Each of these code chunks takes the data from a page on the website and splits it up into 3 columns split by the first 3 spaces. There were five pages.

library(rvest)
## Loading required package: xml2
library(stringr)
## Warning: package 'stringr' was built under R version 3.4.4
library(lubridate)
library(pander)

events = read_html("https://www.onthisday.com/events/date/2017")

rank_data_html <- html_nodes(events,'.mc-closeModal , .event-list__item')

rank_data <- html_text(rank_data_html)
extdata = data.frame(rank_data)

df1 = data.frame(matrix(ncol = 3, nrow = 161))

for(i in 1:161){
  splits = str_split(extdata[i,], " ")
  df1[i,1] = splits[[1]][1]
  df1[i,2] = splits[[1]][2]
  df1[i,3] = str_flatten(splits[[1]][3:length(splits[[1]])], " ")
}

And here is what it looks like:

pander(head(df1))
X1 X2 X3
NA NA NA
NA NA NA
NA NA NA
NA NA NA
NA NA NA
NA NA NA 
events2 = read_html("https://www.onthisday.com/events/date/2017?p=2")

rank_data_html2 <- html_nodes(events2,'.event-list__item')

rank_data2 <- html_text(rank_data_html2)
extdata2 = data.frame(rank_data2)

df2 = data.frame(matrix(ncol = 3, nrow = 166))

for(i in 1:166){
  splits = str_split(extdata2[i,], " ")
  df2[i,1] = splits[[1]][1]
  df2[i,2] = splits[[1]][2]
  df2[i,3] = str_flatten(splits[[1]][3:length(splits[[1]])], " ")
}
events2 = read_html("https://www.onthisday.com/events/date/2017?p=3")

rank_data_html2 <- html_nodes(events2,'.event-list__item')

rank_data2 <- html_text(rank_data_html2)
extdata2 = data.frame(rank_data2)

df3 = data.frame(matrix(ncol = 3, nrow = 172))

for(i in 1:172){
  splits = str_split(extdata2[i,], " ")
  df3[i,1] = splits[[1]][1]
  df3[i,2] = splits[[1]][2]
  df3[i,3] = str_flatten(splits[[1]][3:length(splits[[1]])], " ")
}
events2 = read_html("https://www.onthisday.com/events/date/2017?p=4")

rank_data_html2 <- html_nodes(events2,'.event-list__item')

rank_data2 <- html_text(rank_data_html2)
extdata2 = data.frame(rank_data2)

df4 = data.frame(matrix(ncol = 3, nrow = 121))

for(i in 1:121){
  splits = str_split(extdata2[i,], " ")
  df4[i,1] = splits[[1]][1]
  df4[i,2] = splits[[1]][2]
  df4[i,3] = str_flatten(splits[[1]][3:length(splits[[1]])], " ")
}
events2 = read_html("https://www.onthisday.com/events/date/2016?p=2")

rank_data_html2 <- html_nodes(events2,'.event-list__item')

rank_data2 <- html_text(rank_data_html2)
extdata2 = data.frame(rank_data2)

df5 = data.frame(matrix(ncol = 3, nrow = 145))

for(i in 1:145){
  splits = str_split(extdata2[i,], " ")
  df5[i,1] = splits[[1]][1]
  df5[i,2] = splits[[1]][2]
  df5[i,3] = str_flatten(splits[[1]][3:length(splits[[1]])], " ")
}

Then I needed to combine the day and month column together with a new year column in order to make a date object.

df5$year = 2016

colnames(df5) = c("month", "day", "event","year")
df5 = df5[c("year", "month", "day","event")]

events2017 = dplyr::bind_rows(df1,df2,df3,df4)

colnames(events2017) = c('month','day','event')
events2017$year = 2017
events2017 = events2017[c("year", "month", "day","event")]

events = dplyr::bind_rows(df5, events2017)


events$date = NA

for(i in 1:765){
  events$date[i] = ymd(str_flatten(events[i,1:3], '-'))
}

events$date = as.Date(events[,"date"], origin="1970-01-01")

events = events[c("date", "event")]

events = events[-c(1:91),]

And here is a peak at the complete csv.

pander(head(events))
  date event
92 NA NA
93 NA NA
94 NA NA
95 NA NA
96 NA NA
97 NA NA