Merge pull request #128 from rage/bundle-r-testrunner

bundled tmc-r-testrunner for tests to speed up CI

Author: Heliozoa

.github/workflows/test.yml

@@ -27,8 +27,8 @@ jobs:
           dotnet-version: "5.0.x"
       - name: Install tmc-r-tester
         run: |
-          Rscript -e 'install.packages(c("devtools","testthat", "httr", "curl"),repos="https://ftp.eenet.ee/pub/cran/")'
-          Rscript -e 'devtools::install_github("RTMC/tmc-r-tester/tmcRtestrunner")'
+          Rscript -e 'install.packages(c("testthat", "jsonlite", "R.utils"))'
+          Rscript -e 'install.packages("plugins/r/tests/tmcRtestrunner", repos=NULL, type="source")'
       - name: Build test binary
         run: cargo test --no-run --verbose
       - name: Run tests

plugins/r/tests/tmcRtestrunner/.Rbuildignore

@@ -0,0 +1,2 @@
+^.*\.Rproj$
+^\.Rproj\.user$

plugins/r/tests/tmcRtestrunner/.lintr

@@ -0,0 +1,9 @@
+linters: with_defaults(
+    absolute_paths_linter = NULL, # 21
+    closed_curly_linter = NULL, # 2
+    open_curly_linter = NULL, # 2
+    single_quotes_linter = NULL, # 1
+    NULL,
+    line_length_linter(120),
+    object_length_linter(80)
+  )

plugins/r/tests/tmcRtestrunner/DESCRIPTION

@@ -0,0 +1,27 @@
+Package: tmcRtestrunner
+Type: Package
+Title: Runs TMC R projects
+Version: 0.1.0
+Author: RTMC
+Maintainer: The package maintainer <yourself@somewhere.net>
+Description: More about what it does (maybe more than one line)
+    Use four spaces when indenting paragraphs within the Description.
+License: What license is it under?
+Encoding: UTF-8
+LazyData: true
+Depends:
+  testthat,
+  jsonlite,
+  R.utils
+Imports:
+  testthat,
+  jsonlite,
+  R.utils
+Suggests:
+  lintr
+Collate:
+  'ResultsJsonParser.R'
+  'ResultsCreator.R'
+  'RunTests.R'
+  'GetAvailablePoints.R'
+  'TestEnvironment.R'

plugins/r/tests/tmcRtestrunner/NAMESPACE

@@ -0,0 +1,4 @@
+exportPattern("^[[:alpha:]]+")
+import(testthat)
+import(jsonlite)
+import(R.utils)

plugins/r/tests/tmcRtestrunner/R/GetAvailablePoints.R

@@ -0,0 +1,59 @@
+
+.get_available_points <- function(project_path) {
+  .init_global_vars()
+  test_files <- list.files(path = paste0(project_path, "/tests/testthat"), pattern = "test.*\\.R",
+                                         full.names = T, recursive = FALSE)
+  for (test_file in test_files) {
+    .GlobalEnv$map_to_desc[[.GlobalEnv$counter]] <- list()
+    .GlobalEnv$file_points[[.GlobalEnv$counter]] <- list()
+    test_file(test_file, reporter = "silent", env = .create_counter_env(project_path))
+    .GlobalEnv$counter <- .GlobalEnv$counter + 1
+  }
+  return (.add_points(.GlobalEnv$test_available_points, .GlobalEnv$file_points, .GlobalEnv$map_to_desc))
+}
+
+.init_global_vars <- function() {
+  .GlobalEnv$test_available_points <- list()
+  .GlobalEnv$file_points <- list()
+  .GlobalEnv$map_to_desc <- list()
+  .GlobalEnv$counter <- 1
+}
+
+.add_points <- function(test_available_points, file_points, map_to_desc) {
+  all_available_points <- list()
+  for (i in (1:unlist(.GlobalEnv$counter - 1))) {
+    for (desc in map_to_desc[[i]]) {
+      all_available_points[[desc]] <- c(file_points[[i]], test_available_points[[desc]])
+    }
+  }
+  return (all_available_points)
+}
+
+.create_counter_env <- function(project_path) {
+  test_env <- new.env()
+  .define_counter_functions(test_env, project_path)
+  return (test_env)
+}
+
+.define_counter_functions <- function(test_env, project_path) {
+  .source_files(test_env, project_path)
+  test_env$test <- function(desc, point, code){
+    if (!(desc %in% .GlobalEnv$test_available_points)) {
+      .GlobalEnv$test_available_points[[desc]] <- list()
+    }
+    .GlobalEnv$test_available_points[[desc]] <- c(point)
+    .GlobalEnv$map_to_desc[[.GlobalEnv$counter]] <- c(.GlobalEnv$map_to_desc[[.GlobalEnv$counter]], desc)
+  }
+  test_env$points_for_all_tests <- function(points){
+    .GlobalEnv$file_points[[.GlobalEnv$counter]] <- c(points)
+  }
+}
+
+# Checks the available points for all test in the project without running test. Creates
+# file .available_points.json in the project root.
+run_available_points <- function(project_path = getwd()) {
+  available_points <- .get_available_points(project_path)
+
+  json_results <- .create_available_points_json_results(available_points)
+  .write_json(json_results, paste0(project_path, "/.available_points.json"))
+}

plugins/r/tests/tmcRtestrunner/R/ResultsCreator.R

@@ -0,0 +1,109 @@
+.create_file_results <- function(testthat_file_output,
+                                 tests_points,
+                                 file_points) {
+
+  results <- list()
+  for (test in testthat_file_output) {
+    name <- test$test
+    status <- .get_status_for_test(test)
+    message <- .create_message_for_test(test, status)
+    backtrace <- .create_backtrace_for_test(test, status)
+    points <- .get_points_for_test(name,
+                                   tests_points,
+                                   file_points)
+
+    test_result <- list("name" = name,
+                        "status" = status,
+                        "points" = points,
+                        "message" = message,
+                        "backtrace" = backtrace)
+
+    results[[length(results) + 1]] <- test_result
+  }
+  return(results)
+}
+
+.get_points_for_test <- function(test_name, tests_points, file_points) {
+  if (is.null(tests_points[[test_name]])) {
+    test_points <- vector()
+  } else {
+    test_points <- tests_points[[test_name]]
+  }
+  test_points <- c(file_points, test_points)
+  return(test_points)
+}
+
+.get_status_for_test <- function(test) {
+  if (.check_if_test_passed(test)) {
+    status <- "pass"
+  } else {
+    status <- "fail"
+  }
+  return(status)
+}
+
+#Checks if a single test passed
+.check_if_test_passed <- function(test) {
+  ret <- TRUE
+  for (result in test$results) {
+    if (!.check_if_result_passed(result)) {
+      ret <- FALSE
+      break
+    }
+  }
+  return(ret)
+}
+
+#Check if a single result passed
+.check_if_result_passed <- function(result) {
+  return(format(result) == "As expected")
+}
+
+.message_from_failed_result <- function(result) {
+  message_rows <- strsplit(result$message, "\n")[[1]]
+  return(paste(message_rows, collapse = "\n"))
+}
+
+.create_message_for_test <- function(test, status) {
+  if (status == "pass") return("")
+
+  for (result in test$results) {
+    if (format(result) != "As expected") {
+      return(.message_from_failed_result(result))
+    }
+  }
+  return("")
+}
+
+.create_backtrace_for_test <- function(testthat_test_result, status) {
+  if (status == "pass") return(list())
+
+  for (result in testthat_test_result$results) {
+    if (format(result) != "As expected") {
+      backtrace <- list()
+      i <- 1;
+      for (call in result$call) {
+        backtrace <- append(backtrace, paste0(i, ": ", .create_call_message(call)))
+        i <- i + 1
+      }
+      return(backtrace)
+    }
+  }
+  return(list())
+
+}
+
+.create_call_message <- function(call) {
+  call_str <- format(call)
+  call_srcref <- attributes(call)$srcref
+  srcref_data <- c(call_srcref)
+  srcfile_filename <- attributes(call_srcref)$srcfile$filename
+
+  if (is.null(call_srcref)) {
+    message <- paste0(call_str)
+  } else {
+    message <- paste0(call_str, " in ", srcfile_filename, "#", srcref_data[[1]])
+  }
+
+  return(message)
+}

plugins/r/tests/tmcRtestrunner/R/ResultsJsonParser.R

@@ -0,0 +1,47 @@
+#Creates JSON containing test names and points availble from them, based on the test file.
+.create_available_points_json_results <- function(available_points) {
+  results <- list()
+  for (desc in names(available_points)) {
+    results[[desc]] <- available_points[[desc]]
+  }
+  return (results)
+}
+
+.create_json_run_results <- function(run_results) {
+  json_test_results <- list()
+  for (test_result in run_results$test_results) {
+    json_test_results[[length(json_test_results) + 1]] <- .create_json_test_result(test_result)
+  }
+  json_run_results <- list("runStatus" = unbox(run_results$run_status),
+                       "backtrace" = lapply(run_results$backtrace, unbox), "testResults" = json_test_results)
+  return(json_run_results)
+}
+
+#Creates JSON for each different test case.
+.create_json_test_result <- function(test_result) {
+  test_result <- list(status = unbox(test_result$status),
+                     name = unbox(format(test_result$name)),
+                     message = unbox(test_result$message),
+                     backtrace = lapply(test_result$backtrace, unbox),
+                     points = test_result$points)
+  return(test_result)
+}
+
+#Writes JSON based on the whole test result.
+.write_json <- function(results, file) {
+  #json utf-8 coded:
+  json <- enc2utf8(toJSON(results, pretty = FALSE))
+  json <- prettify(json)
+  #encode json to utf-8 and write file
+  write(json, file)
+}
+
+#Prints results.
+.print_results_from_json <- function(json_result) {
+  for (test in json_result$testResults) {
+    cat(sep = "", test$name, ": ", test$status, "\n")
+    if (test$message != "") {
+      cat(sep = "", "\n", test$message, "\n")
+    }
+  }
+}

plugins/r/tests/tmcRtestrunner/R/RunTests.R

@@ -0,0 +1,75 @@
+# Runs the tests from project directory and writes results JSON to the root of the project
+# as .results.json.
+#
+# Args:
+#  project_path: The absolute path to the root of the project being tested.
+#  print: If TRUE, prints results; if not, not. DEFAULT is FALSE.
+#
+# Returns:
+#   Run results list containing: runStatus (string), backtrace (list), test_results (list)
+run_tests <- function(project_path = getwd(), print = FALSE) {
+  #Runs tests for project and returns the results.
+  #If sourcing_error occurs, .sourcing_error_run_results returns the results.
+  run_results <- tryCatch({.run_tests_project(project_path)},
+                          sourcing_error = .sourcing_error_run_result,
+                          run_error = .run_error_run_result)
+
+  json_run_results <- .create_json_run_results(run_results)
+  .write_json(json_run_results, file.path(project_path, ".results.json"))
+
+  if (print) {
+    .print_results_from_json(json_run_results)
+  }
+
+  invisible(run_results)
+}
+
+.run_tests_project <- function(project_path) {
+  test_results <- list()
+  #Lists all the files in the path beginning with "test" and ending in ".R"
+  test_files <- list.files(path = file.path(project_path, "tests", "testthat"), pattern = "test.*\\.R",
+                           full.names = TRUE, recursive = FALSE)
+
+  for (test_file in test_files) {
+    file_results <- .run_tests_file(test_file, project_path)
+    test_results <- c(test_results, file_results)
+  }
+  return(list("run_status" = "success", "backtrace" = list(), "test_results" = test_results))
+}
+
+.run_tests_file <- function(file_path, project_path) {
+  .GlobalEnv$points <- list()
+  .GlobalEnv$points_for_all_tests <- list()
+
+  test_env = .create_test_env(project_path)
+  test_file_output <- tryCatch({test_file(file_path, reporter = "silent", env = test_env)},
+                               error = .signal_run_error)
+
+  test_file_results <- .create_file_results(test_file_output, points, .GlobalEnv$points_for_all_tests)
+
+  return(test_file_results)
+}
+
+.signal_sourcing_error <- function(error) {
+  sourcing_error <- simpleError(message = error$message, call = error$call)
+  class(sourcing_error) <- c("sourcing_error", class(sourcing_error))
+  signalCondition(sourcing_error)
+}
+
+.sourcing_error_run_result <- function(sourcing_error) {
+  split_message <- strsplit(sourcing_error$message, split = "\n")
+  backtrace <- lapply(split_message[[1]], unbox)
+  return(list("run_status" = "sourcing_failed", "backtrace" = backtrace, "test_results" = list()))
+}
+
+.signal_run_error <- function(error) {
+  run_error <- simpleError(message = error$message, call = error$call)
+  class(run_error) <- c("run_error", class(run_error))
+  signalCondition(run_error)
+}
+
+.run_error_run_result <- function(run_error) {
+  split_message <- strsplit(run_error$message, split = "\n")
+  backtrace <- lapply(split_message[[1]], unbox)
+  return(list("run_status" = "run_failed", "backtrace" = backtrace, "test_results" = list()))
+}