In benjaminguinaudeau/dockeR:

knitr::opts_chunk$set(echo = TRUE)

Docker must be installed

Linux

# Install docker-compose
sudo curl -L "https://github.com/docker/compose/releases/download/1.23.2/docker-compose-$(uname -s)-$(uname -m)" -o /usr/local/bin/docker-compose
sudo chmod +x /usr/local/bin/docker-compose

Docker run

mkdir ~/my_compose
cd ~/my_compose

nano docker-compose.yml

version: '3'
services:
  rstudio:
    image: guiguiplot/rstudio_perso
    container_name: rstudio
    ports:
      - "8787:8787"
    volumes:
     - "/stock:/mydata"
    links:
      - selenium:selenium
  selenium:
    container_name: chrome
    image: selenium/standalone-chrome-debug
    ports:
      - "32002:4444"
      - "32001:5900"

docker-compose up -d

devtools::install_github("benjaminguinaudeau/dockeR")
pacman::p_load(tidyverse, RSelenium, dockeR)

selenium <- remoteDriver(remoteServerAddr = "selenium", port = 4444L, browserName = "chrome")
selenium$open()

selenium %>% go("https://www.google.ca")

selenium %>% screenshot()
selenium %>% screenshot("test.png")

selenium %>%
  element("q", "name") %>%
  send_keys(list("Paris"))

Extension for googlechrom

Connect to port 32001 ; check whether the port is open if problem on a server.

Connect to the server on port 8787.

docker exec -it rstudio /bin/bash
sudo adduser USER 
# Type Password and enter enter enter
sudo adduser USER sudo
sudo chmod o+w  /usr/local/lib/R/site-library

exit

termId <- rstudioapi::terminalExecute("ls")
rstudioapi::terminalExitCode(termId)


devtools::install_github("benjaminguinaudeau/dockeR")
pacman::p_load(tidyverse, RSelenium, dockeR)

selenium <- remoteDriver(remoteServerAddr = "selenium", port = 4444L, browserName = "chrome")
selenium$open()

selenium %>% go("https://www.google.ca")

selenium %>% screenshot()
selenium %>% screenshot("test.png")

selenium %>%
  element("q", "name") %>%
  send_keys(list("Paris"))

tidy_children <- function(elem, id_name = "id_children", text = F){
  if(class(elem)[[1]] != "list"){elems <- list(elems)}
  elem %>%
    find_children("*", "xpath") %>%
    tidy_elements(id_name = {{id_name}}, text = text)
}


tidy_elements <- function(elems, id_name = "id_parent", text = F){
  if(class(elems)[[1]] != "list"){elems <- list(elems)}
  elems %>%
    map_dfr(get_all_attribute, text = text) %>%
    mutate({{id_name}} := 1:n())
}


has_children <- function(elem){
  le <- elem %>% 
    find_children("*", "xpath") %>%
    length
  return(le != 0)
}

get_attribute_children <- function(elems, depth = 1, text = F){

  out <- tidy_elements(elems, text = text) %>%
    mutate(children_1 = map2(element, id_parent,~{
      .x %>%
        find_children("*", "xpath") %>%
        map_dfr(get_all_attribute) %>%
        rename_all(~paste0(.x, "_1")) %>%
        mutate(id_parent = .y,
               id_children_1 = 1:n())
    }))

  if(depth == 1) return(out)

  for(depth_index in 2:depth){


    child_col <- glue::glue("children_{depth_index}")
    parent_col <- dplyr::sym(paste0("children_", depth_index - 1))
    if(all(out %>% pull({{parent_col}}) %>% map_lgl(~length(.x) == 0))){
      message("Max depth was reached")
      out <- out[,-length(out)]
      break()
    }
    message("Current depth: ", depth_index)

    out <- out %>%
      mutate({{child_col}} := imap({{parent_col}}, ~{
        # message(.y)
        if(length(.x) != 0){
          children_attr <- .x %>%
            split(1:nrow(.)) %>%
            imap_dfr(~{
              if(map_lgl(.x[[paste0("element_", depth_index - 1)]], has_children)){
                # print(.y)
                .x %>%
                  select(paste0("element_", depth_index - 1))  %>%
                  .[[1]] %>% 
                  map_dfr(tidy_children) %>% 
                  rename_all(~paste0(.x, "_", depth_index)) %>%
                  cbind(select(.x, matches("id_((parent)|(children))"))) %>%
                  as_tibble 
              }
            })

          if(nrow(children_attr) != 0){return(children_attr)}
        }
      }))
  }
  return(out)
}




join_elements <- function(x, y){
  joining_vars <- names(x)[str_detect(names(x), "id_(parent)|(children)")]
  message("Joining with ", paste(joining_vars, collapse = " "))
  x <- x %>% right_join(y, by = joining_vars)
  return(x)
}


unnest_children <- function(nested.tbl){

  out <- nested.tbl %>% select(-contains("children"))

  for(children_col in str_subset(names(nested.tbl), "children_")){
    out <- join_elements(out, bind_rows(nested.tbl[[children_col]]))
  }

  return(out)
}

unnest_longer_children <- function(nested.tbl){

  out <- nested.tbl %>% select(-contains("children"))

  for(children_col in str_subset(names(nested.tbl), "children_")){
    tmp <- bind_rows(nested.tbl[[children_col]])

    out <- tmp %>%
      rename_at(vars(-contains("id_parent"), -contains("id_children")), ~str_remove(.x, "(?<!=children)_\\d")) %>%
      mutate(layer = max(as.numeric(str_extract(names(tmp), "\\d+")), na.rm = T)) %>%
      bind_rows(out)

  }

  return(out)
}

data <- out %>%
  select(contains("id_children"), contains("href")) %>%
  mutate_all(as.character) %>%
  mutate(id = as.factor(1:n())) %>%
  pivot_longer(-id) %>%
  separate(name, sep = "_(?=\\d)", into = c("type", "layer")) %>%
  pivot_wider(id_cols = c(id, layer), names_from = type, values_from = value) %>%
  mutate(href = is.na(href)) %>%
  group_by_at(vars(layer, id_children, href)) %>%
  count() %>%
  ungroup %>%
  mutate(id = 1:n()) %>%
  pivot_wider(id_cols = c(id, layer), names_from = id_children, values_from = n)


data %>%
  mutate(id_children_2 = paste(id_children_1, id_children_2, sep = "_")) %>%
  mutate(id_children_3 = paste(id_children_2, id_children_3, sep = "_")) %>%
  mutate(id_children_4 = paste(id_children_3, id_children_4, sep = "_")) %>%
  mutate(id_children_5 = paste(id_children_4, id_children_5, sep = "_")) %>%
  ggforce::gather_set_data(1:5) %>%
  ggplot2::ggplot(ggplot2::aes(x, id = as.factor(id), split = y, value = n)) + 
  ggforce::geom_parallel_sets(aes(fill = n), alpha = 0.3, axis.width = 0.1) + 
  # ggforce::geom_parallel_sets_axes(axis.width = 0.1) + 
  # ggforce::geom_parallel_sets_labels(colour = "white") + 
  ggplot2::theme_classic() + 
  ggplot2::theme(legend.position = "none", 
                 line = ggplot2::element_blank()) + 
  ggplot2::labs(x = "", 
                y = "")

chrome <- chrome_init(name = "chrome")

chrome %>% 
  open %>%
  go("https://scholar.google.de/scholar?q=legislative+speeches&hl=de&as_sdt=0,5")
# go("https://lemonde.fr")


elems <- chrome %>% elements(".gs_rt")  
elem <- elems[[1]]

# t <- elems %>% get_attribute_children(depth = 10, text = T)
# t %>% unnest_children()
# s <- t %>% unnest_longer_children()

# s %>% filter(!is.na(href))
# s %>% count(role)
# s %>% count(text)
# s %>% count(title)
# s %>% 
#   filter(title == "Zitieren") %>%
#   pull(element) %>%
#   get_text()

elems <- chrome %>% elements(".gs_rt")  
bashR::wait(1)
elems %>%
  map(~{
    pos <- chrome %>% get_absolute_location(.x)
    doc_key_down("ctrl")
    doc_mouse_moveTo(x = pos$x, y = pos$y)
    doc_mouse_click(button = "left")
    doc_key_up("ctrl")
    # bashR::wait(1)
  })

doc_mouse_position()

art <- "This article examines how national parties and their members position themselves in European Parliament (EP) debates, estimating the principal latent dimension of spoken conflict using word counts from legislative speeches. We then examine whether the estimated ideal points reflect partisan conflict on a left‚Äìright, European integration or national politics dimension. Using independent measures of national party positions on these three dimensions, we find that the corpus of EP speeches reflects partisan divisions over EU integration and national divisions rather than left‚Äìright politics. These results are robust to both the choice of language used to scale the speeches and to a range of statistical models that account for measurement error of the independent variables and the hierarchical structure of the data." 
art %>%
  str_extract_all("((\\b\\w+\\b )|.){1,30}") %>%.[[1]] %>% str_count("(\\b\\w+\\b )")