Update WDI data

.gitignore

@@ -6,3 +6,4 @@ playground.R
 figure/
 docs
 inst/doc
+data-raw/wdi_downloads/

DESCRIPTION

@@ -13,7 +13,7 @@ URL: https://github.com/pik-piam/GDPuc,
     https://pik-piam.github.io/GDPuc/
 BugReports: https://github.com/pik-piam/GDPuc/issues
 Depends:
-    R (>= 2.10)
+    R (>= 3.5)
 Imports:
     cli (>= 2.4.0),
     crayon,
@@ -26,17 +26,14 @@ Imports:
     tidyselect,
     withr
 Suggests:
-    countrycode,
     covr,
     knitr,
     magclass,
     madrat (>= 3.6.4),
     purrr,
     rmarkdown,
-    stringr,
     testthat (>= 3.0.0),
     usethis,
-    WDI,
     zoo
 VignetteBuilder:
     knitr

R/convertGDP.R

@@ -6,6 +6,7 @@
 #' conversion factors (PPPs).
 #'
 #' @details
+#' # Using a custom source
 #' When providing a custom source to the function, a certain format is required.
 #' The source object must be a data frame or tibble with at least the following columns:
 #' \itemize{
@@ -17,43 +18,31 @@
 #'    The base year of the deflator can be any year, and can be country-specific.
 #'    \item a numeric column named "MER (LCU per US$)" with MER values,
 #'    \item a numeric column named "PPP conversion factor, GDP (LCU per international $)"
-#'    wit PPP exchange rate values.
+#'    with PPP exchange rate values.
 #'  }
 #'
-#' @param gdp A tibble, data frame or magpie object, the latter of which
-#'   requires the [magclass](https://github.com/pik-piam/magclass)
-#'   package to be installed. The data-frame needs to have at least 2 columns, in some cases 3:
+#' @param gdp A tibble, data frame or magpie (see [magclass](https://github.com/pik-piam/magclass)) object.
+#' The data-frame needs to have at least 2 columns, in some cases 3:
 #'   \itemize{
 #'     \item a character column with iso3c
 #'     ([wikipedia](https://en.wikipedia.org/wiki/ISO_3166-1_alpha-3)) country codes,
-#'     \item a numeric column with years (only required when converting from or to current currencies),
+#'     \item a numeric column with years (only required when converting from or to "current" currencies),
 #'     \item a numeric column named "value" with GDP values.
 #'  }
-#' @param unit_in A string with the incoming GDP unit, one of:
-#'   \itemize{
-#'     \item "current LCU"
-#'     \item "current Int$PPP"
-#'     \item "current US$MER"
-#'     \item "constant YYYY LCU"
-#'     \item "constant YYYY Int$PPP"
-#'     \item "constant YYYY US$MER"
-#'     \item "constant YYYY €" or "constant YYYY EUR"
-#'     \item "constant YYYY xxx_CU"
-#'   }
-#'   where YYYY should be replaced with a year e.g. "2010" or "2017".
-#' @param unit_out A string with the outgoing GDP unit, one of:
-#'   \itemize{
-#'     \item "current LCU"
-#'     \item "current Int$PPP"
-#'     \item "current US$MER"
-#'     \item "constant YYYY LCU"
-#'     \item "constant YYYY Int$PPP"
-#'     \item "constant YYYY US$MER"
-#'     \item "constant YYYY €" or "constant YYYY EUR"
-#'     \item "constant YYYY xxx_CU"
-#'   }
-#'   where YYYY should be replaced with a year e.g. "2010" or "2017", and xxx with a valid iso3c country code,
-#'   e.g. "JPN_CU" to pick the currency unit of Japan.
+#' @param unit_in,unit_out A string with the units to convert from and to. The following units are available:
+#'
+#' * "current LCU"
+#' * "current Int$PPP"
+#' * "current US$MER"
+#' * "current xxx_CU"
+#' * "constant yyyy LCU"
+#' * "constant yyyy Int$PPP"
+#' * "constant yyyy US$MER"
+#' * "constant yyyy €" or "constant yyyy EUR"
+#' * "constant yyyy xxx_CU"
+#'
+#' Here "LCU" stands for Local Currency Unit, "yyyy" is a placeholder for a year, e.g. "2010" or "2015", and xxx is a
+#' placeholder for a valid iso3c country code with for instance "JPN_CU" designating the currency unit of Japan
 #'
 #' @param source A string referring to a package internal data frame containing the conversion factors, or
 #'   a data-frame that exists in the calling environment.
@@ -124,7 +113,7 @@ convertGDP <- function(gdp,
   # The following line needs to be updated every time the output of convertGDP is affected by an update!
   # This is a trick, so that madrat caching works correctly. For more information, see the documentation of the madrat
   # R-package.
-  "last changes 2025-11-19"
+  "last changes 2026-07-15"
 
   # Save all function arguments as list
   arg <- as.list(environment())

R/print_source_info.R

@@ -22,21 +22,21 @@ print_source_info <- function(source) {
   if (missing(source) || source == "wb_wdi") {
     cli_source_info(name = "wb_wdi",
                     origin = "The World Bank's World Development Indicator Database",
-                    date = "Downloaded on the 30th of April 2024",
+                    date = "Downloaded on the 15th of July 2026",
                     html = "https://databank.worldbank.org/source/world-development-indicators",
                     note = "Uses the GDP deflator.")
   }
   if (missing(source) || source == "wb_wdi_linked") {
     cli_source_info(name = "wb_wdi_linked",
                     origin = "The World Bank's World Development Indicator Database",
-                    date = "Downloaded on the 30th of April 2024",
+                    date = "Downloaded on the 15th of July 2026",
                     html = "https://databank.worldbank.org/source/world-development-indicators",
                     note = "Uses the linked GDP deflator.")
   }
   if (missing(source) || source == "wb_wdi_cpi") {
     cli_source_info(name = "wb_wdi_cpi",
                     origin = "The World Bank's World Development Indicator Database",
-                    date = "Downloaded on the 30th of April 2024",
+                    date = "Downloaded on the 15th of July 2026",
                     html = "https://databank.worldbank.org/source/world-development-indicators",
                     note = "Uses the CPI as deflator.")
   }

README.Rmd

@@ -65,11 +65,15 @@ The `unit_in` and `unit_out` arguments specify the incoming and outgoing GDP uni
 -   current LCU
 -   current US$MER
 -   current Int$PPP
--   constant YYYY LCU
--   constant YYYY US$MER
--   constant YYYY Int$PPP
-
-Here "YYYY" is a placeholder for a year, e.g. "2010" or "2015", and "LCU" stands for Local Currency Unit.
+-   current xxx_CU
+-   constant yyyy LCU
+-   constant yyyy US$MER
+-   constant yyyy Int$PPP
+-   constant yyyy € or constant yyyy EUR
+-   constant yyyy xxx_CU
+
+Here "LCU" stands for Local Currency Unit, "YYYY" is a placeholder for a year, e.g. "2010" or "2015", and xxx is a
+placeholder for a valid iso3c country code with for instance "JPN_CU" designating the currency unit of Japan.
 
 For a quick conversion of a single value use `convertSingle`.
 

README.md

@@ -61,12 +61,17 @@ GDP units. All common GDP units are supported, i.e.:
 - current LCU
 - current US\$MER
 - current Int\$PPP
-- constant YYYY LCU
-- constant YYYY US\$MER
-- constant YYYY Int\$PPP
-
-Here “YYYY” is a placeholder for a year, e.g. “2010” or “2015”, and
-“LCU” stands for Local Currency Unit.
+- current xxx_CU
+- constant yyyy LCU
+- constant yyyy US\$MER
+- constant yyyy Int\$PPP
+- constant yyyy € or constant yyyy EUR
+- constant yyyy xxx_CU
+
+Here “LCU” stands for Local Currency Unit, “YYYY” is a placeholder for a
+year, e.g. “2010” or “2015”, and xxx is a placeholder for a valid iso3c
+country code with for instance “JPN_CU” designating the currency unit of
+Japan.
 
 For a quick conversion of a single value use `convertSingle`.
 

data-raw/create_internal_sources.R

@@ -1,133 +1,50 @@
 # Download data from WDI and save as internal package data
 # !! Don't forget to update the "date" section in print_source_info
 library(magrittr)
-rlang::check_installed(c("countrycode", "WDI", "stringr", "usethis"), reason = "in order to access the WDI database.")
-
-my_vars <- c(
-  "GDP (constant LCU)",
-  "GDP (current LCU)",
-  "GDP: linked series (current LCU)",
-  "GDP (constant 2015 US$)",
-  "GDP (current US$)",
-  "GDP, PPP (constant 2017 international $)",
-  "PPP conversion factor, GDP (LCU per international $)",
-  "GDP, PPP (current international $)",
-  "Population, total",
-  "GDP deflator (base year varies by country)",
-  "GDP deflator: linked series (base year varies by country)",
-  "DEC alternative conversion factor (LCU per US$)",
-  "Consumer price index (2010 = 100)"
-)
-
-my_info <- WDI::WDIsearch(
-  paste0("^",
-         paste0(stringr::str_replace_all(my_vars, c("\\$" = "\\\\$",
-                                                    "\\(" = "\\\\(",
-                                                    "\\)" = "\\\\)",
-                                                    "\\%" = "\\\\%")),
-                collapse = "$|^"),
-         "$"),
-  field = "name",
-  short = FALSE
-) %>%
-  tibble::as_tibble()
-
-# Get codes of only countries (excluding regional aggregates)
-wb_country_codes  <- WDI::WDIcache()$country %>%
-  dplyr::pull(iso3c) %>%
-  countrycode::countrycode("wb", "country.name") %>%
-  `[`(!is.na(.)) %>%
-  countrycode::countrycode("country.name", "wb")
-
-
-# Download data, remove aggregates and do some pivoting and renaming
-my_data <- purrr::map2(my_info$indicator,
-                       my_info$name,
-                       ~ WDI::WDI(indicator = .x, extra = TRUE) %>%
-                         tibble::as_tibble() %>%
-                         dplyr::filter(iso3c %in% wb_country_codes) %>%
-                         #dplyr::filter(!is.na(region) & region != "Aggregates") %>%
-                         dplyr::arrange(iso3c, year) %>%
-                         dplyr::select(iso3c, year, tidyselect::contains(.x)) %>%
-                         tidyr::pivot_longer(cols = tidyselect::contains(.x), names_to = "id") %>%
-                         dplyr::mutate(name = .y) %>%
-                         dplyr::select(iso3c, year, id, name, value)) %>%
-  purrr::list_rbind()
+rlang::check_installed(c("usethis"), reason = "to create the R/sysdata.rda file.")
+
+# my_vars <- c(
+#   "GDP (constant LCU)",
+#   "GDP (current LCU)",
+#   "GDP: linked series (current LCU)",
+#   "GDP (constant 2015 US$)",
+#   "GDP (current US$)",
+#   "GDP, PPP (constant 2021 international $)",
+#   "PPP conversion factor, GDP (LCU per international $)",
+#   "GDP, PPP (current international $)",
+#   "Population, total",
+#   "GDP deflator (base year varies by country)",
+#   "GDP deflator: linked series (base year varies by country)",
+#   "DEC alternative conversion factor (LCU per US$)",
+#   "Consumer price index (2010 = 100)"
+# )
+
+my_data <- readr::read_csv("~/Downloads/wdi_GDPuc_15Jul2026.csv", na = "..", show_col_types = FALSE) %>%
+  tidyr::pivot_longer(tidyselect::starts_with(c("19", "20")),
+                      names_to = "year",
+                      names_transform = ~as.integer(substr(.x, 1, 4))) %>%
+  dplyr::select("iso3c" = "Country Code", "year", "name" = "Series Name", "value") %>%
+  dplyr::arrange(.data$iso3c, .data$name, .data$year)
 
 wb_wdi <- my_data %>%
-  dplyr::select(-id) %>%
   tidyr::pivot_wider(names_from = name) %>%
   dplyr::mutate(`GDP deflator: linked series` = `GDP deflator: linked series (base year varies by country)` / 100,
                 `GDP deflator` = `GDP deflator (base year varies by country)` / 100,
                 `CPI` = `Consumer price index (2010 = 100)` / 100,
                 `MER (LCU per US$)` = `DEC alternative conversion factor (LCU per US$)`)
 
 wb_wdi_linked <- wb_wdi %>%
-  dplyr::select(iso3c,
-                year,
-                `GDP deflator` = `GDP deflator: linked series`,
-                `PPP conversion factor, GDP (LCU per international $)`,
-                `MER (LCU per US$)`)
+  dplyr::select("iso3c",
+                "year",
+                "GDP deflator" = "GDP deflator: linked series",
+                "PPP conversion factor, GDP (LCU per international $)",
+                "MER (LCU per US$)")
 
 wb_wdi_cpi <- wb_wdi %>%
-  dplyr::select(iso3c,
-                year,
-                `GDP deflator` = `CPI`,
-                `PPP conversion factor, GDP (LCU per international $)`,
-                `MER (LCU per US$)`)
+  dplyr::select("iso3c",
+                "year",
+                "GDP deflator" = "CPI",
+                "PPP conversion factor, GDP (LCU per international $)",
+                "MER (LCU per US$)")
 
-# For now, IMF is removed due to copyright issues
-#usethis::use_data(imf_weo, wb_wdi, wb_wdi_linked, internal = TRUE, overwrite = TRUE)
 usethis::use_data(wb_wdi, wb_wdi_linked, wb_wdi_cpi, internal = TRUE, overwrite = TRUE)
-
-
-
-
-#####################################################################
-#####################################################################
-#####################################################################
-#
-##### IMF DATA
-#
-# Path to the IMF WEO report from April 2021, downloaded as .xls file
-# imf_2021_file <- "../../R_projects/gdp_pop_trends/data/source/IMF/WEOApr2021all.xls"
-#
-# imf_weo <- readr::read_tsv(imf_2021_file) %>%
-#   dplyr::rename("iso3c" = ISO) %>%
-#   dplyr::filter(`WEO Subject Code` %in% c(
-#     "NGDP_R",
-#     "NGDP",
-#     "NGDPD",
-#     "PPPGDP",
-#     "NGDPRPPPPC",
-#     "NGDP_D",
-#     "PPPEX",
-#     "LP"
-#   )) %>%
-#   dplyr::select(iso3c, `WEO Subject Code`, tidyselect::starts_with(c("1", "2"))) %>%
-#   dplyr::mutate(dplyr::across(.cols = tidyselect::starts_with(c("1", "2")),
-#                               ~ stringr::str_remove_all(.x, ",") %>%
-#                                 as.double())) %>%
-#   tidyr::pivot_longer(tidyselect::starts_with(c("1", "2")),
-#                       names_to = "year") %>%
-#   dplyr::mutate(year = as.double(year)) %>%
-#   tidyr::pivot_wider(names_from = `WEO Subject Code`) %>%
-#   dplyr::rename(
-#     "GDP (constant LCU)" = NGDP_R,
-#     "GDP (current LCU)" = NGDP,
-#     "GDP (current US$)" = NGDPD,
-#     "GDP, PPP (current international $)" = PPPGDP,
-#     "GDPpc, PPP (constant 2017 international $)" = NGDPRPPPPC,
-#     "GDP deflator" = NGDP_D,
-#     "PPP conversion factor, GDP (LCU per international $)" = PPPEX,
-#     "Population, total" = LP
-#   ) %>%
-#   dplyr::mutate(
-#     `GDP (constant LCU)` = `GDP (constant LCU)` * 1e+9,
-#     `GDP (current LCU)` = `GDP (current LCU)` * 1e+9,
-#     `GDP (current US$)` = `GDP (current US$)` * 1e+9,
-#     `GDP deflator` = `GDP deflator` / 100,
-#     `GDP, PPP (current international $)` = `GDP, PPP (current international $)` * 1e+9,
-#     `MER (LCU per US$)` = `GDP (current LCU)` / `GDP (current US$)`,
-#     `Population, total` = `Population, total` * 1e+6
-#   )