Load jean.dat in R

jean_graph_parser.R

# Read and parse "jean.dat" from the Stanford GraphBase (C) 1993 Stanford University

library(tidyverse)
library(stringr)
library(igraph)
library(tidygraph)

jean <- readLines("http://ftp.cs.stanford.edu/pub/sgb/jean.dat")
jean <- jean[!str_detect(jean, "^\\*")]
jean_sep <- which(jean == "")

jean_nodes <- jean[1:(jean_sep - 1)]
jean_edges <- jean[(jean_sep + 1):length(jean)]

jean_nodes <- 
  jean_nodes %>% 
  str_replace_all("(\t)+$", "") %>%
  str_replace_all("(\t)+", "\t") %>%
  str_replace_all("(?<=^[A-Z]{2}) ", "\t") %>%
  str_replace_all(", ", "\t") %>%
  str_c(collapse = "\n") %>%
  read_tsv(col_names = c("Node_ID", "Node_name", "Node_desc"), col_types = "ccc")

comb_edge_list <- function(x){
  x <- str_split(x, ",")
  x <- unlist(x)
  if(length(x) == 1) x <- c(x, x)
  x <- combn(x, m = 2)
  x <- t(x)
}

jean_edges_chap <- str_extract(jean_edges, "^\\d+\\.\\d+\\.\\d+")

jean_edges <- 
  jean_edges %>% 
  str_extract("(?<=:).*") %>% 
  str_replace_all(pattern = " ", replacement = "") %>% 
  str_replace_all(pattern = ";$|,$", replacement = "") %>% 
  str_split(";") %>%
  map(str_split, pattern = ",") %>% 
  modify_depth(2, comb_edge_list) %>% 
  map(., function(x) do.call("rbind", x)) %>% 
  map2(., jean_edges_chap, cbind) %>% 
  map2(., seq_along(jean_edges_chap), cbind) %>% 
  do.call("rbind", .) %>% 
  as_tibble() %>% 
  transmute(Node_1 = V1, Node_2 = V2,
            Chapter = V3, Chapter_n = as.integer(V4))

jean_graph <-
  jean_edges %>% 
  select(Node_1, Node_2) %>% 
  filter(!is.na(Node_1)) %>% 
  as.matrix() %>%
  graph.edgelist(directed = FALSE) %>% 
  as_tbl_graph()

rm(jean, jean_edges_chap, jean_sep, comb_edge_list)