proust_cedilla.R

# Text analysis of Proust

library(proustr)
library(tidytext)
library(dplyr)
library(stringr)

proust_words = proust_books() %>%
    group_by(book) %>%
    mutate(linenumber = row_number(),
           chapter = cumsum(str_detect(text, regex("^chapitre [\\divxlc]",
                                                   ignore_case = TRUE)))) %>%
    ungroup() %>%
    unnest_tokens(word, text)

mot_avec_cedille = proust_words %>%
    filter(grepl("ç", word, fixed = TRUE) |
               grepl("Ç", word, fixed = TRUE)) %>%
    select(word) %>%
    distinct()

proust_words %>%
    inner_join(mot_avec_cedille, by = "word") %>%
    count(word, sort = TRUE)