parse_html.r

library(RSelenium)
library(dplyr)
library(rvest)
library(tibble)
#add url
url <- "https://www.toimitilat.fi/toimitilahaku/?size_min=&size_max=&deal_type%5B%5D=1&language=fin&result_type=list&advanced=0&gbl=1&ref=main#searchresult"  

# docker run -d -p 4445:4444 selenium/standalone-firefox:2.53.1
remDr <- remoteDriver(
  remoteServerAddr = "127.0.0.1",
  port = 4445L,
  browserName = "firefox"
)
remDr$open()
remDr$navigate(url)
html_source <- remDr$getPageSource()[[1]]

objects <- read_html(html_source) %>% 
         html_nodes("div.infoCont") 

info <- sapply(objects, function (x) x %>% html_nodes("h4") %>% html_text() %>% gsub("[\n|\t]","",.))

price <- sapply(objects, function (x) x %>% html_nodes("div.priceCont") %>% html_text() %>% gsub("[\n|\t]","",.))

ct <- data_frame(title = info, price = price)