Documentation fixes, removed unnecessary functions

Functions removed: get_data(), get_key(), quick_create_SAR()

Author: kmartinet

DESCRIPTION

@@ -17,7 +17,6 @@ Imports:
   mapdata,
   maps,
   reshape2,
-  rgbif,
   segmented,
   stringi,
   tidyr
@@ -32,5 +31,6 @@ Depends:
 Suggests: 
     testthat (>= 3.0.0),
     rmarkdown,
-    knitr (>= 1.46)
+    knitr (>= 1.46),
+    rgbif
 Config/testthat/edition: 3

NAMESPACE

@@ -9,12 +9,9 @@ export(estimate_DR)
 export(estimate_MS)
 export(find_areas)
 export(find_land)
-export(get_data)
 export(get_island_areas)
-export(get_key)
 export(get_presence_absence)
 export(get_sources)
-export(quick_create_SAR)
 export(remove_continents)
 import(mapdata)
 import(maps)

R/create_SAR.R

@@ -3,8 +3,9 @@
 #' Use segmented regression to create a species-area relationship (SAR) plot. 
 #' The X axis represents log(island area) and the Y axis represents log(number 
 #' of species)
-#' @param occurrences The dataframe output by `SSARP::find_areas()` (or if using  
-#' a custom dataframe, ensure that it has the following columns: Species, areas)
+#' @param occurrences The dataframe output by `SSARP::find_areas()` (or if 
+#' using  a custom dataframe, ensure that it has the following columns:
+#' specificEpithet, areas)
 #' @param npsi The maximum number of breakpoints to estimate for model 
 #' selection.  Default: 1
 #' @return A list of 3 including: the summary output, the segmented regression 
@@ -29,6 +30,10 @@ create_SAR <- function(occurrences, npsi = 1) {
   # Checkmate input validation
   checkmate::assertDataFrame(occurrences)
   checkmate::assertNumeric(npsi)
+  checkmate::testSubset(c("specificEpithet", "areas"), names(occurrences))
+  # Ensure columns are correct type
+  checkmate::assertCharacter(occurrences$specificEpithet)
+  checkmate::assertNumeric(occurrences$areas)
 
   #   formula Species ~ Area means to group scientific names by area
   #   function(x) length(unique(x)) tells it to give me the number of unique 

R/create_SpAR.R

@@ -29,6 +29,10 @@ create_SpAR <- function(occurrences, npsi = 1) {
   # Checkmate input validation
   checkmate::assertDataFrame(occurrences)
   checkmate::assertNumeric(npsi)
+  checkmate::testSubset(c("areas", "rate"), names(occurrences))
+  # Ensure columns are correct type
+  checkmate::assertNumeric(occurrences$areas)
+  checkmate::assertNumeric(occurrences$rate)
 
   # The purpose of this function is to create either a linear or segmented 
   #    regression to visualize the relationship between speciation rate and 

R/estimate_BAMM.R

@@ -11,7 +11,7 @@
 #' returned occurrence dataframe.
 #' @param occurrences The occurrence record dataframe output from the SSARP 
 #' pipeline. If you would like to use a custom dataframe, please make sure that 
-#' there are columns titled "Genus" and "Species"
+#' there are columns titled "genericName" and "specificEpithet"
 #' @param edata The eventdata object created by using the 
 #' `BAMMtools::getEventData()` function
 #' @return A dataframe that includes speciation rates for each species in the 
@@ -51,6 +51,10 @@ estimate_BAMM <- function(label_type = "binomial", occurrences, edata) {
   # Checkmate input validation
   checkmate::assertString(label_type)
   checkmate::assertDataFrame(occurrences)
+  checkmate::testSubset(c("specificEpithet", "genericName"), names(occurrences))
+  # Ensure columns are correct type
+  checkmate::assertCharacter(occurrences$specificEpithet)
+  checkmate::assertCharacter(occurrences$genericName)
 
   # Create a named number vector of speciation rates
   speciation_rates <- edata$meanTipLambda

R/find_areas.R

@@ -4,11 +4,12 @@
 #' masses relevant to the taxon of interest.
 #' @param occs The dataframe that is returned by `SSARP::find_land()`. If using 
 #' a custom occurrence record dataframe, ensure that it has the following 
-#' columns in order: "SpeciesName", "Genus", "Species", "Longitude", "Latitude", 
-#' "First", "Second", "Third", "datasetKey". The "datasetKey" column is 
-#' important for GBIF records and identifies the dataset to which the occurrence
-#' record belongs. Custom dataframes without this style of data organization 
-#' should fill the column with placeholder values.
+#' columns: "acceptedScientificName", "genericName", "specificEpithet", 
+#' "decimalLongitude", "decimalLatitude", "First", "Second", "Third", 
+#' "datasetKey". The "datasetKey" column is important for GBIF records and 
+#' identifies the dataset to which the occurrence record belongs. Custom 
+#' dataframes without this style of data organization should fill the column 
+#' with placeholder values.
 #' @param area_custom A dataframe including names of land masses and their 
 #' associated areas. This dataframe should be provided when the user would like 
 #' to bypass using the built-in database of island names and areas. Please 
@@ -41,7 +42,7 @@ find_areas <- function(occs, area_custom = NULL) {
       cli::cli_alert_warning("No data in occurrence record dataframe")
       break
     }
-    if(is.na(occs[i,8]) && is.na(occs[i,7]) && is.na(occs[i,6])) {
+    if(is.na(occs[i,"Third"]) && is.na(occs[i,"Second"]) && is.na(occs[i,"First"])) {
       minus[i] <- i
     }
   }
@@ -74,12 +75,12 @@ find_areas <- function(occs, area_custom = NULL) {
       cli::cli_alert_warning("No data in occurrence record dataframe")
       break
     }
-    if(!is.na(occs[i,8])) {
-      islands[i] <- occs[i,8]
-    } else if (!is.na(occs[i,7])){
-      islands[i] <- occs[i,7]
-    } else if (!is.na(occs[i,6])){
-      islands[i] <- occs[i,6]
+    if(!is.na(occs[i,"Third"])) {
+      islands[i] <- occs[i,"Third"]
+    } else if (!is.na(occs[i,"Second"])){
+      islands[i] <- occs[i,"Second"]
+    } else if (!is.na(occs[i,"First"])){
+      islands[i] <- occs[i,"First"]
     }
   }
 
@@ -141,12 +142,12 @@ find_areas <- function(occs, area_custom = NULL) {
 
   for(i in seq_len(nrow(occs))) {
 
-    if(!is.na(occs[i,8]) && island_dict$has(occs[i,8])){
-      areas[i] <- island_dict$get(occs[i,8])
-    } else if(!is.na(occs[i,7]) && island_dict$has(occs[i,7])){
-      areas[i] <- island_dict$get(occs[i,7])
-    } else if(!is.na(occs[i,6]) && island_dict$has(occs[i,6])){
-      areas[i] <- island_dict$get(occs[i,6])
+    if(!is.na(occs[i,"Third"]) && island_dict$has(occs[i,"Third"])){
+      areas[i] <- island_dict$get(occs[i,"Third"])
+    } else if(!is.na(occs[i,"Second"]) && island_dict$has(occs[i,"Second"])){
+      areas[i] <- island_dict$get(occs[i,"Second"])
+    } else if(!is.na(occs[i,"First"]) && island_dict$has(occs[i,"First"])){
+      areas[i] <- island_dict$get(occs[i,"First"])
     } else {
       areas[i] <- NA
     }

R/find_land.R

@@ -95,10 +95,10 @@ find_land <- function(occurrences, fillgaps = FALSE) {
         cli::cli_alert_warning("Occurrence record dataframe has no entries")
         break
       }
-      if(is.na(occs[i,6])){
+      if(is.na(occs[i,"First"])){
         # Get lon and lat
-        longitude <- occs[i,4]
-        latitude <- occs[i,5]
+        longitude <- occs[i,"decimalLongitude"]
+        latitude <- occs[i,"decimalLatitude"]
 
         # Create Photon URL
         url <- paste0("http://photon.komoot.io/reverse?lon=", 

R/get_data.R

@@ -7,13 +7,13 @@
 #' custom dataframe, ensure that it has the following named columns:
 #' - "areas" containing the areas associated with the land masses of interest
 #' - "Species" containing the names of the species living on those islands
-#' - A 6th column containing locality information. In the SSARP workflow, this 
-#' column is called "First" and contains the country name
-#' - A 7th column containing locality information. In the SSARP workflow, this 
-#' column is called "Second" and contains a province or island name
-#' - An 8th column containing locality information. In the SSARP workflow, this 
-#' column is called "Third" and contains the island name if the 7th column does 
-#' not contain the island name
+#' - "First" containing locality information. In the SSARP workflow, this 
+#' column contains the country name
+#' - "Second" containing locality information. In the SSARP workflow, this 
+#' column contains a province or island name
+#' - "Third" containing locality information. In the SSARP workflow, this 
+#' column contains the island name if the 7th column does not contain the 
+#' island name
 #' @return A dataframe with a row for each island in the given occurrence 
 #' record dataframe and a column for each species. Within each species column, 
 #' a 1 represents the presence of that species on the island corresponding to 
@@ -59,7 +59,7 @@ get_presence_absence <- function(occs) {
 
     # Pick the first occurrence of the ith area to get the name
     area_names <- rbind(area_names, occs[which(occs$areas == new_occs[i,1])[1],
-                                         c(6:8)])
+                                         c("First", "Second", "Third")])
   }
 
   # Now cbind area_names to new_occs