This is a low-level library used as the backbone for all Collection running & Request sending functionality, in the Postman App, and allied systems (Postman Monitoring, Newman)
If you are looking to execute collections, you should be using Newman, this is very low level.
npm run test
: Runs lint, system, unit and integration tests of runtimenpm run test-integration-newman
: This command runs tests of newman with the under-development variant of runtimenpm run test-coverage
: This command runspostman-runtime
tests and generate overall coveragenpm/memory-check.sh
: This bash scripts performs first-level memory usage analysis, and plots a graph out of the results
Postman Runtime supports a lot of options to customize its behavior for different environments and use-cases.
var runner = new runtime.Runner(); // runtime = require('postman-runtime');
// a collection object constructed using the Postman Collection SDK.
var collection = new sdk.Collection();
runner.run(collection, {
// Iteration Data
data: [],
// Timeouts (in ms)
timeout: {
request: 30000,
script: 5000
},
// Number of iterations
iterationCount: 1,
// Control flags (you can only specify one of these):
// - gracefully halts on errors (errors in request scripts or while sending the request)
// calls the `item` and `iteration` callbacks and does not run any further items (requests)
stopOnError: true,
// - abruptly halts the run on errors, and directly calls the `done` callback
abortOnError: true,
// - gracefully halts on errors _or_ test failures.
// calls the `item` and `iteration` callbacks and does not run any further items (requests)
stopOnFailure: true,
// - abruptly halts the run on errors or test failures, and directly calls the `done` callback
abortOnFailure: true,
// Environment (a "VariableScope" from the SDK)
environment: new sdk.VariableScope(),
// Globals (a "VariableScope" from the SDK)
globals: new sdk.VariableScope(),
// Execute a folder/request using id/name or path
entrypoint: {
// execute a folder/request using id or name
execute: 'folderName',
// idOrName in case of execute and path in case of path
// is chosen to specify the folder/request to be executed
lookupStrategy: 'path',
// execute a folder/request using a path
path: ['grand_parent_folder_idOrName', 'parent_folder_idOrName']
},
// Configure delays (in ms)
delay: {
// between each request
item: 1000,
// between iterations
iteration: 1000
},
// Used to fetch contents of files, certificates wherever needed
fileResolver: require('fs'),
// Options specific to the requester
requester: {
// An object compatible with the cookieJar provided by the 'postman-request' module.
// To limit programmatic cookie access to only whitelisted domains, add `allowProgrammaticAccess`
// method to the jar. Example:
// jar.allowProgrammaticAccess = function (domain) { return domain === 'postman-echo.com'; };
cookieJar: jar,
// Controls redirect behavior (only supported on Node, ignored in the browser)
followRedirects: true,
// Redirect with the original HTTP method (only supported on Node, ignored in the browser)
followOriginalHttpMethod: false,
// Maximum number of redirects to follow (only supported on Node, ignored in the browser)
maxRedirects: 10,
// Maximum allowed response size in bytes (only supported on Node, ignored in the browser)
maxResponseSize: 1000000,
// Enable to use WHATWG URL parser and encoder
useWhatWGUrlParser: true,
// Removes the `referer` header when a redirect happens (only supported on Node, ignored in the browser)
removeRefererHeaderOnRedirect: false,
// Enable or disable certificate verification (only supported on Node, ignored in the browser)
strictSSL: false,
// Enable or disable detailed request-response timings (only supported on Node, ignored in the browser)
timings: true,
// Enable or disable verbose level history (only supported on Node, ignored in the browser)
verbose: false,
// Implicitly add `Cache-Control` system header in request (only supported on Node, ignored in the browser)
implicitCacheControl: true,
// Implicitly add `Postman-Token` system header in request (only supported on Node, ignored in the browser)
implicitTraceHeader: true,
// Extend well known "root" CAs with the extra certificates in file. The file should consist of one or more trusted certificates in PEM format. (only supported on Node, ignored in the browser)
extendedRootCA: 'path/to/extra/CA/certs.pem',
// network related options
network: {
hostLookup: { // hosts file configuration for dns lookup
type: 'hostIpMap',
hostIpMap: {
'domain.com': '127.0.0.1',
'ipv6-domain.com': '::1',
}
},
restrictedAddresses: {'192.168.1.1': true} // Allows restricting IP/host in requests
},
// Custom requesting agents (only supported on Node, ignored in the browser)
agents: {
http: {
agentClass: http.Agent,
agentOptions: { keepAlive: true, timeout: 399 }
},
https: new https.Agent({ keepAlive: true })
}
},
// Options specific to the script execution
script: {
// Option to set whether to send console logs in serialized format which can be parsed
// using the `teleport-javascript` serialization library.
serializeLogs: false
},
// A ProxyConfigList, from the SDK
proxies: new sdk.ProxyConfigList(),
// A function that fetches the system proxy for a given URL.
systemProxy: function (url, callback) { return callback(null, {/* ProxyConfig object */}) },
// Opt-out of [proxy configured using environment variables]((https://github.com/postmanlabs/postman-request#controlling-proxy-behaviour-using-environment-variables) ) (only supported on Node, ignored in the browser)
ignoreProxyEnvironmentVariables: false,
// A CertificateList from the SDK
certificates: new sdk.CertificateList(),
// *note* Not implemented yet.
// In the future, this will be used to read certificates from the OS keychain.
systemCertificate: function() {}
}, function (err, run) {
console.log('Created a new Run!');
// Check the section below for detailed documentation on what callbacks should be.
run.start(callbacks);
});
You can pass a series of callbacks for runtime to execute as a collection is being executed.
runner.run(collection, { /* options */ }, function(err, run) {
run.start({
// Called any time we see a new assertion in the test scripts
assertion: function (cursor, assertions) {
// cursor = {
// position: Number,
// iteration: Number,
// length: Number,
// cycles: Number,
// eof: Boolean,
// empty: Boolean,
// bof: Boolean,
// cr: Boolean,
// ref: String,
// scriptId: String,
// eventId: String
// }
// assertions: array of assertion objects
// assertion: {
// error: Error,
// index: Number,
// name: String,
// skipped: Number,
// passed: Number
// }
},
// Called when the run begins
start: function (err, cursor) {
// err: null or Error
// cursor = {
// position: Number,
// iteration: Number,
// length: Number,
// cycles: Number,
// eof: Boolean,
// empty: Boolean,
// bof: Boolean,
// cr: Boolean,
// ref: String
// }
},
// Called before starting a new iteration
beforeIteration: function (err, cursor) {
/* Same as arguments for "start" */
},
// Called when an iteration is completed
iteration: function (err, cursor) {
/* Same as arguments for "start" */
},
// Called before running a new Item (check the postman collection v2 format for what Item means)
beforeItem: function (err, cursor, item) {
// err, cursor: Same as arguments for "start"
// item: sdk.Item
},
// Called after completion of an Item
item: function (err, cursor, item, visualizer) {
// err, cursor, item: Same as arguments for "beforeItem"
// visualizer: null or object containing visualizer result that looks like this:
// {
// -- Tmeplate processing error
// error: <Error>
//
// -- Data used for template processing
// data: <Object>
//
// -- Processed template
// processedTemplate: <String>
// }
},
// Called before running pre-request script(s) (Yes, Runtime supports multiple pre-request scripts!)
beforePrerequest: function (err, cursor, events, item) {
// err, cursor: Same as arguments for "start"
// events: Array of sdk.Event objects
// item: sdk.Item
},
// Called after running pre-request script(s)
prerequest: function (err, cursor, results, item) {
// err, cursor: Same as arguments for "start"
// item: sdk.Item
// results: Array of objects. Each object looks like this:
// {
// error: Error,
// event: sdk.Event,
// script: sdk.Script,
// result: {
// target: 'prerequest'
//
// -- Updated environment
// environment: <VariableScope>
//
// -- Updated globals
// globals: <VariableScope>
//
// data: <Object of data variables>
// return: <Object, contains set next request params, etc>
// }
// }
},
// Called before running test script(s)
beforeTest: function (err, cursor, events, item) {
// err, cursor: Same as arguments for "start"
// events: Array of sdk.Event objects
// item: sdk.Item
},
// Called just after running test script (s)
test: function (err, cursor, results, item) {
// results: Array of objects. Each object looks like this:
// {
// error: Error,
// event: sdk.Event,
// script: sdk.Script,
// result: {
// target: 'test'
//
// -- Updated environment
// environment: <VariableScope>
//
// -- Updated globals
// globals: <VariableScope>
//
// response: <sdk.Response>
// request: <sdk.Request>
// data: <Object of data variables>
// cookies: <Array of "sdk.Cookie" objects>
// tests: <Object>
// return: <Object, contains set next request params, etc>
// }
// }
},
// Called just before sending a request
beforeRequest: function (err, cursor, request, item) {
// err, cursor: Same as arguments for "start"
// item: sdk.Item
// request: sdk.request
},
// Called just after sending a request, may include request replays
request: function (err, cursor, response, request, item, cookies, history) {
// err, cursor: Same as arguments for "start"
// item: sdk.Item
// response: sdk.Response
// request: sdk.request
},
// Called just after receiving the request-response without waiting for
// the response body or, request to end.
// Called once with response for each request in a collection
responseStart: function (err, cursor, response, request, item, cookies, history) {
// err, cursor: Same as arguments for "start"
// item: sdk.Item
// response: sdk.Response
// request: sdk.request
},
// Called once with response for each request in a collection
response: function (err, cursor, response, request, item, cookies, history) {
// err, cursor: Same as arguments for "start"
// item: sdk.Item
// response: sdk.Response
// request: sdk.request
},
exception: function (cursor, err) {
// Called when an exception occurs
// @param {Object} cursor - A representation of the current run state.
// @param {Error} err - An Error instance with name, message, and type properties.
},
// Called at the end of a run
done: function (err) {
// err: null or Error
console.log('done');
},
// Called any time a console.* function is called in test/pre-request scripts
console: function (cursor, level, ...logs) {},
io: function (err, cursor, trace, ...otherArgs) {
// err, cursor: Same as arguments for "start"
// trace: An object which looks like this:
// {
// -- Indicates the type of IO event, may be HTTP, File, etc. Any requests sent out as a part of
// -- auth flows, replays, etc will show up here.
// type: 'http',
//
// -- Indicates what this IO event originated from, (collection, auth flows, etc)
// source: 'collection'
// }
// otherArgs: Variable number of arguments, specific to the type of the IO event.
// For http type, the otherArgs are:
// response: sdk.Response()
// request: sdk.Request()
// cookies: Array of sdk.Cookie()
}
});
});