Update metadata

[sweinand]

• Update country metadata
• Add special aggregate compositions

# Title:    Country metadata and EA/EU compositions
# Author:   Sebastian Weinand
# Date:     20 December 2023

# load packages:
library(openxlsx)
library(data.table)

# countries:
countries <- as.data.table(read.xlsx("U:/HICP/R/package/hicp/dev/countries.xlsx", detectDates=TRUE))
vars <- c("is_eu","is_ea","is_efta","is_candidate")
countries[, (vars) := lapply(.SD, as.logical), .SDcols=vars]

countries[, "produces_hicp" := !is.na(index_dec_publ) & is.na(eu_until)]

# EU composition of the reference period:
EU <- list(
  "code"="EU",
  "name_en"="European Union",
  "name_fr"="Union européenne",
  "name_de"="Europäische Union",
  "index_decimals"=2L,
  "composition"=dt[is_eu==TRUE, code])

# euro area composition of the reference period:
EA <- list(
  "code"="EA",
  "name_en"="Euro area",
  "name_fr"="Zone euro",
  "name_de"="Euroraum",
  "index_decimals"=2L,
  "composition"=dt[is_ea==TRUE, code])


# get special aggregates:
library(xml2)
doc <- read_xml("http:https://publications.europa.eu/resource/authority/ed1/ecoicop/B")
sa.urls <- xml_attr(x = xml_find_all(doc, xpath="https://skos:member"), attr="resource")
sa.codes <- sapply(X=strsplit(sa.urls, split="/", fixed=TRUE), FUN=function(z) z[length(z)])

SPECAGGS <- vector(mode="list", length=length(sa.urls))
for(j in seq_along(SPECAGGS)){

  cat("processing url", j, "/", length(SPECAGGS), "\r")

  sa <- read_xml(x=sa.urls[j])

  SPECAGGS[[j]]$code <- sa.codes[j]

  xpath <- paste0("//rdf:Description[contains(@rdf:about, '", sa.codes[j], "')]")

  xml.node <- xml_find_all(sa, xpath=xpath)

  SPECAGGS[[j]]$name_en <- xml_text(xml_find_all(xml.node, xpath="skos:prefLabel[@xml:lang='en']"))
  SPECAGGS[[j]]$name_fr <- xml_text(xml_find_all(xml.node, xpath="skos:prefLabel[@xml:lang='fr']"))
  SPECAGGS[[j]]$name_de <- xml_text(xml_find_all(xml.node, xpath="skos:prefLabel[@xml:lang='de']"))

  comp <- xml_attr(xml_find_all(xml.node, xpath="skos:member"), "resource")
  SPECAGGS[[j]]$composition <- sapply(X=strsplit(comp, split="/", fixed=TRUE), FUN=function(z) z[length(z)])

  Sys.sleep(runif(n=1, min=0.1, max=0.5))

}