默认设置
概述
Nightwatch 有一个 默认配置 对象,其中包含预定义的值。这些值可以根据需要覆盖。
设置
const filename_format = function ({testSuite = '', testCase = '', isError = false, dateObject = new Date()} = {}) {
const fileName = [];
const dateParts = dateObject.toString().replace(/:/g, '').split(' ');
dateParts.shift();
const dateStamp = dateParts.slice(0, 5).join('-');
if (testSuite) {
fileName.push(testSuite);
}
if (testCase) {
fileName.push(testCase);
}
return `${fileName.join('/')}${isError ? '_ERROR' : '_FAILED'}_${dateStamp}.png`;
};
module.exports = {
// Location(s) where custom commands will be loaded from.
custom_commands_path: null,
// Location(s) where custom assertions will be loaded from.
custom_assertions_path: null,
// Location(s) where page object files will be loaded from.
page_objects_path: null,
// Location of an external globals module which will be loaded and made available to the test as a
// property globals on the main client instance.
globals_path: null,
// An object which will be made available on the main test api, throughout the test execution
globals: {
// this controls whether to abort the test execution when an assertion failed and skip the rest
// it's being used in waitFor commands and expect assertions
abortOnAssertionFailure: true,
// this controls whether to abort the test execution when an element cannot be located; an error
// is logged in all cases, but this also enables skipping the rest of the testcase;
// it's being used in element commands such as .click() or .getText()
abortOnElementLocateError: false,
// this will overwrite the default polling interval (currently 500ms) for waitFor commands
// and expect assertions that use retry
waitForConditionPollInterval: 500,
// default timeout value in milliseconds for waitFor commands and implicit waitFor value for
// expect assertions
waitForConditionTimeout: 5000,
// this will cause waitFor commands on elements to throw an error if multiple
// elements are found using the given locate strategy and selector
throwOnMultipleElementsReturned: false,
// By default a warning is printed if multiple elements are found using the given locate strategy
// and selector; set this to true to suppress those warnings
suppressWarningsOnMultipleElementsReturned: false,
// controls the timeout value for async hooks. Expects the done() callback to be invoked within this time
// or an error is thrown
asyncHookTimeout: 10000,
// controls the timeout value for when running async unit tests. Expects the done() callback to be invoked within this time
// or an error is thrown
unitTestsTimeout: 2000,
// controls the timeout value for when executing the global async reporter. Expects the done() callback to be invoked within this time
// or an error is thrown
customReporterCallbackTimeout: 20000,
// Automatically retrying failed assertions - You can tell Nightwatch to automatically retry failed assertions until a given timeout is reached, before the test runner gives up and fails the test.
retryAssertionTimeout: 5000,
reporter: function(results, cb) {cb(results)}
},
// configuration settings for the dotenv module - a zero-dependency module that loads environment variables from a .env file into process.env. More details on https://npmjs.net.cn/package/dotenv
dotenv: {},
// persist the same globals object between runs or have a (deep) copy of it per each test;
// this can be useful when persisting data between test suites is needed, such as a cookie or session information
persist_globals: false,
// The location where the JUnit XML report files will be saved. Set this to false if you want to disable XML reporting
output_folder: 'tests_output',
// A string or array of folders (excluding subfolders) where the tests are located.
src_folders: null,
// Used when running in parallel to determine if the output should be collected and displayed at the end.
live_output: false,
// disable support of loading of typescript files for backwards compatibility with test suites
disable_typescript: false,
// Used to disable colored output in the terminal.
disable_colors: false,
// Used when running in parallel to specify the delay (in milliseconds) between starting the child processes
parallel_process_delay: 10,
// An object containing Selenium Server related configuration options
selenium: {
start_process: false,
cli_args: {},
server_path: null,
log_path: '',
port: undefined,
check_process_delay: 500,
max_status_poll_tries: 15,
status_poll_interval: 200
},
// Whether or not to automatically start the Selenium/WebDriver session. If running unit tests, this should be set tot false.
start_session: true,
// End the session automatically when the test is being terminated, usually after a failed assertion.
end_session_on_fail: true,
// Skip the remaining test cases from the current test suite, when one test case fails.
skip_testcases_on_fail: undefined,
// Whether or not to run individual test files in parallel.
test_workers: false,
`/*`
test_workers: {
enabled: true,
// automatically compute the number of workers based on CPU cores
workers: 'auto',
// manually specify the number of workers
workers: 4,
// pass node arguments to individual workers (all of the process.execArgv)
node_options: 'auto',
// selectively pass node arguments to individual worker processes
node_options: ['--inspect']
},
`*/`
// Specifies which test runner to use: default|mocha
test_runner: 'default',
// Defines options used to connect to the WebDriver/Selenium server
webdriver: {
start_process: false,
cli_args: {},
server_path: null,
log_path: '',
use_legacy_jsonwire: undefined,
// Time to wait (in ms) before starting to check the Webdriver server is up and running
check_process_delay: 100,
// Maximum number of ping status check attempts before returning a timeout error
max_status_poll_tries: 10,
// Interval (in ms) to use between status ping checks when checking if the Webdriver server is up and running
status_poll_interval: 200,
// The entire time (in ms) to wait for the Node.js process to be created and running (default is 2 min), including spawning the child process and checking the status
process_create_timeout: 120000,
host: undefined,
port: undefined,
ssl: undefined,
proxy: undefined,
timeout_options: {
timeout: undefined,
retry_attempts: undefined
},
default_path_prefix: undefined,
username: undefined,
access_key: undefined
},
test_settings: {
},
// A url which can be used later in the tests as the main url to load.
launch_url: '',
// set to false if you want to show the extended http traffic command logs from the WebDriver server.
silent: true,
// Used to disable terminal output completely.
output: true,
// Set this to false if you'd like to only see the test case name displayed and pass/fail status.
detailed_output: true,
// Set this to true if you'd like to see timestamps next to the logging output
output_timestamp: false,
// Set this to iso if you'd like to see timestamps as ISO strings
timestamp_format: '',
// Set this to true if you'd like to not display errors during the execution of the test (they are shown at the end always).
disable_error_log: false,
// By default, API command errors that don't deal with DOM elements (e.g. cookie) are ignored, unless they are thrown by Node.js (e.g. ECONNRESET errors)
report_command_errors: false,
// Take error and failure screenshots during test execution
screenshots: {
enabled: false,
filename_format,
path: '',
on_error: true,
on_failure: true
},
// Used to enable showing the Base64 image data in the (verbose) log when taking screenshots.
log_screenshot_data: false,
desiredCapabilities: {
browserName: 'firefox'
},
// An array of folders or file patterns to be skipped (relative to the main source folder).
exclude: null,
// Folder or file pattern to be used when loading the tests. Files that don't match this pattern will be ignored.
filter: null,
// Skip a group of tests (a subfolder); can be a list of comma-separated values (no space)
skipgroup: '',
sync_test_names: true,
// Skip tests by tag name; can be a list of comma-separated values (no space)
skiptags: '',
// Use xpath as the default locator strategy
use_xpath: false,
parallel_mode: false,
report_prefix: '',
unit_tests_mode: false,
default_reporter: 'junit',
// In Nightwatch v1.x, when used with "await" operator, API commands will return the full result object as {value: `<VALUE>`}
// whereas in v2, the value is return directly; if using a callback, the behaviour remains unchanged
backwards_compatibility_mode: false,
// disable the global apis like "browser", "element()", "expect()"; this might be needed if using Nightwatch with third-party libraries
disable_global_apis: false,
// enable aborting the test run execution when the first test failure occurs; the remaining test suites will be skipped
enable_fail_fast: false
}