108 make output messages optional and reanimate caching

* adding progress bar for loading

* trying to get caching/output option to work

* passing output through

* reviving caching

* fixing counting

* roxygen update

* adding quiet option

* checking cached file

* documenting caching functionality

* getting message working

* removing cutting edge arrow

* reverting change back to cran, too soon

* nope arrow github not working yet

R/APCalign-package.R

@@ -75,6 +75,8 @@ utils::globalVariables(
     "taxon_ID",
     "taxon_ID_aligned",
     "taxon_rank",
+    "txtProgressBar",
+    "setTxtProgressBar",
     "taxonomic_status",
     "taxonomic_status_aligned",
     "taxonomic_status_genus",

R/align_taxa.R

@@ -90,8 +90,16 @@ align_taxa <- function(original_name,
           .default = readr::col_character()
         )
       )
-
-    # TODO: check taxa_ raw has correct columns
+    correct_names <- c("original_name", "aligned_name", "accepted_name", "suggested_name", 
+                           "genus", "family", "taxon_rank", "taxonomic_dataset", "taxonomic_status", 
+                           "taxonomic_status_aligned", "aligned_reason", "update_reason", 
+                           "subclass", "taxon_distribution", "scientific_name", "taxon_ID", 
+                           "taxon_ID_genus", "scientific_name_ID", "canonical_name", "row_number", 
+                           "number_of_collapsed_taxa", "checked", "known")
+    if(!identical(names(taxa_raw), correct_names)) {
+      stop("Your output file already exists and it's not in the right format. 
+         Please check that the file you are passing in to the output option.")
+    }
   }
   else {
     taxa_raw <-
@@ -149,29 +157,29 @@ align_taxa <- function(original_name,
     # take unique values so each name only processed once
     dplyr::filter(!duplicated(original_name))
 
-  if (all(taxa$tocheck$checked)) {
+  if (all(taxa$tocheck$checked)|all(is.na(taxa$tocheck$checked))) {
     message("  - all taxa are already checked, yay!")
     return(invisible(taxa$tocheck))
   }
 
   # move all checked taxa to "checked"
   taxa <- redistribute(taxa)
 
+  if (!is.null(output) && file.exists(output) && !all(taxa$tocheck$checked)) {
   # check unknown taxa
   message(
     "  -> ",
-    crayon::blue(sum(taxa$tocheck$known, na.rm = T)),
+    crayon::blue(sum(!is.na(taxa$checked$accepted_name), na.rm = T)),
     " names already matched; ",
     crayon::blue(sum(
-      taxa$tocheck$checked &
-        !taxa$tocheck$known,
+      is.na(taxa$checked$accepted_name),
       na.rm = T
     )),
-    " names checked but without a match; ",
-    crayon::blue(sum(!taxa$tocheck$checked)),
+    " names checked but without a species-level match; ",
+    crayon::blue(sum(!is.na(taxa$tocheck$original_name))),
     " taxa yet to be checked"
   )
-
+  }
   # do the actual matching
   taxa <- 
     match_taxa(taxa, resources, fuzzy_abs_dist, fuzzy_rel_dist, fuzzy_matches, imprecise_fuzzy_matches, APNI_matches, identifier) %>%
@@ -199,8 +207,10 @@ align_taxa <- function(original_name,
   ## save outputs to file, useful for caching results 
   if (!is.null(output)) {
     dir.create(dirname(output), FALSE, TRUE)
+    taxa$checked<-TRUE
+    taxa$known<-!is.na(taxa$aligned_name)
     readr::write_csv(taxa, output)
-    message("  - output saved in file: ", output)
+    #message("  - output saved in file: ", output)
   }
 
   return(taxa)

R/create_taxonomic_update_lookup.R

@@ -15,7 +15,7 @@
 #' @param APNI_matches Name matches to the APNI (Australian Plant Names Index) are turned off as a default. 
 #' @param imprecise_fuzzy_matches Imprecise fuzzy matches are turned on as a default.
 #' @param identifier A dataset, location or other identifier, which defaults to NA.
-#' @param output file path to save the intermediate output to
+#' @param output file path to save the output. If this file already exists, this function will check if it's a subset of the species passed in and try to add to this file. This can be useful for large and growing projects. 
 #' @return A lookup table containing the accepted and suggested names for each original name input, and additional taxonomic information such as taxon rank, taxonomic status, taxon IDs and genera. 
 #' - original_name: the original plant name.
 #' - aligned_name: the input plant name that has been aligned to a taxon name in the APC or APNI by the align_taxa function.
@@ -66,7 +66,8 @@ create_taxonomic_update_lookup <- function(taxa,
     align_taxa(taxa, resources = resources, 
                APNI_matches = APNI_matches, 
                identifier = identifier, 
-               imprecise_fuzzy_matches = imprecise_fuzzy_matches)
+               imprecise_fuzzy_matches = imprecise_fuzzy_matches,
+               output=output)
 
   updated_data <- 
     update_taxonomy(aligned_data, 

R/load_taxonomic_resources.R

@@ -10,6 +10,8 @@
 #' @param version The version number of the dataset to use. Defaults to the default version.
 #'
 #' @param reload A logical indicating whether to reload the dataset from the data source. Defaults to FALSE.
+#' 
+#' @param quiet A logical indicating whether to print status of loading to screen. Defaults to FALSE.
 #'
 #' @return The taxonomic resources data loaded into the global environment.
 #' @export
@@ -23,20 +25,29 @@
 load_taxonomic_resources <-
   function(stable_or_current_data = "stable",
            version = default_version(),
-           reload = FALSE) {
-    message("Loading resources...", appendLF = FALSE)
-    on.exit(message("...done"))
+           reload = FALSE,
+           quiet = FALSE) {
+
+
 
     taxonomic_resources <- dataset_access_function(
       version = version,
       path = tools::R_user_dir("APCalign"),
       type = stable_or_current_data
     )
-
+
+
+    total_steps <- 3  # Define how many steps you expect in the function
+    pb <- utils::txtProgressBar(min = 0, max = total_steps, style = 2)
+    if(!quiet){
+      message("Loading resources into memory...")
+      utils::setTxtProgressBar(pb, 1)  
+    }
     if(is.null(taxonomic_resources)) {
       return(NULL)
     }
 
+
     # Give list names
     names(taxonomic_resources) <- c("APC", "APNI")
 
@@ -167,6 +178,8 @@ load_taxonomic_resources <-
       dplyr::filter(taxonomic_status != "accepted") %>%
       dplyr::mutate(taxonomic_dataset = "APC")
 
+
+    if(!quiet) utils::setTxtProgressBar(pb, 2) 
     # Repeated from above - bionomial, tronomials etc
     taxonomic_resources[["APNI names"]] <-
       taxonomic_resources$APNI %>%
@@ -236,6 +249,7 @@ load_taxonomic_resources <-
       dplyr::mutate(taxonomic_dataset = "APC") %>%
       dplyr::distinct(canonical_name, .keep_all = TRUE)
 
+    if(!quiet) utils::setTxtProgressBar(pb, 3) 
     taxonomic_resources[["genera_APNI"]] <-
       taxonomic_resources$APNI %>%
       dplyr::select(
@@ -269,6 +283,8 @@ load_taxonomic_resources <-
       taxonomic_resources$APC %>%
       dplyr::filter(taxon_rank %in% c("family"), taxonomic_status == "accepted")
 
+    close(pb)
+    if(!quiet) message("...done")
     return(taxonomic_resources)
   }
 
@@ -460,6 +476,7 @@ dataset_get <- function(version = default_version(),
   path_to_apni <- file.path(path, paste0("apni", version, ".parquet"))
 
   APC <- if (!file.exists(path_to_apc)) {
+    message("Downloading...")
     download_and_read_parquet(apc.url, path_to_apc)
   } else {
     arrow::read_parquet(path_to_apc)

R/update_taxonomy.R

@@ -191,6 +191,8 @@ update_taxonomy <- function(aligned_data,
   taxa_out <- taxa_out  %>% dplyr::arrange(row_number)
 
   if (!is.null(output)) {
+    taxa_out$checked<-TRUE
+    taxa_out$known<-!is.na(taxa_out$accepted_name)
     readr::write_csv(taxa_out, output)
     message("  - output saved in file: ", output)
   }