Merge pull request epiforecasts#417 from epiforecasts/estonia-data

Add Estonia data

.github/workflows/Estonia.yaml

@@ -0,0 +1,48 @@
+on:
+  schedule:
+    - cron: '36 12 * * *'
+  workflow_dispatch:
+
+name: Estonia
+
+jobs:
+  Estonia:
+    runs-on: macOS-latest
+    env:
+      GITHUB_PAT: ${{ secrets.GITHUB_TOKEN }}
+    steps:
+      - uses: actions/checkout@v2
+
+      - uses: r-lib/actions/setup-r@v1
+
+      - name: Query dependencies
+        run: |
+          install.packages('remotes')
+          saveRDS(remotes::dev_package_deps(dependencies = TRUE), ".github/depends.Rds", version = 2)
+          writeLines(sprintf("R-%i.%i", getRversion()$major, getRversion()$minor), ".github/R-version")
+        shell: Rscript {0}
+
+      - name: Cache R packages
+        uses: actions/cache@v2
+        with:
+          path: ${{ env.R_LIBS_USER }}
+          key: ${{ runner.os }}-${{ hashFiles('.github/R-version') }}-1-${{ hashFiles('.github/depends.Rds') }}
+          restore-keys: ${{ runner.os }}-${{ hashFiles('.github/R-version') }}-1-
+
+      - name: Install dependencies
+        run: |
+          install.packages(c("remotes"))
+          remotes::install_deps(dependencies = TRUE)
+          install.packages("devtools")
+        shell: Rscript {0}
+
+      - name: Install package
+        run: R CMD INSTALL .
+
+      - name: Test dataset
+        run: |
+          options("testDownload" = TRUE)
+          options("testSource" = "Estonia")
+          devtools::load_all()
+          testthat::test_file("tests/testthat/test-regional-datasets.R", reporter = c("summary", "fail"))
+        shell: Rscript {0}

DESCRIPTION

@@ -1,6 +1,6 @@
 Package: covidregionaldata
 Title: Subnational Data for COVID-19 Epidemiology
-Version: 0.9.2.1000
+Version: 0.9.2.2000
 Authors@R:
     c(person(given = "Joseph",
              family = "Palmer",
@@ -121,4 +121,4 @@ Encoding: UTF-8
 Language: en-gb
 LazyData: true
 Roxygen: list(markdown = TRUE)
-RoxygenNote: 7.1.1
+RoxygenNote: 7.1.2

NAMESPACE

@@ -10,6 +10,7 @@ export(Covid19DataHub)
 export(Cuba)
 export(DataClass)
 export(ECDC)
+export(Estonia)
 export(France)
 export(Germany)
 export(Google)
@@ -73,6 +74,7 @@ importFrom(dplyr,slice_tail)
 importFrom(dplyr,starts_with)
 importFrom(dplyr,summarise)
 importFrom(dplyr,tally)
+importFrom(dplyr,transmute)
 importFrom(dplyr,ungroup)
 importFrom(dplyr,vars)
 importFrom(httr,GET)

NEWS.md

@@ -1,3 +1,11 @@
+# covidregionaldata 0.9.3
+
+This release is currrently under development
+
+## New data sets
+
+- Support for level 1 region data in Estonia (thanks to @RichardMN). See `?Estonia` for details. 
+
 # covidregionaldata 0.9.2
 
 This release adds support for the Covid19 Data Hub which includes Google and Apple mobility data amongst a large range of other data sets, data from the European Commission's Joint Research Centre which is at both the regional and national level, and individual sources for regional data from several countries. Package updates have been made in line with a software review at the [Journal of Open Source Software](https://github.com/openjournals/joss-reviews/issues/3290). Finally, this release exposes more of the testing infrastructure to users and adds a package hexsticker. 

R/Estonia.R

@@ -0,0 +1,90 @@
+#' Estonia Class for downloading, cleaning and processing notification data
+#' @description Information for downloading, cleaning
+#'  and processing COVID-19 region data for Estonia
+#'
+# nolint start
+#' @source \url{https://www.terviseamet.ee/et/koroonaviirus/avaandmed}
+# nolint end
+#' @export
+#' @concept dataset
+#' @family subnational
+#' @examples
+#' \dontrun{
+#' region <- Estonia$new(verbose = TRUE, steps = TRUE, get = TRUE)
+#' region$return()
+#' }
+Estonia <- R6::R6Class("Estonia",
+  inherit = DataClass,
+  public = list(
+
+    # Core Attributes
+    #' @field origin name of origin to fetch data for
+    origin = "Estonia",
+    #' @field supported_levels A list of supported levels.
+    supported_levels = list("1"),
+    #' @field supported_region_names A list of region names in order of level.
+    supported_region_names = list("1" = "county"),
+    #' @field supported_region_codes A list of region codes in order of level.
+    supported_region_codes = list("1" = "iso_3166_2"),
+    #' @field common_data_urls List of named links to raw data.
+    # nolint start
+    common_data_urls = list(
+      "main" = "https://opendata.digilugu.ee/opendata_covid19_test_county_all.csv" # nolint
+    ),
+    # nolint end
+    #' @field source_data_cols existing columns within the raw data
+    source_data_cols = c("cases_new", "tested_new",
+                         "cases_total", "tested_total"),
+    #' @field source_text Plain text description of the source of the data
+    source_text = "Estonian Ministry of Social Affairs",
+    #' @field source_url Website address for explanation/introduction of the
+    #' data
+    source_url = "https://www.terviseamet.ee/et/koroonaviirus/avaandmed",
+
+
+    #' @description Set up a table of region codes for clean data
+    #' @importFrom tibble tibble
+    set_region_codes = function() {
+      self$codes_lookup$`1` <- tibble(
+        code = c("EE-37",
+          "EE-39", "EE-45", "EE-52", "EE-50", "EE-56", "EE-60", "EE-68",
+          "EE-64", "EE-71", "EE-74", "EE-79", "EE-81", "EE-84", "EE-87",
+          NA),
+        region = c("Harju", "Hiiu", "Ida-Viru", "J\u00e4rva",
+          "J\u00f5geva", "L\u00e4\u00e4ne", "L\u00e4\u00e4ne-Viru",
+          "P\u00e4rnu", "P\u00f5lva", "Rapla", "Saare", "Tartu",
+          "Valga", "Viljandi", "V\u00f5ru", "Unknown")
+      )
+    },
+
+    #' @description Estonia specific state level data cleaning
+    #' @importFrom dplyr select mutate transmute
+    #' @importFrom tidyr pivot_wider
+    #' @importFrom rlang .data
+    #'
+    clean_common = function() {
+      self$data$clean <- self$data$raw[["main"]] %>%
+        pivot_wider(
+          id_cols = c("LastStatisticsDate", "StatisticsDate", "Country",
+                      "CountryEHAK", "County", "CountyEHAK"),
+          names_from = ResultValue,
+          values_from = c("DailyTests", "TotalTests",
+                          "DailyCases", "TotalCases")) %>%
+        select(-Country, -CountryEHAK) %>%
+        mutate(TestPositivity =
+                 ifelse((DailyTests_N + DailyTests_P) > 0,
+                        (DailyTests_P / (DailyTests_N + DailyTests_P)),
+                        NA)) %>%
+      transmute(
+        level_1_region = gsub(" maakond", "", .data$County),
+        level_1_region_code = gsub("00", "EE-", .data$CountyEHAK),
+        date = .data$StatisticsDate,
+        cases_new = .data$DailyCases_P,
+        tested_new = .data$DailyTests_P + .data$DailyTests_N,
+        cases_total = .data$TotalCases_P,
+        tested_total = .data$TotalTests_P + .data$TotalTests_N,
+        test_positivity = .data$TestPositivity
+      )
+    }
+  )
+)