couchdb-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From michellephung <...@git.apache.org>
Subject [GitHub] couchdb-fauxton pull request: grunt works :) with $ grunt nightwat...
Date Mon, 10 Nov 2014 23:03:29 GMT
Github user michellephung commented on the pull request:

    https://github.com/apache/couchdb-fauxton/pull/136#issuecomment-62472667
  
    @robertkowalski Regarding manually deleting databases after failed tests:
    
    This is tricky. 
    
    Sometimes during a test failure, Nightwatch will just bail. It will end all subsequent
test suites, not run anything else, and when this happens, you'll get leftover databases to
clean up, since on error, it will exit the program all together. I've noticed that this happens
mostly when I am writing a javascript callback inside a function like, .getText(), where [something
went wrong, and] the page isn't there. 
    
    Sometimes it's because of the functions we are using. Many of the browser behaviors we
use are assertions, that need to be passed in order for the rest of the test to continue.

    When I want to check if the page has a certain element or bit of text I use 
    
            client.verify.visible(".fauxton");
    
    instead of
    
            client.assert.visible(".fauxton");
    
    `.verify` will fail that particular test, but will keep going and clean up, even though
that checkpoint failed,  
    whereas `.assert` will end the browser testing altogether, without clean up.
    
    Also, functions like:
    
           .waitForElementVisible('#js-new-database-name', waitTime, false)
    
    have a third callback that is for AbortOnFailure, meaning if it is unset, will exit Nightwatch
on failure. Setting this to `false` will continue running the program, and clean up.
    
    Originally I set up the timestamps because Nightwatch can have multiple browser tests
running in parallel, and I didn't want to Firefox's browser test to interfere with Chrome's
test, etc. I'm not sure if we will revisit this idea of parallel testing or not, so I've renamed
all the databases used in the test to 'fauxton-selenium-tests', per your suggestion. 
    
    I am still thinking about your other comment, of deleting a database [that may or may
not be there to delete], then creating it again....



---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

Mime
View raw message