Die Einführung in Phantomjs ist gut geschrieben, mit vielen leicht nachvollziehbaren Beispielen.Leider haben sich einige Dinge inzwischen verändert, manche Methoden sind nun in anderen Klassen verfügbar, was beim Nachvollziehen der Beispiele dann schon mal zu Frusterlebnissen führt.Eine aktualisierte Neuauflage wäre wünschenswert. Aber sie käme zu spät, ich habe das Buch schon gekauft. :D

This book is exactly what you want in a cookbook. It is informative, concise and practical. Testing is usually an afterthought when it comes to JavaScript. This book, with its straight forward writing style, gives you many actionable ways to create a suite of tools which makes testing code not only easier, but enjoyable. I HIGHLY recommend this book to anyone who has an interest in learning more about analyzing and creating performant JavaScript applications.

This is much more than a cookbook, it's pretty much the bible of Phantom JS development.PhantomJS is remarkably thorough (a subtitle might be "no use-case left behind") and yet it's not at all intimidating. In fact the first two chapters are squarely aimed at the beginner. We're introduced to PhantomJS and its core modules with no assumptions re. experience, other than a basic command of JavaScript.From there Friesel takes us through unit testing (all mainstream frameworks are covered), integration testing, CI, using test data for performance metrics and rendering page content.Friesel's writing is clear and straightforward throughout. If you only get one PhantomJS book, get this one.

I read one of the books’ recipes Automating performance analysis with YSlow and PhantomJS on the Packt website, and was thrilled when given the opportunity to review it.The first three chapters of this book explain how to get Phantom installed, use its core modules and begin working with webpages. The next four chapters cover specific use cases: unit, functional and performance testing and generating screenshots and documents. The final chapter describes how to setup Jenkins (the continuous integration server) to run your PhantomJS code.A couple of things stood out in this book:This book is not designed to be read straight through from cover to cover – like I did. Each of the recipes is designed to be self-contained (though this does lead to several paragraphs being repeated in most recipes – how to start the books demo site and run the example script). Making it easy to cherry pick the recipes that fit your needs and ignore those that don’t (i.e., the unit testing chapter contains recipes for running QUnit, Mocha and Jasmine tests – I doubt anyone is using all of these).The author clearly understands Phantom, both its strengths and weaknesses, at a very deep level. Several times he pointed out functions that lack documentation or where the documentation is incorrect. This type of information can save countless hours of debugging and frustration. Likewise, he clearly states where the use of an additional library (i.e., Casper) can make things simpler. When the reader may be left questioning how or why to use a specific recipe (i.e., Preventing CSS from downloading) he provides a clear and simple use case (to view the site when styles are disabled).The book makes the most of its cross references and external links. When a topic is introduced, but is outside the book’s scope the author provides a link to additional information (i.e., how to install Jenkins). Likewise, each recipe contains a “See also” section that references other recipes from the book, though in the eBook version not all of these were hyperlinked to the corresponding section. This was made worse by the fact that the table of contents only lists the chapter titles, not individual recipes, resulting in a lot of swiping to find the correct page.I did see a few minor issues:Deprecated code is not treated consistantly. In one instance (Running a PhantomJS script with arguments) phantom.args and phantom.scriptName provide the main functionality. The author notes that these functions are deprecated and their use therefore discouraged, and adding “Although using them for quick one-off exploratory scripts is fine, neither of these properties should go into any library that we intend to maintain or distribute. He also mentions the now preferred method of getting the same info using the system module. Which left me with several questions – why explore code using a method not suited to production? Why include the deprecated functions at all? Why not just rewrite the example to use the system module? However, when discussing the WebPage constructor he includes an example of the now deprecated form of the constructor noting that it should not be used, but was included because the reader may see it in older code. In my opinion this was a much better method of dealing with the issues presented by deprecated code.In the testing chapters I would have liked to have seen the results of the non-happy path (not all tests pass every time - if yours do your doing it wrong). What happens when the tests fail? Where can I get help interpreting the failure message? Are there online resources I can turn to like StackOverflow, a mailing list or IRC?I found it curious that when using Selenium the author chose to use the JAVA bindings and not the JS bindings – this may be a case of the author being more familiar with the java bindings, but since the book is based on a JS lib I would have thought that a better choice.In one recipe he suggests solving an error by deleting and reinstalling the node modules – but includes no discussion of why or how this solves the problem. While this will likely have no negative effects I still like to know why I am doing something.As I have said the recipes in this book are designed to be self-contained, in a handful of cases the author breaks from his pattern, most notably when using the confess.js library. He only discusses the config file after using it/mentioning it several times, making this information much harder to find.Overall the good far outweighs the bad in this book. For me the chapters on performance testing and continuous integration easily repaid the time invested in reading this book and implementing several of the book’s recipes.Since this book is not designed to be read from cover to cover, the question of value is a little harder to answer; but in the end, if you’re already using Phantom JS and are trying to maximize your return on investment, implementing only a couple of the book’s recipes should make it a worthwhile investment.

http://www.packtpub.com/phantomjs-cookbook/book?utm_source=Mention.com&utm_medium=Blog%2FSM&utm_campaign=MarketingThe PhantomJS Cookbook starts explaining the basics of PhantomJS.Later on, a set of recipes explains the core modules.How to interact with webpage objects comes next, including http actions, injecting js into the browser, simulationuf user input actions, web sockets and an interesting way of blocking css from downloading, amongst others.Then we are presented with how to drive unit tests usign PhantomJS as the main driver, with recipes for a wideset of technologies: jasmine, karma, qunit, grunt for tasks automation. It also presents how to integrate with codecoverage reports tool like Istanbul.The reading continues with Functional and end-to-end testing recipes, with Selenium, GhostDriver, Poltergeist, CapybaraCasper JS, just to name some of them. I've found the ideas for the recipes on how to perform visual regressions usingPhantomCSS very useful.The latest chapters provides good examples for Network monitoring and Performance Analysis, generation of images anddocuments for the reporting part of your testing and development efforts, and finally how to seamessly integratePhantomJS with your CI server.I would recommend this book to readers who wants to dig in how to integrate headless testing to your currentbattery of automated tests. It also explains other concepts, besides PhantomJS, that adds value to the overall proposalof this publication.One little note, though, is the repetition on all the chapters of how to start the example application. Please refactor! :P