Merge pull request #24 from nih-cfde/23-bespoke-function-to-create-excel-spreadsheet-of-results

Export Tabular Results

Author: the-mayer

.Rbuildignore

@@ -11,4 +11,5 @@ terraform/
 DEVELOPER.md
 ^README\.Rmd$
 ^CODE_OF_CONDUCT\.md$
+.xlsx$
 

.gitignore

@@ -5,3 +5,5 @@ docs
 terraform/
 /.quarto/
 **/*.quarto_ipynb
+inst/secret/cfde-access-keyfile.json
+*.xlsx

DESCRIPTION

@@ -1,9 +1,11 @@
 Package: programets
 Title: Collect and analyze academic impact metrics from various sources
-Version: 0.3.0
+Version: 0.3.1
 Authors@R: 
-    c(person("Sean", "Davis", , "[email protected]", role = c("aut", "cre"), comment=c(ORCID = "0000-0002-8991-6458")),
-    person("David", "Mayer", , "[email protected]", role = c("aut")))
+    c(
+        person("Sean", "Davis", , "[email protected]", role = c("aut", "cre"), comment=c(ORCID = "0000-0002-8991-6458")),
+        person("David", "Mayer", , "[email protected]", role = c("aut"), comment=c(ORCID = "0000-0002-6056-9771"))
+        )
 Description: 
     Collect and analyze academic impact metrics from various sources. 
     The package provides functions to collect and analyze data from NIH reporter, 
@@ -28,7 +30,10 @@ Imports:
     tibble,
     glue,
     ghql,
-    rvest
+    openxlsx,
+    tidyr,
+    rvest,
+    readr
 Suggests:
     gargle,
     gitcreds,

DEVELOPER.md

@@ -38,9 +38,9 @@ Step 3: Encrypt the google service account json file
 
 ```{r}
 gargle::secret_encrypt_json(
-  path = "path/to/ga4-acess-keyfile.json",
+  path = "path/to/ga4-access-keyfile.json",
   key = "GARGLE_ENCRYPTION_KEY",
-  output = "inst/secret/ga4-acess-keyfile.json"
+  output = "inst/secret/ga4-access-keyfile.json"
 )
 ```
 This will create an encrypted version of the json file in the `inst/secret` directory.
@@ -51,7 +51,7 @@ Step 4: Use the encrypted file in your code
 library(gargle)
 googleAnalyticsR::ga_auth(
   json_file = gargle::secret_decrypt_json(
-    path = system.file("secret/ga4-acess-keyfile.json", package = "programets"),
+    path = system.file("secret/ga4-access-keyfile.json", package = "programets"),
     key = "GARGLE_ENCRYPTION_KEY"
   )
 )

NAMESPACE

@@ -2,10 +2,13 @@
 
 export(cfde_opportunity_numbers)
 export(epmc_search)
+export(export_tabular)
 export(ga_dataframe)
 export(ga_meta_simple)
 export(ga_query_explorer)
 export(get_core_project_info)
+export(get_ga_basic)
+export(get_ga_meta_by_id)
 export(get_github_by_topic)
 export(get_github_by_topic_graphql)
 export(icite)
@@ -23,6 +26,8 @@ importFrom(dplyr,tibble)
 importFrom(ghql,GraphqlClient)
 importFrom(glue,glue)
 importFrom(glue,glue_collapse)
+importFrom(googleAnalyticsR,ga_account_list)
+importFrom(googleAnalyticsR,ga_auth)
 importFrom(httr2,req_auth_bearer_token)
 importFrom(httr2,req_body_json)
 importFrom(httr2,req_error)
@@ -36,12 +41,18 @@ importFrom(httr2,request)
 importFrom(httr2,resp_body_json)
 importFrom(httr2,resp_status)
 importFrom(jsonlite,fromJSON)
+importFrom(openxlsx,addWorksheet)
+importFrom(openxlsx,createWorkbook)
+importFrom(openxlsx,saveWorkbook)
+importFrom(openxlsx,writeData)
 importFrom(purrr,discard)
+importFrom(purrr,map)
 importFrom(purrr,map2_dbl)
 importFrom(purrr,map_chr)
 importFrom(purrr,map_dbl)
 importFrom(purrr,map_dfr)
 importFrom(purrr,pmap)
+importFrom(readr,write_csv)
 importFrom(rlang,"%||%")
 importFrom(rlang,.data)
 importFrom(rlang,abort)
@@ -54,4 +65,6 @@ importFrom(stats,na.omit)
 importFrom(stats,setNames)
 importFrom(stringr,regex)
 importFrom(stringr,str_detect)
+importFrom(stringr,str_remove)
 importFrom(tibble,tibble)
+importFrom(tidyr,separate)

R/export.R

@@ -0,0 +1,61 @@
+#' Export to Tabular
+#' 
+#' @param core_project_numbers A character vector of NIH Core Project Numbers
+#' @param token The token required for authentication with the GitHub API
+#' @param service_account_json A character string containing the path to a JSON file containing a Google service account
+#' @param dir A character string containing the path to directory where the Excel file will be written
+#' @param csv A logical indicating whether to write a CSV file
+#' 
+#' @importFrom openxlsx createWorkbook addWorksheet writeData saveWorkbook
+#' @importFrom readr write_csv
+#' @importFrom rlang .data
+#' @export
+#' 
+#' @examples
+#' \dontrun{
+#' test_projects <-c("OT2OD030545")
+#' }
+#' 
+export_tabular <- function(core_project_numbers, token = gitcreds::gitcreds_get()$password, service_account_json = 'cfde-access-keyfile.json', dir, csv = FALSE) {
+
+  ## Create Excel Workbook
+  wb <- createWorkbook()
+
+  ## Add NIH Project Info
+  addWorksheet(wb, "project_info")
+  proj_info <- get_core_project_info(core_project_numbers)
+  writeData(wb = wb, sheet = "project_info", x = proj_info, na.string = "")
+  if (csv) {
+    write_csv(proj_info, file.path(dir, paste0("programets_proj_info_", Sys.Date(), ".csv", sep = "")))
+  }
+  
+  ## Add Assosciated Publications
+  addWorksheet(wb, "pub_info")
+  pmids <- proj_info |> 
+    filter(.data$found_publication) |> 
+    pull('pmid')
+  pub_info <- icite(pmids)
+  writeData(wb = wb, sheet = "pub_info", x = pub_info, na.string = "")
+  if (csv) {
+    write_csv(pub_info, file.path(dir, paste0("programets_pub_info_", Sys.Date(), ".csv", sep = "")))
+  }
+
+  ## Add GitHub
+  addWorksheet(wb, "github_info")
+  github_info <- get_github_by_topic_graphql(core_project_numbers, token = token)
+  writeData(wb = wb, sheet = "github_info", x = github_info, na.string = "")
+  if (csv) {
+    write_csv(github_info, file.path(dir, paste0("programets_github_info_", Sys.Date(), ".csv", sep = "")))
+  }
+
+  ## Add Google Analytics
+  addWorksheet(wb, "ga_info")
+  ga_info <- get_ga_basic(core_project_numbers = core_project_numbers, service_account_json = service_account_json)
+  writeData(wb = wb, sheet = "ga_info", x = ga_info, na.string = "")
+  if (csv) {
+    write_csv(ga_info, file.path(dir, paste0("programets_ga_info_", Sys.Date(), ".csv", sep = "")))
+  }
+
+  ## Save Workbook
+  saveWorkbook(wb, file.path(dir, paste0("programets_", Sys.Date(), ".xlsx", sep = "")))
+}

R/ga_meta_simple.R

@@ -26,4 +26,37 @@
 #' @export
 ga_meta_simple <- function() {
   tibble::as_tibble(googleAnalyticsR::ga_meta(version = "data"))
+}
+
+#' Google Analytics metadata dataframe by property ID
+#'
+#' This function retrieves Google Analytics metadata by property ID
+#' and returns it as a dataframe. The metadata includes
+#' information about metrics, dimensions, and other
+#' attributes available in Google Analytics.
+#'
+#' This function is a wrapper around the
+#' `googleAnalyticsR::ga_meta()` function. It retrieves
+#' metadata for the Google Analytics API version 4.
+#'
+#' @param property_id The property ID for which to retrieve
+#' metadata.
+#'
+#' @note This function requires first authenticating to
+#' Google Analytics using the `ga_auth()` function.
+#'
+#' @family Google Analytics
+#'
+#' @examples
+#' \dontrun{
+#' res = get_ga_meta_by_id("123456789")
+#' head(res)
+#' dplyr::glimpse(res)
+#' }
+#'
+#' @return A tibble containing Google Analytics metadata.
+#'
+#' @export
+get_ga_meta_by_id <- function(property_id) {
+  tibble::as_tibble(googleAnalyticsR::ga_meta(version = "data", propertyId = property_id))
 }

R/get_ga_basic.R

@@ -0,0 +1,74 @@
+#' Get Basic Google Analytics Info
+#'
+#' This function takes a character vector of NIH Core Project Numbers and 
+#' returns a data frame containing the any Google Analytics properties associated
+#' with the Core Project Numbers.
+#'
+#' @param core_project_numbers A character vector of NIH Core Project Numbers
+#' @param service_account_json A character string containing the path to a JSON file containing the 
+#' Google service account credentials. If no file is provided, interactive authentication is used. 
+#' Defaults to "cfde-access-keyfile.json"
+#' 
+#' @importFrom googleAnalyticsR ga_account_list ga_auth
+#' @importFrom purrr map map_chr
+#' @importFrom stringr str_remove
+#' @importFrom tidyr separate
+#' @importFrom rlang .data
+#' 
+#' @return A data frame containing the associated Google Analytics data
+#' @export
+get_ga_basic <- function(core_project_numbers, service_account_json = 'cfde-access-keyfile.json') {
+  ## This function requires authentication, check for existing creds
+  ## Package Credentials
+  if(file.exists(system.file("secret", service_account_json, package = "programets")) &&
+    !is.null(Sys.getenv("CFDE_ENCRYPTION_KEY"))){
+    programets_service_account <- gargle::secret_decrypt_json(
+      path = system.file(
+        "secret",
+        service_account_json,
+        package = "programets"
+      ),
+      key = "CFDE_ENCRYPTION_KEY"
+    )
+    googleAnalyticsR::ga_auth(
+      json_file = programets_service_account
+    )
+  ## User SA Credentials
+  } else if (file.exists(service_account_json)) {
+    ga_auth(json_file = service_account_json)
+  ## Interactive Auth
+  } else {
+    ga_auth()
+  }
+  
+  ## Get All Analytics Properties
+  core_project_regex <- paste0(unique(tolower(core_project_numbers)), collapse = "|")
+  account_list <- ga_account_list("ga4") |> 
+    mutate(
+      property_meta = suppressMessages(map(.data$propertyId, get_ga_meta_by_id)),
+      core_project_num = map_chr(
+        .data$property_meta,
+        ~{
+          res <- .x |>
+            filter(str_detect(apiName, regex(core_project_regex, ignore_case = TRUE))) |>
+            tidyr::separate(apiName, into = c("api", "value"), sep = ":", remove = FALSE) |>
+            pull(value)
+          if (length(res) == 0) {
+            NA_character_
+          } else {
+            res |>
+              str_remove("^cfde_") |>
+              unique() |>
+              paste(collapse = ",")
+          }
+        }
+      )
+    ) |> 
+    ## Filter to those with the requested Core Project Numbers
+    filter(!is.na(.data$core_project_num)) |> 
+    select(-'property_meta')
+  if(nrow(account_list) == 0) {
+    rlang::inform(rlang::format_error_bullets(c(i = "No Google Analytics properties found for the requested Core Project Numbers")))
+  }
+  return(account_list)
+  }

R/nih_reporter.R

@@ -215,8 +215,8 @@ get_core_project_info <- function(core_project_numbers) {
     proj_results_tbl |> 
     full_join(all_results, by = c('core_project_num' = 'core_project_number', 'appl_id' = 'applid')) |> 
     filter(.data$core_project_num != "") |>
-    relocate(.data$core_project_num, .before = .data$appl_id) |> 
-    relocate(.data$found_publication, .after = .data$core_project_num)
+    relocate('core_project_num', .before = 'appl_id') |> 
+    relocate('found_publication', .after = 'core_project_num')
 
   return(all_results_combined)
 }