handle excel sheets #190

NEWS.md

@@ -1,10 +1,14 @@
 # bcdata (development version)
 
+### IMPROVEMENTS
 * Geometry predicates can now take a `bbox` object as well as an `sf*` object (#176)
 * Rename `selectable` column from `bcdc_describe_feature` to `sticky` and modify corresponding docs and tests (#180)
-* Fix `select`, `filter` and `mutate` roxygen so that bcdata specific documentation to these methods is available
 * Add `head` and `tail` methods for `bcdc.promise` objects. Thanks to @hgriesbauer for the suggestion! (#182, #186)
 * Provide `as_tibble` as an alias for `collect` in line with `dbplyr` behaviour (#166)
+* When reading in excel files, `bcdc_get_data` now outputs a messages indicating the presence and names of any sheets (#190)
+
+### BUG FIXES
+* Fix `select`, `filter` and `mutate` roxygen so that bcdata specific documentation to these methods is available
 
 # bcdata 0.1.2
 

R/get_data.R

@@ -26,14 +26,15 @@
 #' @param resource optional argument used when there are multiple data files
 #' within the same record. See examples.
 #' @param ... arguments passed to other functions. Tabular data is passed to a function to handle
-#' the import based on the file extension. `bcdc_read_functions()` provides details on which functions
+#' the import based on the file extension. [bcdc_read_functions()] provides details on which functions
 #' handle the data import. You can then use this information to look at the help pages of those functions.
 #' See the examples for a workflow that illustrates this process.
-#' For spatial Web Service data the `...` arguments are passed to `bcdc_query_geodata()`.
+#' For spatial Web Service data the `...` arguments are passed to [bcdc_query_geodata()].
 #' @param verbose When more than one resource is available for a record,
 #' should extra information about those resources be printed to the console?
 #' Default `TRUE`
 #'
+#'
 #' @return An object of a type relevant to the resource (usually a tibble or an sf object)
 #' @export
 #'
@@ -57,16 +58,20 @@
 #' bcdc_get_data('d7e6c8c7-052f-4f06-b178-74c02c243ea4')
 #'
 #' ## From bcdc_get_record we realize that the data is in xlsx format
-#' bcdc_get_record('d7e6c8c7-052f-4f06-b178-74c02c243ea4')
+#' bcdc_get_record('8620ce82-4943-43c4-9932-40730a0255d6')
 #'
 #' ## bcdc_read_functions let's us know that bcdata
 #' ## uses readxl::read_excel to import xlsx files
 #' bcdc_read_functions()
 #'
-#' ## If you read the help page for readxl::read_excel,
-#' ## it seems likely that we need to skip the first row:
-#' bcdc_get_data('d7e6c8c7-052f-4f06-b178-74c02c243ea4', skip = 1)
+#' ## bcdata let's you know that this resource has
+#' ## multiple worksheets
+#' bcdc_get_data('8620ce82-4943-43c4-9932-40730a0255d6')
+#'
+#' ## we can control what is read in from an excel file
+#' ## using arguments from readxl::read_excel
 #'
+#' bcdc_get_data('8620ce82-4943-43c4-9932-40730a0255d6', sheet = 'Regional Districts')
 #' }
 #'
 #' @export

R/utils.R

@@ -197,7 +197,6 @@ read_from_url <- function(resource, ...){
   if (!reported_format %in% formats_supported()) {
     stop("Reading ", reported_format, " files is not currently supported in bcdata.")
   }
-
   cli <- bcdc_http_client(file_url)
 
   ## Establish where to download file
@@ -220,10 +219,13 @@ read_from_url <- function(resource, ...){
   # where that's not the case
   message("Reading the data using the ", fun$fun, " function from the ",
           fun$package, " package.")
+  handle_excel(tmp, ...)
+
   tryCatch(do.call(fun$fun, list(tmp, ...)),
            error = function(e) {
-             stop("Could not read data set. The file can be found here:\n '",
-                  tmp, "'\n if you would like to try to read it manually.\n\n",
+             stop("Reading the data set failed with the following error message:\n\n  ", e,
+                  "\nThe file can be found here:\n  '",
+                  tmp, "'\nif you would like to try to read it manually.\n",
                   call. = FALSE)
            })
 }
@@ -281,6 +283,21 @@ handle_zip <- function(x) {
   files
 }
 
+handle_excel <- function(tmp, ...) {
+  if (!is_filetype(tmp, c("xls", "xlsx"))) {
+    return(invisible(NULL))
+  }
+
+  sheets <- readxl::excel_sheets(tmp)
+  if (length(sheets) > 1L) {
+    message(paste0("\nThis .", tools::file_ext(tmp), " resource contains the following sheets: \n",
+                   paste0(" '", sheets,"'", collapse = "\n")))
+    if (!methods::hasArg("sheet")) {
+      message("Defaulting to the '", sheets[1], "' sheet. See ?bcdc_get_data for examples on how to specify a sheet.\n")
+    }
+  }
+}
+
 
 unique_temp_dir <- function(pattern = "bcdata_") {
   dir <- tempfile(pattern = pattern)

tests/testthat/test-get-data.R

@@ -100,7 +100,7 @@ test_that("fails informatively when can't read a file", {
   skip_on_cran()
   expect_error(bcdc_get_data(record = '523dce9d-b464-44a5-b733-2022e94546c3',
                              resource = '4cc98644-f6eb-410b-9df0-f9b2beac9717'),
-               "Could not read data set")
+               "Reading the data set failed with the following error message:")
 })
 
 test_that("bcdc_get_data can return the wms resource when it is specified by resource",{
@@ -156,3 +156,11 @@ test_that("bcdc_get_data fails when >1 resource not specified & noninteractive",
                "The record you are trying to access appears to have more than one resource.")
 })
 
+test_that("bcdc_get_data handles sheet name specification", {
+  expect_message(bcdc_get_data('8620ce82-4943-43c4-9932-40730a0255d6'), 'This .xlsx resource contains the following sheets:')
+  expect_error(bcdc_get_data('8620ce82-4943-43c4-9932-40730a0255d6', sheet = "foo"), "Error: Sheet 'foo' not found")
+  out <- capture.output(bcdc_get_data('8620ce82-4943-43c4-9932-40730a0255d6', sheet = "Single Detached"), type = 'message')
+  expect_false(any(grepl('This .xlsx resource contains the following sheets:', out)))
+
+})
+