the full rasters will be used for sampling.
 #' @param paired Logical. If \code{TRUE} the corresponding pixels will be used in the overlap.
 #' @param returnFunctions Logical. If \code{TRUE} the matching functions will be returned instead of applying them to \code{x}. 
-#' @param ... Further arguments to be passed to \link[raster]{writeRaster}.
+#' @param ... Further arguments to be passed to \link[terra]{writeRaster}.
 #' @param forceInteger Logical. Force integer output.
 #' @note \code{x} and \code{ref} must have the same number of layers.
 #' @return A SpatRaster of \code{x} adjusted to the histogram of \code{ref}. If \code{returnFunctions  = TRUE} a list of functions (one for each layer) will be returned instead.

R/mesma.R

@@ -11,7 +11,7 @@
 #' @param iterate Integer. Set maximum iteration per pixel. Processing time could increase the more iterations are made possible. Default is 400.
 #' @param tolerance Numeric. Tolerance limit representing a nearly zero minimal number. Default is 1e-8. 
 #' @param verbose Logical. Prints progress messages during execution.
-#' @param ... further arguments passed to \link[raster]{writeRaster}.
+#' @param ... further arguments passed to \link[terra]{writeRaster}.
 #' 
 #' @return SpatRaster. The object will contain one band per endmember, with each value representing the estimated presence probability of the endmember per pixel (0 to 1), and an RMSE band.
 #' 

R/rasterCVA.R

@@ -21,13 +21,14 @@
 #' Returns a SpatRaster with two layers: change vector angle and change vector magnitude
 #' @export 
 #' @examples
+#' \donttest{
 #' library(terra)
 #' pca <- rasterPCA(lsat)$map
 #'
 #' ## Do change vector analysis
 #' cva <- rasterCVA(pca[[1:2]], pca[[3:4]])
 #' cva
-#' plot(cva)
+#' }
 rasterCVA <- function(x, y, tmf = NULL, nct = NULL,  ...) {
 	x <- .toTerra(x)
 	y <- .toTerra(y)

R/rasterPCA.R

@@ -20,10 +20,11 @@
 #' @param spca Logical. If \code{TRUE}, perform standardized PCA. Corresponds to centered and scaled input image. This is usually beneficial for equal weighting of all layers. (\code{FALSE} by default)
 #' @param maskCheck Logical. Masks all pixels which have at least one NA (default TRUE is reccomended but introduces a slow-down, see Details when it is wise to disable maskCheck). 
 #' Takes effect only if nSamples is NULL.
-#' @param ... further arguments to be passed to \link[raster]{writeRaster}, e.g. filename.
+#' @param ... further arguments to be passed to \link[terra]{writeRaster}, e.g. filename.
 #' @return Returns a named list containing the PCA model object ($model) and a SpatRaster with the principal component layers ($object).
 #' @export 
 #' @examples
+#' \donttest{
 #' library(ggplot2)
 #' library(reshape2)
 #' ggRGB(rlogo, 1,2,3)
@@ -42,6 +43,7 @@
 #'   plots <- lapply(1:3, function(x) ggR(rpc$map, x, geom_raster = TRUE))
 #'   grid.arrange(plots[[1]],plots[[2]], plots[[3]], ncol=2)
 #' }
+#' }
 rasterPCA <- function(img, nSamples = NULL, nComp = nlyr(img), spca = FALSE,  maskCheck = TRUE, ...){
     img <- .toTerra(img)
 
@@ -54,7 +56,7 @@ rasterPCA <- function(img, nSamples = NULL, nComp = nlyr(img), spca = FALSE,  ma
         ellip[["norm"]] <- NULL
     }
 
-    if(nComp > terra::nlyr(img)) nComp <- terra::nlyr(img)
+    if(nComp > nlyr(img)) nComp <- nlyr(img)
 
     if(!is.null(nSamples)){
         trainData <- terra::spatSample(img, size = nSamples, na.rm = TRUE)
@@ -70,7 +72,7 @@ rasterPCA <- function(img, nSamples = NULL, nComp = nlyr(img), spca = FALSE,  ma
         covMat <- terra::layerCor(img, "cov", na.rm = TRUE)
         model  <- stats::princomp(covmat = covMat$covariance, cor = spca)
         model$center <- covMat$mean
-        model$n.obs  <- t(global(img, "sum", na.rm = TRUE))
+        model$n.obs  <- ncell(any(!is.na(img)))
 
         if(spca) {    
             ## Calculate scale as population sd like in in princomp
@@ -79,7 +81,8 @@ rasterPCA <- function(img, nSamples = NULL, nComp = nlyr(img), spca = FALSE,  ma
         }
     }
     ## Predict
-    out <- .paraRasterFun(img, predict, args = list(model = model, index = 1:nComp), wrArgs = ellip)
+    out   <- .paraRasterFun(img, terra::predict, args = list(model = model, na.rm = TRUE, index = 1:nComp), wrArgs = ellip)
+
     names(out) <- paste0("PC", 1:nComp)
     structure(list(call = match.call(), model = model, map = out), class = c("rasterPCA", "RStoolbox"))  
 

R/readMeta.R

@@ -268,7 +268,7 @@ readMeta <- function(file, raw = FALSE){
 #' @param na Numeric vector. No-data value per band
 #' @param vsat Numeric vector. Saturation value per band
 #' @param scal Numeric vector. Scale factor per band. e.g. if data was scaled to 1000*reflectance for integer conversion.
-#' @param dtyp Character vector. Data type per band. See \code{\link[raster]{dataType}} for options.
+#' @param dtyp Character vector. Data type per band. See \code{\link[terra]{dataType}} for options.
 #' @param radRes Numeric vector. Radiometric resolution per band.
 #' @param spatRes Numeric vector. Spatial resolution per band.
 #' @param calrad data.frame. Calibration coefficients for dn->radiance conversion. Must have columns 'gain' and 'offset'. Rows named according to \code{bands}.

R/topCor.R

@@ -3,15 +3,15 @@
 #' account and correct for changes in illumination due to terrain elevation.
 #' 
 #' @param img SpatRaster. Imagery to correct
-#' @param dem SpatRaster. Either a digital elevation model as a RasterLayer or a RasterStack/Brick with pre-calculated slope and aspect (see \link[raster]{terrain}) in which case the layers must be named 'slope' and 'aspect'.
+#' @param dem SpatRaster. Either a digital elevation model as a RasterLayer or a RasterStack/Brick with pre-calculated slope and aspect (see \link[terra]{terrain}) in which case the layers must be named 'slope' and 'aspect'.
 #' Must have the same dimensions as \code{img}.
 #' @param metaData Character, ImageMetaData. Either a path to a Landsat meta-data file (MTL) or an ImageMetaData object (see \link{readMeta}) 
 #' @param solarAngles Numeric vector containing sun azimuth and sun zenith (in radians and in that order). Not needed if metaData is provided   
 #' @param method Character. One of c("cos", "avgcos", "minnaert", "C", "stat", "illu"). Choosing 'illu' will return only the local illumination map.
 #' @param stratImg RasterLayer or SpatRaster to define strata, e.g. NDVI. Or the string 'slope' in which case stratification will be on \code{nStrat} slope classes. Only relevant if \code{method = 'minnaert'}.
 #' @param nStrat Integer. Number of bins or quantiles to stratify by. If a bin has less than 50 samples it will be merged with the next bin. Only relevant if \code{method = 'minnaert'}.
 #' @param illu SpatRaster. Optional pre-calculated ilumination map. Run topCor with method="illu" to calculate an ilumination map
-#' @param ... arguments passed to \code{\link[raster]{writeRaster}}
+#' @param ... arguments passed to \code{\link[terra]{writeRaster}}
 #' @details
 #' For detailed discussion of the various approaches please see Riano et al. (2003).
 #' 

R/unsuperClass.R

@@ -16,7 +16,7 @@
 #' Clustering is done using \code{\link[stats]{kmeans}}. This can be done for all pixels of the image (\code{clusterMap=FALSE}), however this can be slow and is
 #' not memory safe. Therefore if you have large raster data (> memory), as is typically the case with remote sensing imagery it is advisable to choose clusterMap=TRUE (the default).
 #' This means that a kmeans cluster model is calculated based on a random subset of pixels (\code{nSamples}). Then the distance of *all* pixels to the cluster centers 
-#' is calculated in a stepwise fashion using \code{\link[raster]{predict}}. Class assignment is based on minimum euclidean distance to the cluster centers.   
+#' is calculated in a stepwise fashion using \code{\link[terra]{predict}}. Class assignment is based on minimum euclidean distance to the cluster centers.
 #' 
 #' The solution of the kmeans algorithm often depends on the initial configuration of class centers which is chosen randomly. 
 #' Therefore, kmeans is usually run with multiple random starting configurations in order to find a convergent solution from different starting configurations.
@@ -122,7 +122,7 @@ unsuperClass <- function(img, nSamples = 10000, nClasses = 5, nStarts = 25, nIte
 #' @param object unsuperClass object
 #' @param img Raster object. Layernames must correspond to layernames used to train the superClass model, i.e. layernames in the original raster image.
 #' @param output Character. Either 'classes' (kmeans class; default) or 'distances' (euclidean distance to each cluster center).
-#' @param ... further arguments to be passed to \link[raster]{writeRaster}, e.g. filename
+#' @param ... further arguments to be passed to \link[terra]{writeRaster}, e.g. filename
 #' @export
 #' @return
 #' Returns a raster with the K-means distances base on your object passed in the arguments.

tests/testthat/test-panSharpen.R

@@ -3,9 +3,6 @@ context("panSharpen")
 library(terra)
 
 test_that("panSharpen methods", {
-	skip_on_cran()
-	skip_on_covr()
-	skip_on_ci()
 	suppressWarnings({
 		agg     <- aggregate(lsat, 10)
 		pan     <- sum(lsat[[1:3]])