R/cleaning_strings_and_renaming_columns.R

In BillyEhrenberg/FTutilityfuncs:

  #' A function to clean names to remove funny characters from strings
  #'
  #' Swaps whitespaces for underscores, retains underscores, removes special/accented characters
  #' @param string a string
  #' @export
  #' @examples
  #' cleanNames()



cleanNames <- function(string){
  string %<>% gsub('\\s','AWHITESPACE',.) %>% gsub('\\_','ANUNDERSCORE',.) %>%  iconv(.,to='ASCII//translit') %>%
    gsub('[^[:alnum:]]','',.) %>% gsub('AWHITESPACE|ANUNDERSCORE','_',.) %>%
    tolower
}


#' A function to clean column names in a df
#'
#'Swaps whitespaces for underscores, retains underscores, removes special/accented characters
#' @param df a df or tibble
#' @export
#' @examples
#' clean_col_Names()

clean_col_Names <- function(df){

  names(df) = tolower(cleanNames(names(df) %>% stringr::str_replace('/','_')))

  df

}


#' A function to merge multi-row column headers into one row
#'
#' Choose a first and last row, all rows between will be merged
#' @param df a data.frame or tibble
#' @param first_row the first row to merge
#' @param last_row the last row to merge
#' @param fill_gaps_in_rows indices of rows with missing values that you want filled.
#' @export
#' @examples
#' merge_col_headers()

merge_col_headers <- function(df, first_row, last_row, fill_gaps_in_rows = NULL){


  if(!is.null(fill_gaps_in_rows)){


    df <- fill_gaps_in_rows %>%
      purrr::map_dfr(~{

        if(is.na(df[1,1])){
          df[1,1] = '||holding_value||'
        }

        df[1,] = df[1,] %>%
          as.character %>%
          zoo::na.locf(.) %>%
          ifelse(. == '||holding_value||',
                 NA,
                 .)

        df

      })

  }

  headers <- df %>%
    .[first_row:last_row,]



  new_names <- names(headers) %>%
    purrr::map_chr(~{
      pull(headers,.x) %>%
        .[!is.na(.)] %>%
        myCollapse('_') %>%
        cleanNames()
    })

  df %>%
    .[(last_row+1):nrow(.),] %>%
    purrr::set_names(new_names)


}