Last active
November 10, 2019 19:53
-
-
Save ganzuul/2cae364c0b9f7ae7beae5de44bacc328 to your computer and use it in GitHub Desktop.
en.wikipedia.org/wiki/Special:WhatLinksHere to R via REST API
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| #install.packages("httr") #install.packages("jsonlite") #install.packages("tidyverse") | |
| #install.packages("stringr") #install.packages("furrr") #install.packages("igraph") | |
| require(httr) | |
| require(jsonlite) | |
| require(tidyverse) | |
| require(stringr) | |
| #require(furrr) #require(igraph) | |
| url_base <- "https://en.wikipedia.org/w/api.php?action=query&format=json&prop=linkshere&lhlimit=500" | |
| articles <- c("Maximal and minimal elements", "Maxima and minima") | |
| titles <- function(v) {paste("&titles=", v, sep = "")} | |
| ids <- function(v) {paste("&pageids=", v, sep = "")} | |
| query <- function(v) {paste(url_base, v, sep = "")} | |
| get_lh <- function(v) {as.data.frame(fromJSON(content(GET(v), "text"), flatten = FALSE))} | |
| map(articles, URLencode) %>% | |
| map(titles) %>% | |
| map(query) %>% | |
| map(getLh) %>% | |
| map(~select(.x, c(4, 5, 7, 8))) %>% | |
| map(~as_tibble(.x, .name_repair = "minimal")) -> tibs | |
| map(tibs, function(v) {filter(v, complete.cases(v))}) %>% # Convenient NAs in data. | |
| map(~select(.x, c(2))) %>% | |
| map(flatten_chr) %>% | |
| map(ids) %>% | |
| map(as.list) %>% | |
| map(query) %>% | |
| map(as.list) %>% | |
| map(function(v) {v %>% map(get_lh)}) -> df1 | |
| df1 %>% | |
| map(function(v) {v %>% map(~select(.x, c(ends_with(".pageid"), ends_with(".title"))))}) -> tibs2 | |
| map(df1, function(v) {v %>% map(~select(.x, ends_with(".title")))}) %>% | |
| map(function(v) {v %>% map(~as.matrix(.x))}) -> g1 | |
| g1[[1]] %>% | |
| map(graph_from_edgelist) -> g | |
| #map(g1, function(v){v %>% map(~graph_from_edgelist(.x))}) -> g | |
| #truncCol <- function(v){names(v) %>% map(~str_trunc(.x, str_length(.x)-12, side = c("left"),ellipsis=""))} | |
| #tib0 <- map(tibs, function(v){names(v) <- truncCol(v);return(v)}) # TO-DO Why is this return special? | |
| #df <- graph.edgelist(matrix |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment