First package functions and documentation

DESCRIPTION

@@ -16,6 +16,9 @@ License: GPL (>= 2)
 Encoding: UTF-8
 Roxygen: list(markdown = TRUE)
 RoxygenNote: 7.3.1
+Imports: 
+    dplyr,
+    stats
 Suggests: 
     knitr,
     rmarkdown,

NAMESPACE

@@ -1,3 +1,7 @@
 # Generated by roxygen2: do not edit by hand
 
-export(print_msg)
+export(sts_get_metadata)
+export(sts_get_sads)
+importFrom(dplyr,left_join)
+importFrom(stats,aggregate)
+importFrom(stats,reshape)

R/sts_get_metadata.R

@@ -0,0 +1,62 @@
+#' @title Prepare Data For Analyses
+#' 
+#' @description Simple function to get species abundances and plot
+#'   metadata together for analyses
+#' 
+#' @param x the SynTreeSys database
+#' @param output character. Should the plot metadat be returned (i.e.
+#'   'Plots') or the the species table (i.e. 'SpeciesObservations')
+#'   with the plot metadata? Defaults to 'SpeciesObservations'.
+#'
+#' @return A data frame
+#' 
+#' @importFrom dplyr left_join
+#' 
+#' @export
+#' 
+sts_get_metadata <- function(x, output = "SpeciesObservations") {
+
+  ## Isolating the tables
+  plots <- x$Plots
+  comms <- x$CommunityObservations
+  spps <- x$SpeciesObservations
+
+  ## Checking table contents
+  plots_plots <- unique(plots$uniquePlotID)
+  plots_comms <- unique(comms$uniquePlotID)
+  plots_spps <- unique(spps$uniquePlotID)
+
+  check1 <- setdiff(plots_plots, plots_comms)
+  if (length(check1) > 0L) {
+    warning("One or more plots in the CommunityObservations are not in the Plots table. Removing them from the CommunityObservations table")
+    comms <- comms[!comms$uniquePlotID %in% check1, ]
+  }
+
+  check2 <- setdiff(plots_plots, plots_spps)
+  if (length(check2) > 0L) {
+    warning("One or more plots in the SpeciesObservations are not in the Plots table. Removing them from the Plots table")
+    plots <- plots[!plots$uniquePlotID %in% check2, ]
+  }
+
+  ## Checking for duplicated entries
+  dup_comms_plot <- comms$uniquePlotID[duplicated(comms$uniquePlotID)]
+  if  (length(dup_comms_plot) > 0L){
+    warning("One or more plots in CommunityObservations are duplicated. Removing: ", 
+            paste(dup_comms_plot,collapse=", "))
+    comms <- comms[!duplicated(comms$uniquePlotID),]
+  }
+
+  ## Getting the necessary descriptions for the species observations
+  dados1 <- dplyr::left_join(plots, comms,
+                             by = "uniquePlotID")
+
+  if (output == "Plots") {
+    return(dados1)    
+  }
+
+  if (output == "SpeciesObservations") {
+    spps1 <- dplyr::left_join(spps, dados1,
+                              by = "uniquePlotID")
+    return(spps1)    
+  }
+}

R/sts_get_sads.R

@@ -0,0 +1,62 @@
+#' @title Get the Species Abundances Distributions per Group
+#' 
+#' @param x the SynTreeSys database
+#' @param minN numerical. The minimum number of individuals per plot.
+#'   Defaults to zero.
+#' @param maxN numerical. The maximum number of individuals per plot.
+#'   Defaults to Inf.
+#' @param minCutOff numerical. The minimum size criteria cutoff of the
+#'   plot (in mm). Defaults to zero.
+#' @param group.name character. The STS column name with the categorie
+#'   used for grouping abundances.
+#' @param spp.name character. The STS column name with the
+#'   (morpho)species name. Defaults to "organismNameCurated".
+#' @param remove.indet logical. Should morphospecies not identified at
+#'   the species level be removed? Defaults to TRUE
+#' @param abund.metric character. The STS column name with the
+#'   abundance metric to be used to build the SAD. Defaults to
+#'   "countsMeasurement"
+#' @param strata.names vector of characters. The names of the strata
+#'   the should be kept in the analyses.
+#' 
+#' 
+#' @importFrom stats aggregate reshape
+#' 
+#' 
+#' @export
+#' 
+sts_get_sads <- function(x, 
+                     minN = 0, 
+                     maxN = Inf, 
+                     minCutOff = 0, 
+                     group.name = NULL,
+                     spp.name = "organismNameCurated",
+                     remove.indet = TRUE,
+                     abund.metric = "countsMeasurement",
+                     strata.names = NULL) {
+
+  ids2keep <- x$individuals >= minN & x$sizeCutoffMin >= minCutOff
+  plot2keep <- x$uniquePlotID[ids2keep]
+  x1 <- x[x$uniquePlotID %in% plot2keep,]
+
+  if (remove.indet)
+    x1 <- x1[x1$organismNameTaxonRank %in% "species", ]
+
+  if(!is.null(strata.names)){
+    ## TO BE FINSHED ##
+  }
+
+  sad.per.group <- stats::aggregate(x1[[abund.metric]], 
+                             list(x1[[spp.name]], x1[[group.name]]), 
+                             sum, na.rm = TRUE)
+  names(sad.per.group) <- c(spp.name, group.name, abund.metric)
+  sad.per.group1 <- stats::reshape(sad.per.group, direction = "wide",
+                              idvar = spp.name,
+                              timevar = group.name,
+                              v.names = abund.metric)
+  colnames(sad.per.group1) <- gsub("countsMeasurement.", "",
+                                     colnames(sad.per.group1))
+  sad.per.group1[is.na(sad.per.group1)] <- 0
+
+  return(sad.per.group1)  
+}

tests/testthat/test-demo.R

@@ -1,14 +1,4 @@
 ## Test demo ----
 
 test_that("Test demo", {
-  expect_invisible(print_msg())
-
-  x <- print_msg()
-  expect_equal(x, "Hello world")
-
-  x <- print_msg("Bonjour le monde")
-  expect_equal(x, "Bonjour le monde")
-
-  expect_error(print_msg(c("Hello", "world")))
-  expect_error(print_msg(1), "Argument 'x' must be a character of length 1.")
 })