Google Forms Data Extraction

NAMESPACE

@@ -15,13 +15,13 @@ export(get_ga_user)
 export(get_github)
 export(get_github_metrics)
 export(get_github_user)
-export(get_google_files)
 export(get_repo_list)
 export(get_repos_metrics)
 export(get_youtube_stats)
 export(gh_repo_wrapper)
 export(list_calendly_events)
 export(request_ga)
+export(request_google_forms)
 importFrom(assertthat,assert_that)
 importFrom(assertthat,is.string)
 importFrom(dplyr,"%>%")

R/auth.R

@@ -71,6 +71,7 @@ authorize <- function(app_name = NULL,
       scope = scopes_list,
       ...
     )
+    googledrive::drive_auth(token = token)
   }
   set_token(token = token, app_name = app_name)
   return(invisible(token))

R/google-forms.R

@@ -2,9 +2,251 @@
 
 #' Get Google Forms
 #' @description This is a function to get the Calendly API user info
-#' @param token You can provide the a googledrive link that you'd like to narrow the search to
-#' @return Calendly REST API response as a list
-#' @importFrom utils menu installed.packages
-#' @importFrom httr oauth_app oauth_endpoints oauth2.0_token
+#' @param url The endpoint URL for the request
+#' @param token credentials for access to Google using OAuth. `authorize("google")`
+#' @param body_params The body parameters for the request
+#' @param query_params The body parameters for the request
+#' @param return_request Should a list of the request be returned as well?
+#' @importFrom httr config accept_json content
+#' @importFrom jsonlite fromJSON
+#' @importFrom assertthat assert_that is.string
 #' @export
-#' @examples
+request_google_forms <- function(token, url, query = NULL, body_params = NULL, query_params = NULL,
+                                 return_request = TRUE) {
+  if (is.null(token)) {
+    # Get auth token
+    token <- get_token(app_name = "google")
+  }
+  config <- httr::config(token = token)
+
+  result <- httr::GET(
+    url = url,
+    body = body_params,
+    query = query,
+    config = config,
+    httr::accept_json(),
+    encode = "json"
+  )
+
+  request_info <- list(
+    url = url,
+    token = token,
+    body_params = body_params,
+    query_params = query_params
+  )
+
+  if (httr::status_code(result) != 200) {
+    httr::stop_for_status(result)
+    return(result)
+  }
+
+  # Process and return results
+  result_content <- httr::content(result, "text")
+  result_list <- jsonlite::fromJSON(result_content)
+
+  if (return_request) {
+    return(list(result = result_list, request_info = request_info))
+  } else {
+    return(result_list)
+  }
+}
+
+
+#' Get Google Forms
+#' @description This is a function to get the Calendly API user info
+#' @param form_id The form ID we need to get
+#' @param token credentials for access to Google using OAuth. `authorize("google")`
+#' @param dataformat What format would you like the data? Options are "raw" or "dataframe". "dataframe" is the default.
+#' @importFrom httr config accept_json content
+#' @importFrom jsonlite fromJSON
+#' @importFrom assertthat assert_that is.string
+#' @examples \dontrun{
+#'
+#' authorize("google")
+#' form_info <- get_google_form("https://docs.google.com/forms/d/1Z-lMMdUyubUqIvaSXeDu1tlB7_QpNTzOk3kfzjP2Uuo/edit")
+#'
+#' ### OR You can give it a direct form id
+#' form_info <- get_google_form("1Z-lMMdUyubUqIvaSXeDu1tlB7_QpNTzOk3kfzjP2Uuo")
+#' }
+get_google_form <- function(form_id, token = NULL, dataformat = "dataframe") {
+  # If a URL is supplied, only take the ID from it.
+  if (grepl("https:", form_id[1])) {
+    form_id <- gsub("\\/viewform$|\\/edit$", "", form_id)
+    form_id <- gsub("https://docs.google.com/forms/d/e/|https://docs.google.com/forms/d/", "", form_id)
+  }
+
+  form_info_url <- gsub("\\{formId\\}", form_id, "https://forms.googleapis.com/v1/forms/{formId}")
+  form_response_url <- gsub("\\{formId\\}", form_id, "https://forms.googleapis.com/v1/forms/{formId}/responses")
+
+  message(paste0("Trying to grab form: ", form_id))
+
+  form_info <- request_google_forms(
+    url = form_info_url,
+    token = token
+  )
+
+  response_info <- request_google_forms(
+    url = form_response_url,
+    token = token
+  )
+
+  result <- list(
+    form_metadata = form_info,
+    response_info = response_info
+  )
+
+  if (dataformat == "dataframe") {
+
+    metadata <- get_question_metadata(form_info)
+
+    if (length(result$response_info$result) > 0) {
+      answers_df <- extract_answers(result)
+    } else {
+      answers_df <- "no responses yet"
+    }
+    return(list(title = result$form_metadata$result$info$title,
+                metadata = metadata,
+                answers = answers_df))
+  }
+  return(result)
+}
+
+
+#' Get multiple Google forms
+#' @description This is a function to get the Calendly API user info
+#' @param form_ids a vector of form ids you'd like to retrieve information for
+#' @param token credentials for access to Google using OAuth. `authorize("google")`
+#' @importFrom purrr map
+#' @importFrom janitor make_clean_names
+#' @examples \dontrun{
+#'
+#' authorize("google")
+#' form_list <- googledrive::drive_find(shared_drive = googledrive::as_id("0AJb5Zemj0AAkUk9PVA"), type = "form")
+#'
+#' multiple_forms <- get_multiple_forms(form_ids = form_list$id)
+#' }
+get_multiple_forms <- function(form_ids = NULL, token = NULL) {
+
+  # Get all the forms info
+  all_form_info <- sapply(form_ids, function(form_id) {
+    get_google_form(
+      form_id = form_id,
+      token = token
+    )
+  }, simplify = FALSE, USE.NAMES = TRUE)
+
+
+  # Set up the names
+  titles <- purrr::map(all_form_info, ~ .x$title)
+  titles <- janitor::make_clean_names(titles)
+
+  # Set as names
+  names(all_form_info) <- titles
+
+  all_form_info
+}
+
+get_question_metadata <- function(form_info) {
+  metadata <- data.frame(
+    question_id = form_info$result$items$itemId,
+    title = form_info$result$items$title
+  )
+
+  if (length(form_info$result$items$questionItem$question$textQuestion) > 0) {
+    metadata <- data.frame(
+      metadata,
+      paragraph = form_info$result$items$questionItem$question$textQuestion
+    )
+  }
+  if (length(form_info$result$items$questionItem$question$choiceQuestion$type) > 0) {
+    metadata <- data.frame(
+      metadata,
+      choice_question = form_info$result$items$questionItem$question$choiceQuestion$type,
+      text_question = is.na(form_info$result$items$questionItem$question$choiceQuestion$type)
+    )
+  }
+
+  return(metadata)
+}
+
+extract_answers <- function(form_info) {
+  questions <- form_info$response_info$result$responses$answers
+
+  if (length(questions) > 0) {
+    # Extract the bits we want
+    answers <- purrr::map(
+      questions,
+      ~ .x$textAnswers$answers
+    )
+
+    question_id <- purrr::map(
+      questions,
+      ~ .x$questionId
+    )
+
+    # Reformat the answer info
+    answers <- purrr::map_depth(answers, 2, ~ ifelse(is.null(.x),
+      data.frame(value = "NA"),
+      .x
+    ))
+
+    answers <- purrr::map_depth(answers, -1, ~ ifelse(length(.x) > 1,
+      paste0(.x, collapse = "|"),
+      .x
+    ))
+    answers <- lapply(answers, purrr::map, -1)
+
+    # Turn into data frames
+    answers_df <- lapply(answers, paste0) %>% dplyr::bind_cols()
+    question_df <- lapply(question_id, paste0) %>% dplyr::bind_cols()
+
+    colnames(answers_df) <- paste0(colnames(answers_df), "_answers")
+    colnames(question_df) <- paste0(colnames(question_df), "_question")
+
+    # Put it all in a data.frame we will keep
+    info_df <- data.frame(
+      reponse_id = rep(form_info$response_info$result$responses$responseId, length(questions)),
+      answers_df,
+      question_df
+    )
+  } else {
+    info_df <- data.frame(value = "no responses yet")
+  }
+
+  return(info_df)
+}
+
+
+google_pagination <- function(first_page_result) {
+  # Set up a while loop for us to store the multiple page requests in
+  cummulative_pages <- first_page_result$result$files
+  page <- 1
+
+  next_pg <- try(next_google(first_page_result), silent = TRUE)
+
+  while (!grepl("Error", next_pg$result[1])) {
+    cummulative_pages <- dplyr::bind_rows(cummulative_pages, next_pg$result$files)
+    next_pg <- try(next_google(first_page_result), silent = TRUE)
+    page <- page + 1
+  }
+  return(cummulative_pages)
+}
+
+
+next_google <- function(page_result) {
+  ## TODO: Next page request is not working! Not sure why. It doesn't throw an error,
+  ## but it just gives the same result everytime!
+  body_params <- c(page_result$request_info$body_params,
+    pageToken = page_result$result$nextPageToken
+  )
+
+  result <- request_google_forms(
+    token = token,
+    url = url,
+    body_params = body_params,
+    query_params = query_params,
+    return_request = TRUE
+  )
+
+  return(result)
+}

R/token-handlers.R

@@ -68,7 +68,8 @@ get_token <- function(app_name, try = FALSE) {
 get_stored_token <- function(app_name) {
   if (app_name == "calendly") token <- getOption("calendly")
   if (app_name == "github") token <- getOption("github")
-  if (app_name == "google") token <- try(readRDS(file.path(cache_secrets_folder(), "google.RDS")), silent = TRUE)
+  if (app_name == "google") token <- getOption("google")
+
   return(token)
 }
 

R/utils.R

@@ -1,5 +1,5 @@
 utils::globalVariables(c(
-   "result", "num", "test_name", "scopes", "set_token", "browseURL", "remove_token", "get_token", "get_github", "get_calendly", "%>%"
+  "result", "num", "test_name", "scopes", "set_token", "browseURL", "remove_token", "get_token", "get_github", "get_calendly", "%>%"
 ))
 #' Supported endpoints
 #' @description This is function stores endpoints and supported app names