Cross-browser Testing <3 Provided by
- Jasmine 1 is no longer supported
- If you get
Error: TypeError: Cannot set property 'searchSettings' of undefined
use at least version 1.2.7, where this bug has been fixed
- Browser's Logs (only for Chrome)
- Stack Trace (with suspected line highlight)
- Screenshot
- Screenshot only on failed spec
- Search
- Filters (can display only Passed/Failed/Pending/Has Browser Logs)
- Inline Screenshots
- Details (Browser/Session ID/OS)
- Duration time for test cases (only Jasmine2)
- HTML Dump
Need some feature? Let me know or code it and propose Pull Request :) But you might also look in our WIKI/FAQ where we present some solutions how you can enhance the reporter by yourself.
Does not work with protractor-retry
or protractor-flake
. The collection of results currently assumes only one continuous
run.
This is built on top of protractor-angular-screenshot-reporter, which is built on top of protractor-html-screenshot-reporter, which is built on top of protractor-screenshot-reporter.
protractor-beautiful-reporter
still generates a HTML report, but it is Angular-based and improves on the original formatting.
The protractor-beautiful-reporter
module is available via npm:
$ npm install protractor-beautiful-reporter --save-dev
In your Protractor configuration file, register protractor-beautiful-reporter
in Jasmine.
No longer supported
Jasmine 2.x introduced changes to reporting that are not backwards compatible. To use protractor-beautiful-reporter
with Jasmine 2, please make sure to use the getJasmine2Reporter()
compatibility method introduced in protractor-beautiful-reporter@0.1.0
.
var HtmlReporter = require('protractor-beautiful-reporter');
exports.config = {
// your config here ...
onPrepare: function() {
// Add a screenshot reporter and store screenshots to `/tmp/screenshots`:
jasmine.getEnv().addReporter(new HtmlReporter({
baseDirectory: 'tmp/screenshots'
}).getJasmine2Reporter());
}
}
You have to pass a directory path as parameter when creating a new instance of the screenshot reporter:
var reporter = new HtmlReporter({
baseDirectory: 'tmp/screenshots'
});
If the given directory does not exists, it is created automatically as soon as a screenshot needs to be stored.
The function passed as second argument to the constructor is used to build up paths for screenshot files:
var path = require('path');
new HtmlReporter({
baseDirectory: 'tmp/screenshots'
, pathBuilder: function pathBuilder(spec, descriptions, results, capabilities) {
// Return '<browser>/<specname>' as path for screenshots:
// Example: 'firefox/list-should work'.
return path.join(capabilities.caps_.browser, descriptions.join('-'));
}
});
If you omit the path builder, a GUID is used by default instead.
Caution: The format/structure of these parameters (spec, descriptions, results, capabilities) differs between Jasmine 2.x and Jasmine 1.x.
(removed because was only used in jasmine 1.x)
You can modify the contents of the JSON meta data file by passing a function jasmine2MetaDataBuilder
as part of the options parameter.
Note: We have to store and call the original jasmine2MetaDataBuilder also, else we break the "whole" reporter.
The example is a workaround for the jasmine quirk angular/jasminewd#32 which reports pending() as failed.
var originalJasmine2MetaDataBuilder = new HtmlReporter({'baseDirectory': './'})["jasmine2MetaDataBuilder"];
jasmine.getEnv().addReporter(new HtmlReporter({
baseDirectory: 'tmp/screenshots'
jasmine2MetaDataBuilder: function (spec, descriptions, results, capabilities) {
//filter for pendings with pending() function and "unfail" them
if (results && results.failedExpectations && results.failedExpectations.length>0 && "Failed: => marked Pending" === results.failedExpectations[0].message) {
results.pendingReason = "Marked Pending with pending()";
results.status = "pending";
results.failedExpectations = [];
}
//call the original method after my own mods
return originalJasmine2MetaDataBuilder(spec, descriptions, results, capabilities);
},
preserveDirectory: false
}).getJasmine2Reporter());
You can store all images in subfolder by using screenshotsSubfolder
option:
new HtmlReporter({
baseDirectory: 'tmp/screenshots'
, screenshotsSubfolder: 'images'
});
If you omit this, all images will be stored in main folder.
You can store all JSONs in subfolder by using jsonsSubfolder
option:
new HtmlReporter({
baseDirectory: 'tmp/screenshots'
, jsonsSubfolder: 'jsons'
});
If you omit this, all images will be stored in main folder.
You can change default sortFunction
option:
new HtmlReporter({
baseDirectory: 'tmp/screenshots'
, sortFunction: function sortFunction(a, b) {
if (a.cachedBase === undefined) {
var aTemp = a.description.split('|').reverse();
a.cachedBase = aTemp.slice(0).slice(0,-1);
a.cachedName = aTemp.slice(0).join('');
};
if (b.cachedBase === undefined) {
var bTemp = b.description.split('|').reverse();
b.cachedBase = bTemp.slice(0).slice(0,-1);
b.cachedName = bTemp.slice(0).join('');
};
var firstBase = a.cachedBase;
var secondBase = b.cachedBase;
for (var i = 0; i < firstBase.length || i < secondBase.length; i++) {
if (firstBase[i] === undefined) { return -1; }
if (secondBase[i] === undefined) { return 1; }
if (firstBase[i].localeCompare(secondBase[i]) === 0) { continue; }
return firstBase[i].localeCompare(secondBase[i]);
}
var firstTimestamp = a.timestamp;
var secondTimestamp = b.timestamp;
if(firstTimestamp < secondTimestamp) return -1;
else return 1;
}
});
If you omit this, all specs will be sorted by timestamp (please be aware that sharded runs look ugly when sorted by default sort).
Alternatively if the result is not good enough in sharded test you can try and sort by instanceId (for now it's process.pid) first:
function sortFunction(a, b) {
if (a.instanceId < b.instanceId) return -1;
else if (a.instanceId > b.instanceId) return 1;
if (a.timestamp < b.timestamp) return -1;
else if (a.timestamp > b.timestamp) return 1;
return 0;
}
You can set excludeSkippedSpecs
to true
to exclude reporting skipped test cases entirely.
new HtmlReporter({
baseDirectory: `tmp/screenshots`
, excludeSkippedSpecs: true
});
Default is false
.
You can define if you want report screenshots from skipped test cases using the takeScreenShotsForSkippedSpecs
option:
new HtmlReporter({
baseDirectory: 'tmp/screenshots'
, takeScreenShotsForSkippedSpecs: true
});
Default is false
.
Also you can define if you want capture screenshots only from failed test cases using the takeScreenShotsOnlyForFailedSpecs:
option:
new HtmlReporter({
baseDirectory: 'tmp/screenshots'
, takeScreenShotsOnlyForFailedSpecs: true
});
If you set the value to true
, the reporter for the passed test will still be generated, but, there will be no screenshot.
Default is false
.
If you want no screenshots at all, set the disableScreenshots
option to true.
new HtmlReporter({
baseDirectory: 'tmp/reports'
, disableScreenshots: true
});
Default is false
.
Also you can define a document title for the html report generated using the docTitle:
option:
new HtmlReporter({
baseDirectory: 'tmp/screenshots'
, docTitle: 'my reporter'
});
Default is Test results
.
Also you can change document name for the html report generated using the docName:
option:
new HtmlReporter({
baseDirectory: 'tmp/screenshots'
, docName: 'index.html'
});
Default is report.html
.
You can change stylesheet used for the html report generated using the cssOverrideFile:
option:
new HtmlReporter({
baseDirectory: 'tmp/screenshots'
, cssOverrideFile: 'css/style.css'
});
If you want to add small customizations without replaceing the whole css file:
new HtmlReporter({
baseDirectory: 'tmp/screenshots'
customCssInline:`
.mediumColumn:not([ng-class]) {
white-space: pre-wrap;
}
`
});
This example will enable line-wrapping if the tests spec contains newline characters
You can preserve (or clear) the base directory using preserveDirectory:
option:
new HtmlReporter({
baseDirectory: 'tmp/screenshots'
, preserveDirectory: false
});
Default is true
.
You can gather browser logs using gatherBrowserLogs:
option:
new HtmlReporter({
baseDirectory: 'tmp/screenshots'
, gatherBrowserLogs: false
});
Default is true
.
If you do not want all buttons in the search filter pressed by default you can modify the default state via searchSettings:
option:
For example: We filter out all passed tests when report page is opened
new HtmlReporter({
baseDirectory: 'tmp/screenshots'
, clientDefaults:{
searchSettings:{
allselected: false,
passed: false,
failed: true,
pending: true,
withLog: true
}
}
});
Default is every option is set to true
If you do not want to show all columns by default you can modify the default choice via columnSettings:
option:
For example: We only want the time column by default
new HtmlReporter({
baseDirectory: 'tmp/screenshots'
, clientDefaults:{
columnSettings:{
displayTime:true,
displayBrowser:false,
displaySessionId:false,
displayOS:false,
inlineScreenshots:false
}
}
});
Default is every option except inlineScreenshots
is set to true
Additionally you can customize the time values for coloring the time column. For example if you want to mark the time orange when the test took longer than 1 second and red when the test took longer than 1.5 seconds add the following to columnSettings (values are in milliseconds scale):
new HtmlReporter({
baseDirectory: 'tmp/screenshots'
, clientDefaults:{
columnSettings:{
warningTime: 1000,
dangerTime: 1500
}
}
});
If you want to show the total duration in the header or footer area...
new HtmlReporter({
baseDirectory: 'reports'
, clientDefaults:{
showTotalDurationIn: "header",
totalDurationFormat: "hms"
}
});
For all possible values for showTotalDurationIn
and totalDurationFormat
refer to the wiki entry Options for showing total duration of e2e test
By default the raw data of all tests results from the e2e session are embedded in the main javascript file (array results in app.js).
If you add useAjax:true
to clientDefaults
the data is not embedded in the app.js but is loaded from the file combined.json
which contains all the raw test results.
Currently the reason for this feature is a better testability in unit tests. But you could benefit from ajax loading, if you want to polish/postprocess tests results (e.g. filter duplicated test results). This would not be possible if the data is embedded in the app.js file.
Upon running Protractor tests with the above config, the screenshot reporter will generate JSON and PNG files for each test.
In addition, a small HTML/Angular app is copied to the output directory, which cleanly lists the test results, any errors (with stacktraces), and screenshots.
Click More Details to see more information about the test runs.
Use Search Input Field to narrow down test list.
Click View Stacktrace to see details of the error (if the test failed). Suspected line is highlighted.
Click View Browser Log to see Browser Log (collects browser logs also from passed tests)
Click View Screenshot to see an image of the webpage at the end of the test.
Click Inline Screenshots to see an inline screenshots in HTML report.
Please see the examples
folder for sample usage.
To run the sample, execute the following commands in the examples
folder
$ npm install
$ protractor protractor.conf.js
After the test run, you can see that, a screenshots folder will be created with all the reports generated.
You like it? You can buy me a cup of coffee/glass of beer :)
Copyright (c) 2017 Marcin Cierpicki zycienawalizkach@gmail.com
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.