gender_count.R

# This script is quick and dirty code. It is very preliminary and may be improved one day ;)
library("jsonlite")
library("tidyverse")
library("httr")
library("WikidataR")

paste0(
 "https://www.wikidata.org/w/api.php?action=wbgetentities&", 
 "ids=Q6581097|Q6581072", 
 "&languages=fr&props=labels&format=json"
 ) %>% 
  GET() %>% 
  content() %>%
  pluck("entities")

get_claims <- function(x) {
  paste0(
    "https://www.wikidata.org/w/api.php?action=wbgetentities&titles=", 
    URLencode(x), 
    "&sites=frwiki&format=json&props=claims"
  ) %>%
    GET() %>% 
    content() %>%
    pluck("entities", 1, "claims")
}
get_claims("Ada Lovelace")

get_gender <- function(x) {
  gender <- get_claims(x) %>%
    pluck("P21") %>%
    pluck(1,"mainsnak", "datavalue", "value", "id") 
  gender <- ifelse(is.null(gender), "", gender)
  tibble(x, gender)
}
get_gender("Ada Lovelace")

get_genderlinks <- function(title) {
  df <- paste0(
    "https://fr.wikipedia.org/w/api.php?", 
    "format=json&origin=*&action=query&prop=links&titles=", 
    URLencode(title), 
    "&pllimit=1000&plnamespace=0") %>%
    GET() %>%
    content() %>%
    pluck("query", "pages", 1, "links") %>%
    map_df(.x =  ., .f = as_tibble) %>%
    pull("title") %>% 
    set_names() %>%
    map_df(.x = ., .f = get_gender) 
    print(nrow(df))

  return(
    df %>%
      filter(gender != "")
    )
}


get_genderlinks(title = "Sociologie") %>% count(gender)
get_genderlinks(title = "Science politique") %>% count(gender)
get_genderlinks(title = "Microéconomie") %>% count(gender)