R/datexp_import.R

In NicolasJBM/datexp: Interface to facilitate data exploration

#' Function to import and concatenate datasets having the same format.
#' @param x           Tibble for csv, excel, and text; Character vector for google.
#' @param spreadsheet Character string. For excel and google, specify the name of the spreadsheet to be imported.
#' @param filetype    Character string. specify whether "", "", "", or "" are imported.
#' @return A tibble contaning the name of the source and all the related segment(s).
#' @importFrom tibble as_tibble
#' @importFrom tibble tibble
#' @importFrom tibble rownames_to_column
#' @importFrom purrr map
#' @importFrom purrr safely
#' @importFrom stringr str_sub
#' @importFrom dplyr mutate
#' @importFrom dplyr select
#' @importFrom dplyr filter
#' @importFrom dplyr mutate_all
#' @importFrom tidyr unnest
#' @importFrom RefManageR ReadBib
#' @importFrom googlesheets gs_title
#' @importFrom googlesheets gs_read_csv
#' @importFrom readxl read_excel
#' @importFrom utils read.delim
#' @importFrom readr read_csv
#' @export


datexp_import <- function(x,
                          spreadsheet = 1,
                          filetype) {
  
  # Bind variables
  name <- NULL
  datapath <- NULL
  extension <- NULL
  data <- NULL
  notimported <- NULL
  V1 <- NULL
  line <- NULL
  
  
  if(length(x)>1 & filetype != "google") x <- as_tibble(x) else {
    x <- tibble(name = x, datapath = x)
  }
  
  if (filetype == "csv") {
    safe_read <- safely(read_csv, quiet = T)
    import <- x %>%
      select(name, datapath) %>%
      mutate(extension = str_sub(datapath, -4)) %>%
      filter(extension == ".csv") %>%
      select(-extension) %>%
      mutate(data = map(datapath, safe_read)) %>%
      mutate(data = map(data, function(x) mutate_all(x$result, as.character))) %>%
      mutate(notimported = map(data, is.null)) %>%
      filter(notimported == FALSE) %>%
      select(-notimported) %>%
      unnest()
  }

  if (filetype == "excel") {
    safe_read <- safely(read_excel, quiet = T)
    import <- x %>%
      select(name, datapath) %>%
      mutate(extension = str_sub(datapath, -5)) %>%
      filter(extension == ".xlsx") %>%
      select(-extension) %>%
      mutate(data = map(datapath, safe_read, sheet = spreadsheet)) %>%
      mutate(data = map(data, function(x) mutate_all(x$result, as.character))) %>%
      mutate(notimported = map(data, is.null)) %>%
      filter(notimported == FALSE) %>%
      select(-notimported) %>%
      unnest()
  }

  if (filetype == "google") {
    safe_title <- safely(gs_title, quiet = T)
    safe_read <- safely(gs_read_csv, quiet = T)
    import <- x %>%
      mutate(data = map(datapath, safe_title)) %>%
      mutate(data = map(data, function(x) x$result)) %>%
      mutate(notimported = map(data, is.null)) %>%
      filter(notimported == FALSE) %>%
      select(-notimported) %>%
      mutate(data = map(data, safe_read, ws = spreadsheet, col_types = cols(.default = "c"))) %>%
      mutate(data = map(data, function(x) mutate_all(x$result, as.character))) %>%
      mutate(notimported = map(data, is.null)) %>%
      filter(notimported == FALSE) %>%
      select(-notimported) %>%
      unnest()
  }

  if (filetype == "bibtex") {
    read <- function(x) {
      ReadBib(x) %>% 
        as.data.frame() %>%
        rownames_to_column("refid") %>%
        as_tibble()
    }
    safe_read <- safely(read, quiet = T)
    import <- x %>%
      select(name, datapath) %>%
      mutate(extension = str_sub(datapath, -4)) %>%
      filter(extension == ".bib") %>%
      select(-extension) %>%
      mutate(data = map(datapath, safe_read)) %>%
      mutate(data = map(data, function(x) mutate_all(x$result, as.character))) %>%
      mutate(notimported = map(data, is.null)) %>%
      filter(notimported == FALSE) %>%
      select(-notimported) %>%
      unnest()
  }

  if (filetype == "text") {
    options(warn = -1)
    safe_read <- safely(read.delim, quiet = T)
    import <- x %>%
      select(name, datapath) %>%
      mutate(extension = str_sub(datapath, -4)) %>%
      filter(extension == ".txt") %>%
      select(-extension) %>%
      mutate(data = map(datapath, safe_read, header = F)) %>%
      mutate(data = map(data, function(x) mutate_all(x$result, as.character))) %>%
      mutate(notimported = map(data, is.null)) %>%
      filter(notimported == FALSE) %>%
      select(-notimported) %>%
      mutate(data = map(data, function(x)
        x %>% rownames_to_column("line") %>%
          mutate(V1 = as.character(x$V1)))
      ) %>%
      unnest() %>%
      select(source = name, line, segments = V1)
  }

  import
}