Major upgrades for new version of AusTraits (#64)

* Testing R CMD check on other machines via GitHub Actions

* Bumped version

* Removed ubuntu #40

* Using austraits_lite in vignettes

* Using austraits_lite in structure vignette #40

* Implementing a new solution so dictionary can be precompiled before turning into a vignette #40

* Rebuilt site, vignettes are building now #40

* Changing global options in load-austraits #41

* Bug fix for parsing definitions data in dictionary.rmd #43

* Rebuilt pkgdown .htmls

* Bumped version number to minor release and minor edit to load_austraits

* xlab was not being imported properly, fixed

* Bumped the version number for dev version

* Updated README to make it clear that users need to install all dependencies and may need to upgrade exisiting packages that they may have for smooth install

* Updated pkgdown vignettes

* Moved janitor to Imports #47

* Rebuilt website using precompiled vignettes, new dir for figures

* More tests for plots #48

* Recreated austraits_lite so class is austraits, all extracted subsets are class austraits too, tweaked relevant tests #45

* Since australia_map_raster is a df, we can remove the raster import function in the plot_site_location function

* Bumped minor patch version number for PR

* Removed docs/ from PR
#49

* Revert docs to master

* Rm new files

* Rm new files in man

* Precompiled code for vignettes, figures for .RMD, removed old figs from README

* Updated README for lightweight install #49

* Updated file path to man/figures

* Updated file path to man/figures

* Added a test to detect #52

* ifelse for converting to numeric

* ifelse for converting to numeric #52

* Updated news folliwngfix #52

* Added site changes in NEWS and changed call to site_name to loc_name #54

* Fixed typo in wide_table, added to news, austraits$locations #54

* site_property to location_property #54

* Changed sites to locations in as_wide_table #54

* Renames plot_site_locations #54, added switch for join_taxonomy

* Convert versions to character strings

* Amended Join sites with version switch #55, #54

* Notes in join for changes

* Expanded join taxonomy to all versions

* Join methods tweaks for all versions

* Added alias for join_location and join_site, both functions for all versions

* Fixing contexts join

* updated and saved green .svg

* Condtional switch to print.austraits

* Upgrades to extract_dataset id, works with print.austraits #57

* #55 Join contexts working for multiple link vals

* Added two versions of join_methods

* Added switch for extract_trait #57

* Updated plot beeswarm to work with new and old versions of Austraits

* Added version switch for plot_locations with depracation warning for old plot_sites #59

* Added distinct to join_methods

* Added deprecation statement in join_sites #59

* Collapsed dataset ids in taxnomomic updates, beginnings of switches for trait_pivot_wider and internalised some switch functions

* Started to switch for extract_taxa

* Implemented taxon_name for extract_taxa #51

* Allowing extract_taxa to accept a string of family/genus/taxon_names #51

* Added switch for trait pivot

* changed to str_subset for more reaable code

* Plot_locations, accepts austraits or traits table

* Plot_locations, accepts austraits or traits table

* Added first pass as trait_pivot_wider #60

* Added switch for trait pivot longer #60

* Started to tinker with as_wide_table, removed vars option for join_contexts

* Fixed typo in as_wide_table1 #61

* Small changes to as_wide_table

* Added arg to collapse contexts for join_contexts, added as_wide_table upgrades #61

* Minor tweaks for methods#61

* Minor tweaks to roxygen tags #61

* Upgraded extract_taxa, and minor fix to extract_trait

* Minor tweaks

* Enabled lifecycle badges for soft deprecation for some functions, updated version numbers and news.md

* Minor fixes to rd files to rid warnings

* Created another austraits lite for testing new release

* Tests for as_wide_table

* Tweaked imports and removes .data in wide_table and added to tests

* rm .data$

* Rm import from rlang

* Commented out mem heavy test

* Removed files not needed and updated documentation

* Checked v number for testcoverage github actions

* Added missing topucs and search bar

* Updated version number for GHA

* Rm formatting lines from Rstudio

* Bumped v number

* Fixing up vignettes and adding to NEWS

* Tweaks to NAMESPACE, fixing vignettes so new version ofpackage works with austraits lite, new what_versions function

* Added establishment_means to join_taxonomy

* Changes species to taxa in print

* Added dplyr dependency

.github/workflows/R-CMD-check.yml

@@ -26,8 +26,8 @@ jobs:
       GITHUB_PAT: ${{ secrets.GITHUB_TOKEN }}
     steps:
       - uses: actions/checkout@v2
-      - uses: r-lib/actions/setup-r@v1
-      - uses: r-lib/actions/setup-pandoc@v1
+      - uses: r-lib/actions/setup-r@v2
+      - uses: r-lib/actions/setup-pandoc@v2
       - name: Install dependencies
         run: |
           install.packages(c("remotes", "rcmdcheck"))

.github/workflows/pkgdown_deploy.yml

@@ -17,9 +17,9 @@ jobs:
     steps:
       - uses: actions/checkout@v2
 
-      - uses: r-lib/actions/setup-r@v1
+      - uses: r-lib/actions/setup-r@v2
 
-      - uses: r-lib/actions/setup-pandoc@v1
+      - uses: r-lib/actions/setup-pandoc@v2
 
       - name: Query dependencies
         run: |

.github/workflows/test-coverage.yml

@@ -18,9 +18,9 @@ jobs:
     steps:
       - uses: actions/checkout@v2
 
-      - uses: r-lib/actions/setup-r@master
+      - uses: r-lib/actions/setup-r@v2
 
-      - uses: r-lib/actions/setup-pandoc@master
+      - uses: r-lib/actions/setup-pandoc@v2
 
       - name: Query dependencies
         run: |

.gitignore

@@ -51,3 +51,4 @@ data/austraits/
 
 #
 *.pdf
+*.svg

DESCRIPTION

@@ -1,6 +1,6 @@
 Package: austraits
 Title: Helpful functions to access, summarise and wrangle austraits data
-Version: 1.1.1
+Version: 2.1.1
 Authors@R:
     c(person(given = "Daniel",
              family = "Falster",
@@ -22,7 +22,7 @@ Encoding: UTF-8
 Language: en
 LazyData: true
 Roxygen: list(markdown = TRUE)
-RoxygenNote: 7.2.0
+RoxygenNote: 7.2.1
 Depends: 
     R (>= 4.0.0),
     RefManageR
@@ -38,7 +38,8 @@ Imports:
     jsonlite,
     utils,
     magrittr,
-    janitor
+    janitor,
+    lifecycle
 Suggests: 
     ggplot2,
     knitr,

NAMESPACE

@@ -11,6 +11,7 @@ export(get_version_latest)
 export(get_versions)
 export(join_all)
 export(join_contexts)
+export(join_locations)
 export(join_methods)
 export(join_sites)
 export(join_taxonomy)
@@ -20,6 +21,7 @@ export(lookup_trait)
 export(my_kable_styling_html)
 export(my_kable_styling_markdown)
 export(my_kable_styling_pdf)
+export(plot_locations)
 export(plot_site_locations)
 export(plot_trait_distribution_beeswarm)
 export(separate_trait_values)
@@ -28,8 +30,15 @@ export(summarise_trait_means)
 export(trait_pivot_longer)
 export(trait_pivot_wider)
 import(RefManageR)
+importFrom(dplyr,arrange)
+importFrom(dplyr,filter)
+importFrom(dplyr,group_by)
+importFrom(dplyr,select)
+importFrom(lifecycle,deprecated)
 importFrom(magrittr,"%>%")
-importFrom(rlang,.data)
-importFrom(rlang,abort)
 importFrom(stats,family)
+importFrom(stringr,str_detect)
+importFrom(tidyr,pivot_longer)
+importFrom(tidyr,pivot_wider)
+importFrom(tidyselect,all_of)
 importFrom(utils,methods)

NEWS.md

@@ -1,7 +1,24 @@
+# austraits 2.1.1
 
-# austraits 1.1.1
-
-* Improved vignette building process
-* Improved extract_ functions
-* Rebuilt website
-
+-   Minor bug fix in `extract_trait`
+-   Minor bug fix in `as_wide_table`
+-   `site` & `site_name` renamed to `location` & `location_name` within functions
+-   `austraits$sites` renamed to `austraits$locations`
+-   `plot_site_locations` is deprecated, use `plot_locations`
+-   `join_sites` is deprecated, use `join_locations`
+-   The following functions have been updated to work with AusTraits > v3.0.2, as well <=3.0.2
+    - `join_taxonomy`
+    - `join_locations`
+    - `join_contexts`
+    - `join_methods`
+    - `join_all`
+    - `trait_pivot_wider`
+    - `trait_pivot_longer`
+    - `as_wide_table`
+    - `extract_taxa`
+    - `extract_trait`
+    - `extract_dataset`
+    - `plot_locations`
+    - `plot_trait_distribution_beeswarm`
+    - `print.austraits`
+

R/as_wide_table.R

@@ -2,21 +2,120 @@
 #'
 #' @param austraits austraits data object
 #'
-#' @return A single wide table with collapsed contexts and sites text and with 
-#' some cols renamed for alignment with other reosurces
+#' @return A single wide table with collapsed contexts and locations text and with 
+#' some cols renamed for alignment with other resources
 #' @export
 #'
 #' @examples
 #' \dontrun{
 #' data <- austraits
 #' data %>% as_wide_table()
 #' }
-#' @importFrom rlang .data
 #' @importFrom stats family
 #' @importFrom utils methods
+
 as_wide_table <- function(austraits){
+  # Switch for different versions
+  version <- what_version(austraits)
+
+  switch (version,
+          'new' = as_wide_table2(austraits),
+          'old' = as_wide_table1(austraits),
+  )
+
+}
+
+#' Turning entire AusTraits object into wide table >3.0.2
+#' @noRd
+#' @keywords internal
+as_wide_table2 <- function(austraits){
+
+  # Function to collapse columns in locations and contexts into single column
+  process_table2 <- function(data) {
+    data %>% 
+      tidyr::pivot_wider(names_from = property, values_from = value) %>% 
+      tidyr::nest(data=-dplyr::any_of(c("dataset_id", "location_id", "latitude (deg)", "longitude (deg)"))) %>%
+      dplyr::mutate(location = purrr::map_chr(data, collapse_cols)) %>%
+      dplyr::select(-data) 
+  }
+
+  ################################################################################
+  # Define and adapt each table in the list of austraits to prepare for the wide table format 
+
+  # The contexts table needs the contexts collapsed to one context name per site
+  austraits %>% 
+    join_contexts(collapse_context = TRUE) -> austraits
 
+  # Getting rid of the columns that will soon be deleted in the next austraits release and renaming the description column
+  austraits$methods <- 
+    austraits$methods %>% 
+    #  -----------
+    # TODO: this section can be removed for next release
+    # Some studies have multiple records per traits. This breaks things when joining
+    # For now select first
+    # dplyr::group_by(dataset_id, trait_name) %>% 
+    # dplyr::slice(1) %>%
+    # dplyr:: ungroup() %>%
+    #----------
+    dplyr::rename(c("dataset_description" = "description"))  
+
+  # collapse into one column
+  austraits$locations <- 
+    austraits$locations %>% 
+    dplyr::filter(value!="unknown") %>% 
+    dplyr::rename(c("property" = "location_property")) %>%
+    split(., .$dataset_id) %>%
+    purrr::map_dfr(process_table2)
+
+  # rename taxonomic_reference field to reflect the APC/APNI name matching process better
+  austraits$taxa <- 
+    austraits$taxa %>% 
+    dplyr::rename(c("taxonNameValidation" = "taxonomic_reference"))
+
+  austraits_wide <- 
+    austraits$traits %>% 
+    dplyr::left_join(by=c("dataset_id", "location_id"), austraits$locations) %>%
+    dplyr::left_join(by=c("dataset_id", "trait_name"), austraits$methods) %>%
+    dplyr::left_join(by=c("taxon_name"), austraits$taxa)
+
+    # reorder the names to be more intuitive
+    austraits_wide %>% dplyr::select(
+
+    # The most useful (if you are filtering for just one taxon_name)
+      dataset_id, observation_id, trait_name, taxon_name, value, unit, 
+      entity_type, population_id, individual_id,
+      value_type, basis_of_value, 
+      replicates, 
+    # tissue, trait_category,  # Add after new zenodo release
+
+    # More stuff you can filter on
+    collection_date, basis_of_record, life_stage, sampling_strategy, 
+    treatment_id, temporal_id, 
+
+    #stuff relating to locations
+    `latitude (deg)`, `longitude (deg)`, location, plot_id,
+
+    #stuff relating to contexts and methods
+    context, methods, method_id, original_name,
+
+    #the citations
+    dataset_description, source_primary_citation, source_secondary_citation,
+
+    #the taxa details
+    taxonomic_status, taxon_distribution, 
+    taxon_rank, genus, family, #accepted_name_usage_id, 
+    scientific_name_authorship
+    )
+
+  austraits_wide
+}
 
+#' Turning entire AusTraits object into wide table <=3.0.2
+#' @noRd
+#' @keywords  internal
+as_wide_table1 <- function(austraits){
+
+
   ################################################################################
   # TODO: this updated with next zenodo release
   # Load the trait classification doc - classifies the tissue type and type of trait based on the trait_name data field
@@ -34,27 +133,17 @@ as_wide_table <- function(austraits){
   # 
   # Function to collapse columns in sites and contexts into single column
   process_table <- function(data) {
-
-    # worker function called intop worklfow below
-    # for a df, combine all column names and values
-    collapse_cols <- function(data) {
-
-      if(ncol(data) ==0) return(NA_character_)
-
-      data %>% purrr::imap_dfr(~ sprintf("%s='%s'",.y,.x)) %>%
-        tidyr::unite("text", sep="; ") %>% dplyr::pull(.data$text)
-    }
-
+
     data %>% 
-      tidyr::pivot_wider(names_from = .data$property, values_from = .data$value) %>% 
+      tidyr::pivot_wider(names_from = property, values_from = value) %>% 
       tidyr::nest(data=-dplyr::any_of(c("dataset_id", "site_name", "context_name", "latitude (deg)", "longitude (deg)"))) %>%
       dplyr::mutate(site = purrr::map_chr(data, collapse_cols)) %>%
       dplyr::select(-data) 
   }
 
   ################################################################################
   # Define and adapt each table in the list of austraits to prepare for the wide table format 
-
+  
   # the trait table needs little prep. Rename the value columns as value
   austraits$traits <- 
     austraits$traits %>% 
@@ -67,31 +156,31 @@ as_wide_table <- function(austraits){
     split(austraits$contexts$dataset_id) %>%
     purrr::map_dfr(process_table)  %>% 
     dplyr::rename(c("context" = "site"))
-
+  
   # Getting rid of the columns that will soon be deleted in the next austraits release and renaming the description column
   austraits$methods <- 
     austraits$methods %>% 
     #  -----------
-    # TODO: this section can be removed for next release
-    # Some studies have multiple records per traits. This breaks things when joining
-    # For now select first
-    dplyr::group_by(.data$dataset_id, .data$trait_name) %>% 
+  # TODO: this section can be removed for next release
+  # Some studies have multiple records per traits. This breaks things when joining
+  # For now select first
+  dplyr::group_by(dataset_id, trait_name) %>% 
     dplyr::slice(1) %>%
     dplyr:: ungroup() %>%
     #------------
-    dplyr::select(-.data$year_collected_start, -.data$year_collected_end) %>% 
+  dplyr::select(-year_collected_start, -year_collected_end) %>% 
     dplyr::rename(c("dataset_description" = "description"))  
 
   # collapse into one column
   austraits$sites <- 
     austraits$sites %>% 
-    dplyr::filter(.data$value!="unkown") %>% 
+    dplyr::filter(value!="unknown") %>% 
     # next line is a fix -- one dataset in 3.0.2 has value "site_name"
-    dplyr::mutate(site_property = gsub("site_name", "name", .data$site_property)) %>%
+    dplyr::mutate(site_property = gsub("site_name", "name", site_property)) %>%
     dplyr::rename(c("property" = "site_property")) %>%
     split(., .$dataset_id) %>%
     purrr::map_dfr(process_table)
-
+  
   # rename source data field to reflect the APC/APNI name matching process better
   austraits$taxa <- 
     austraits$taxa %>% 
@@ -103,32 +192,43 @@ as_wide_table <- function(austraits){
     dplyr::left_join(by=c("dataset_id", "site_name"), austraits$sites) %>%
     dplyr::left_join(by=c("dataset_id", "trait_name"), austraits$methods) %>%
     dplyr::left_join(by=c("taxon_name"), austraits$taxa) %>%
-
+    
     # reorder the names to be more intuitive
     dplyr::select(
 
-    # The most useful (if you are filtering for just one taxon_name)
-      .data$dataset_id, .data$observation_id, .data$trait_name, .data$taxon_name, .data$trait_value, .data$unit, 
-      .data$value_type, .data$replicates, 
-    # tissue, trait_category,  # Add after new zenodo release
-
-    # More stuff you can filter on
-    .data$date, .data$collection_type, .data$sample_age_class, .data$sampling_strategy, 
-
-    #stuff relating to sites
-    .data$`latitude (deg)`, .data$`longitude (deg)`, .data$site_name, .data$site,
-
-    #stuff relating to contexts and methods
-    .data$context_name, .data$context, .data$methods, .data$original_name,
-
-    #the citations
-    .data$dataset_description, .data$source_primary_citation, .data$source_secondary_citation,
-
-    #the taxa details
-    .data$taxonomicStatus, .data$taxonDistribution, 
-    .data$taxonRank, .data$genus, .data$family, .data$acceptedNameUsageID, 
-    .data$scientificNameAuthorship, .data$ccAttributionIRI
+      # The most useful (if you are filtering for just one taxon_name)
+      dataset_id, observation_id, trait_name, taxon_name, trait_value, unit, 
+      value_type, replicates, 
+      # tissue, trait_category,  # Add after new zenodo release
+      
+      # More stuff you can filter on
+      date, collection_type, sample_age_class, sampling_strategy, 
+      
+      #stuff relating to sites
+      `latitude (deg)`, `longitude (deg)`, site_name, site,
+      
+      #stuff relating to contexts and methods
+      context_name, context, methods, original_name,
+      
+      #the citations
+      dataset_description, source_primary_citation, source_secondary_citation,
+      
+      #the taxa details
+      taxonomicStatus, taxonDistribution, 
+      taxonRank, genus, family, acceptedNameUsageID, 
+      scientificNameAuthorship, ccAttributionIRI
     )
 
   austraits_wide
 }
+
+#' Collapse columns into text string
+#' @keywords internal
+#' @noRd
+collapse_cols <- function(data) {
+
+  if(ncol(data) ==0) return(NA_character_)
+
+  data %>% purrr::imap_dfr(~ sprintf("%s='%s'",.y,.x)) %>%
+    tidyr::unite("text", sep="; ") %>% dplyr::pull(text)
+}