A small tip to reduce iOS App automation execution time in Appium

 Everyone knows that XPATH is very time-consuming, specially when used in Appium, some times they take more than 3 seconds to find the locator.

If you don't believe me, then take a look at the attached video. Observe the time taken by Xpath vs iOS Predicate and decide yourself.

Hence, I started looking for an alternative. 

Please note that this TIP is only applicable for iOS. Instead of using XPATH, if you use iOS Predicate, it will save your time a lot. IOS predicates are very easy to use as they also worked on attributes, just like XPATH, just that their syntax is different from XPath



For example

driver.findElementByXPath('//XCUIElementTypeTable[@name="table"]');
//will translate to:
driver.findElementByIosNsPredicate('type == "XCUIElementTypeTable" AND name == "table"');


driver.findElementByXPath('//XCUIElementTypeTable[@name="table" or @label="tableLabel"]');
//will translate to:
driver.findElementByIosNsPredicate('type == "XCUIElementTypeTable" AND (name == "table" OR label == "tableLabel")');



* Above examples taken from 

https://developers.perfectomobile.com/display/TT/Finding+Elements+on+XCUITest+devices


Comments

Post a Comment

Popular posts from this blog

How to mock a response in Selenium

Image
 First, let us understand why there is a need to mock?  UI automation is an expensive thing to do as it's flaky and time-consuming not only in development but as well as in execution. Use Case 1 Imagine a situation, where we have an e-Commerce application where there is a new rating and review functionality. Now, in order to test this functionality, I need a lot of variations of test data from 0 reviews and ratings to probably a higher number. If I have to automate this scenario, that means, I need to  create multiple users Add their basic details such as Address, payments, etc Add that particular product in all the multiple users who will give ratings and reviews And finally, add ratings and reviews To achieve the above, we have 2 options 1. Using API's  directly and follow the above steps 2. Mock the response of the API which is getting these ratings and reviews details. If you are using Cypress, it is very much possible. todo using their intercept command. For Se...
Read more

Some useful links for testing

Image
  AI AI Needs Testing  —  Jason Arbon Testing Language Models (and Prompts) Like We Test Software  —  Marco Túlio Ribeiro  and  Scott Lundberg Testing LLM-Based Applications: Strategy and Challenges  —  Xin Chen   Basics Bug Severity and Priority Matrix  —  Dilara Atesogullari QA 101: how to manage product quality right from the start  —  Olya Kolyada QA Wiki  —  Luc Gagan Shifting left and what it means for QA  — Reddit   Career How much do testers make?  — Reddit Is It Time You Moved On from Quality Engineering?  —  Stuart Thomas What your QA career progression has been like?  — Reddit   Future Is Manual Testing Dying?  — Reddit Language for QA Automation in 2024?  — Reddit Should We Fear AI in Test Automation?  —  Richard Bradshaw What Does QA Work Look Like in the Era of AI?  —  Patrick Ramser   Guides Bugs backlog automation. RICE for...
Read more

Generic Checklist in case you want to consider Cypress

Tets automation tool space is hot right now. Selenium, which is considered as a de fact tool is now being challenges by Cypress as well as Playwright. We will talk about Playwright later but the first few things about Cypress THE GOOD PART easy peasy and impressive setup ( just a few commands and you get a meaningful as well as standardized framework structure ) It has a nice dashboard to view reports, recordings, and analytics Talking in 2022, it's more mature now, and very good community support Now supports multi-domain ( version 9 onwards) Nice sleek UI runner which has capability to find locators as well time travel ( very helpful in debugging) Nice way to handle API re, response and mocking Full support to component testing Since its syntax is close to jQuery, front end dev finds it easy to use Good documentation Automatic waiting THE NOT SO GOOD PART Re-evaluate if your AUT supports Safari and your end-user uses Safari a lot No direct support to tabs ( workaround exists) Lim...
Read more