Merge pull request #29 from ecohealthalliance/feature/uid_in_spreadsh…

…eets

Feature/uid in spreadsheets

.gitignore

@@ -4,3 +4,4 @@
 .Ruserdata
 inst/doc
 docs
+auth

DESCRIPTION

@@ -1,7 +1,7 @@
 Package: ohcleandat
 Type: Package
 Title: One Health Data Cleaning and Quality Checking Package
-Version: 0.2.3
+Version: 0.2.4
 Authors@R: c(
     person("Collin", "Schwantes", email = "schwantes@ecohealthalliance.org", role = c("cre", "aut"), comment = c(ORCID = "0000-0003-4014-4896")),
     person("Johana", "Teigen", email = "teigen@ecohealthalliance.org", role = "aut", comment = c(ORCID = "0000-0002-6209-2321")),

R/read_excel_all_sheets.R

@@ -4,11 +4,60 @@
 #' For a given excel file, this will detect all sheets, and iteratively read
 #' all sheets and place them in a list.
 #'
-#' @param file character File path to an excel file
+#' If primary keys are added, the primary key is the triplet of the file,
+#' sheet name, and row number e.g. "file_xlsx_sheet1_1". Row numbering is based
+#' on the data ingested into R. R automatically skips empty rows at the beginning
+#' of the spreadsheet so id 1 in the primary key will belong to the first row
+#' with data.
+#'
+#' @note The primary key method is possible because Excel forces sheet names
+#'  to be unique.
+#'
+#' @param add_primary_key_field Logical. Should a primary key field be added?
+#' @param primary_key character. The column name for the unique identifier to be added to the data.
+#' @param file character. File path to an excel file
 #'
 #' @return list
 #' @export
-read_excel_all_sheets <- function(file){
+#'
+#' @examples
+#'  \dontrun{
+#' # Adding primary key field
+#' read_excel_all_sheet(file = "test_pk.xlsx",add_primary_key_field = TRUE)
+#'
+#' # Don't add primary key field
+#' read_excel_all_sheet(file = "test_pk.xlsx")
+#'
+#'     }
+#'
+read_excel_all_sheets <- function(file, add_primary_key_field = FALSE, primary_key = "primary_key"){
   sheets <- readxl::excel_sheets(file)
-  purrr::map(sheets, ~readxl::read_excel(file, sheet = .x))
+
+  if(!add_primary_key_field){
+    out <- purrr::map(sheets, ~readxl::read_excel(file, sheet = .x))
+    return(out)
+  }
+
+  if(add_primary_key_field){
+    purrr::map2(sheets,file,function(sheet,file){
+      df  <- readxl::read_excel(file, sheet = sheet)
+
+      file_name <-  gsub("\\.","_",basename(file))
+      row_ids <- paste(file_name,sheet,1:nrow(df),sep = "_")
+
+      if(primary_key%in%names(df)){
+
+        msg <- sprintf("primary_key - %s - is already a column in the dataframe.
+                     \nPlease choose a column name that isn't present in the data.",primary_key)
+        rlang::abort(msg)
+      }
+
+      out <- df %>%
+        dplyr::mutate({{primary_key}} := {{row_ids}})
+
+      return(out)
+    })
+  }
+
 }
+

R/read_googlesheets.R

@@ -6,36 +6,51 @@
 #' @param sheet Sheet to read, in the sense of "worksheet" or "tab".
 #' @param ss Something that identifies a Google Sheet such as drive id or URL
 #' @param ... other arguments passed to `googlesheets4::range_read()`
+#' @param add_primary_key_field Logical. Should a primary key field be added?
+#' @param primary_key character. The column name for the unique identifier to be added to the data.
 #'
 #' @return tibble
 #' @export
 #' @seealso [googlesheets4::range_read()]
 #' @examples
 #' \dontrun{
-#' read_googlesheets(ss = kzn_animal_ship_sheets, sheet = "all")
+#' read_googlesheets(ss = kzn_animal_ship_sheets, sheet = "all",)
 #' }
 #'
 read_googlesheets <-
   function(key_path,
            sheet = "all",
            ss,
+           add_primary_key_field = FALSE,
+           primary_key = "primary_key",
            ...) {
     # Handle authentication
     googledrive::drive_auth(path = key_path)
     googlesheets4::gs4_auth(path = key_path)
 
-    # decide of all sheets are read and combined, or just one.
+    # decide if all sheets are read and combined, or just one.
     if (sheet == "all") {
       sheet_names <- googlesheets4::sheet_names(ss = ss)
       sheet_names <- rlang::set_names(sheet_names, sheet_names)
       dat <-
         purrr::map(sheet_names,
-                   \(x) googlesheets4::range_read(
-                     ss = ss,
-                     sheet = x,
-                     na = c("", "NA", "NULL", "-"),
-                     ...
-                   ))
+                   function(x){
+                    df  <- googlesheets4::range_read(
+                       ss = ss,
+                       sheet = x,
+                       na = c("", "NA", "NULL", "-"),
+                       ...
+                     )
+
+                    if(add_primary_key_field){
+                      #ss_sheet_rownum
+                      drive_id <- googledrive::as_id(ss)
+                      row_ids <- paste(drive_id,x,1:nrow(df),sep = "_")
+                      df <- df %>%
+                        dplyr::mutate({{primary_key}} := {{row_ids}})
+                    }
+                    return(df)
+                   } )
 
     } else{
       dat <-
@@ -46,6 +61,14 @@ read_googlesheets <-
           ...
         )
 
+      if(add_primary_key_field){
+        #ss_sheet_rownum
+        drive_id <- googledrive::as_id(ss)
+        row_ids <- paste(drive_id,sheet,1:nrow(dat),sep = "_")
+        dat <- dat %>%
+          dplyr::mutate({{primary_key}} := {{row_ids}})
+      }
+
     }
 
     return(dat)