R/get711.R

In leoluyi/CRMaddress: Crawling address in Taiwan

#' 7-ELEVEN
#'
#' 7-11 shop address.
#'
#' @return data.table
#' @export
get711 <- function() {
  url <- "http://emap.pcsc.com.tw/EMapSDK.aspx"
  res <- GET(url,
             query = list(
               commandid = "SearchStore",
               StoreName = "%"
             ))
  res_xml <- res %>%
    content(encoding = "UTF-8") %>%
    XML::xmlParse(encoding = "UTF-8", options = XML::HUGE) %>%
    .['//GeoPosition']
  dt_711 <- res_xml %>%
    XML::xmlToDataFrame(stringsAsFactors = FALSE) %>%
    as.data.table()

  ## Data cleansing
  dt_711[
    , c("X", "Y") := lapply(.SD,
                            function(x) {
                              as.numeric(as.character(x))*10^(-6)
                            }),
    .SDcols = c("X", "Y")]

  ## add brand name, keyword
  dt_711[, `:=`(brand_nm = "統一超商", keyword = "7-11")]

  ## rename variables
  setnames(dt_711,
           c("Address", "POIName", "Telno", "X", "Y"),
           c("addr", "store_nm", "tel_no", "lon_x", "lat_y"))

  # names(dt_711) %>% str_replace_all("([a-z])([A-Z])", "\\1_\\L\\2")
  names(dt_711) <- names(dt_711) %>%
    str_replace_all("([a-z])([A-Z])", "\\1_\\2") %>%
    str_to_lower()

  ## add url, time, key
  try({setnames(dt_711,"is7wi_fi", "is_7wifi")}, silent = T)
  dt_711[, store_url := "http://emap.pcsc.com.tw/emap.aspx"]
  dt_711[, data_time := format(Sys.time(), "%Y-%m-%dT%H:%M:%S%z")]
  dt_711[, full_nm := paste0(brand_nm, store_nm)]


  ## move brand_nm, key_word to first two colmuns
  setcolorder(dt_711, c("full_nm", "brand_nm", "keyword", "store_nm", "addr",
                        setdiff(names(dt_711), c("full_nm", "brand_nm", "keyword", "store_nm", "addr"))))

  dt_711
}