Every tutorial I have seen for mocha-phantomjs shows having a test harness html file, and a separate javascript file that gets included.
Is this the correct way to do this for each test? I want to create a separate test for each page in my website, but it seems like overkill/crazy to duplicate an html file for every test case.
Granged, this is my first time trying to use mocha-phantomjs, but still, it seems really odd to create an html file and a js for every test case.
What is the standard for doing this sort of thing? I have been googling for about an hour now and can't find any good examples.
I know it seems weird, but... yes.
You need fixture (or harness) files in the "/test" directory. By default, Mocha looks in this directory for filenames with a .html extension, starting with test.html.
Make sure to include the script (and css) tags for 1) mocha, 2) chai (or whatever other assertion library you want), 3) and your specific test suites.
Personally, I've found it helps to use it with a modular bootloader like RequireJS. That way all your fixture files can point to a single configuration file: less maintenance.
Related
I have tried many option but any of them not work as i need. I have created some feature file as i required time by time. so i have a disorder feature file.
I want to run the feature file as i wanted. like.
fil1.feature
fil2.feature
fil3.feature
fil4.feature
so i want to run in this sequence : file3.feature->fil4.feature->fil1.feature.
I have tried #tag, #feature in junit test runner option but its maintain the sequence its run only 3,4 but can't run 1.
So can you tell me how to run feature file randomly???
Cucumber picks up the feature files in alphabetic order from the folder given in the features parameter of the CucumberOptions. So one options would be to rename your feature files alphabetically in the order you want.
After the feature files in the initial folder are read, then the sub folders in the location are picked up alphabetically and the feature files inside them are read. So you can place the file you want to be used later into a sub-folder.
Saying all this, it is not a very good idea to have any dependency between tests which requires a sequence to be maintained.
I got the same stop. Have you solved this problem?
I did some test, use --tags #XXX to controll the sequence is useless, it's just action on different scenarios in a feature file.
so far, I have to do it like this cucumber features/c.feature features/a.feature features/b.feature I think this is a little bit better than rename feature files.
But you may say that: "If there is a lot of feature files......" In my project(ROR), I assigned the statement to ./config/cucumber.yml
In cucumber.yml I define a profile, test_dev: features/c.feature features/a.feature features/b.feature after that, I just to use cucumber -p test_dev and the feature files execute in order.
If you have a butter way, please share with us.
Is it possible to specify a relative path, rather than an absolute one, when specifying the location of a data file when doing data driven development with the SelBlocks add-on?
I'm working on a small team and we'd love to be able to have our tests be portable and eventually become part of our production process. We are currently using the Selenium IDE to write the tests and we have some people on the team using Macs, some using Windows, etc. So, right now, the 'forXML' command requires an absolute path, like:
forXML|file:///C:/data.xml
This isn't terribly portable. It would be handy to just do something like:
forXML|file:///./data.xml
and that would pull up a data.xml file located in the same directory as the test case or the current test suite or something. I've played with various versions of the syntax and I can't figure out anything that will pull up a file via some 'relative' path. Is this even possible?
Thanks
Paths are relative to the running script itself. Examples:
forXML|data.xml
forXML|data_dir/data.xml
See documentation here: http://refactoror.wikia.com/wiki/Selblocks_Reference
Paths are relative to the running script itself. Examples:
forXML|data.xml
forXML|data_dir/data.xml
This is not working for me, works only using an absolute path like forXML|file:///C:/dataDir/data.xml
Having it relative, or at least configurable as a parameter ([yourpath]filename.xml)would be great
I am writing test cases for ROR. The code to test one model/controller is too much . Is there anyway to break the file in different files for testing the same model/controller.
Yes you can split up tests into different files. I believe test::unit requires files to be *_test.rb.
So say you had a test file for a User model. You could have your tests broken up like:
user_validations_test.rb
user_login_test.rb
...
I know you're using test::unit but, same thing goes with RSpec, you can break up your tests into *_spec.rb files.
If you are using rspec for testing, you can simply add more than one file describing the same class. It might be reasonable to create in spec/ directory matching subdirectory, and place all spec files in there.
I've been trying to figure out if there is already an accepted method for testing file io operations in Haskell, but I have yet to find any information that is useful for what I am trying to do.
I'm writing a small library that performs various file system operations (recursively traverse a directory and return a list of all files; sync multiple directories so that each directory contains the same files using inodes as the equality test and hardlinks...) and I want to make sure that they actually work, but the only way I can think of to test them is to create a temporary directory with a known structure and compare the results from the functions executed on this temporary directory with the known results. The thing is, I would like to get as much test coverage as possible while still being mainly automated: I don't want to have create the directory structure by hand.
I have searched google and hackage, but the packages that I have seen on hackage do not use any testing -- maybe I just picked the wrong ones -- and anything I find on google does not deal with IO testing.
Any help would be appreciated
Thanks, James
Maybe you can find a way to make this work for you.
EDIT:
the packages that I have seen on hackage do not use any testing
I have found an unit testing framework for Haskell on Hackage. Including this framework, maybe you could use assertions to verify that the files you require are present in the directories that you want them to be and they correspond to their intended purpose.
HUnit is the usual library for IO-based tests. I don't know of a set of properties/combinators for file actions -- that would be useful.
There is no reason why your test code cannot create a temporary directory, and check its contents after running your impure code.
If you want mainly automated testing of monadic code, you might want to look into Monadic QuickCheck. You can write down properties that you think should be true, such as
If you create a file with read permission, it will be possible to open the file for reading.
If you remove a file, it won't open.
Whatever else you think of...
QuickCheck will then generate random tests.
I've been reading O'Reilly book "Dojo - The Definitive Guid" but somethings are still not definitive to me.
They talk about "bootstrapping" and getting the dojo.css from the AOL CDN".
When I'm testing on my machine, should I use the CDN? Or should I wait and use that only when I deploy?
Secondly, the book talks about CDN for dojo, but not for dijit.
I'm developing on Google App Engine (GAE) - so having the 2000+ Dojo/Dijit files in my Javascript directory is a little annoying, because it slows down my upload to GAE each time.
Firebug is giving me this error:
GET http://localhost:8080/dijit/nls/dijit-all_en-us.js 404 not Found
GET http://localhost:8080/dijit/_editor/plugins/FontChoice.js 404 not Found
I downloaded the sample from here:
http://archive.dojotoolkit.org/nightly/dojotoolkit/dijit/themes/themeTester.html?theme=soria
and I'd like to "simply" get it to run on my machine under local google app engine (which is the localhost:8080 that you see in the URLs above).
I see this statement which probably is causing the second 404 above:
dojo.require("dijit._editor.plugins.FontChoice");
One other error:
cannot access optimized closure
preload("en-us") dijit-all.js (line 479)
anonymous("dijit.nls.dijit-all", ["ROOT", "ar", "ca", 40 more... 0=ROOT 1=ar 2=ca 3=cs 4=da 5=de 6=de-de 7=el 8=en 9=en-gb])dijit-all.js (line 489)
dijit-all.js()
dojo.i18n._searchLocalePath(locale, true, function(loc){\n
To continue for now, I'm going to try to copy the entire dijit library, but is there a solution short of that?
My current script includes look like this:
<script type="text/javascript" src="/javascript/dijit.js"></script>
<script type="text/javascript" src="/javascript/dijit-all.js" charset="utf-8"></script>
I got the dijit.js file by copying and renaming dijit.js.uncompressed.js to dijit.js.
You have a few options actually:
You could use the CDN for everything (though using the full source locally does give you better error messages). Google has them as well. Dijit is here: http://ajax.googleapis.com/ajax/libs/dojo/1.3.2/dijit/dijit.js FYI. This has many advantages in my opinion. User caching of the JS being the primary one.
Build a layered file. I think the O'Reilly book has a section about it but the PragProg book is better in this regard IMO. There's also this doc on dojocampus.org about building. This will trim down the files you need to upload to GAE and speed up your app loading. This is actually what I do in order to cut down on HTTP requests.
Keep doing what you are doing. :)
Regarding the errors you are seeing about 404 for en-us files are essentially harmless. Here's a better description.
You also might be reloading dijit files by using dijit.uncompressed.js and dijit-all.js and causing problems in the process...but I'm not sure about this one.
I just want to clarify that when using CDN all you need to include is the main Dojo script. The rest will be pulled in automatically when you dojo.require() them.
If for some (technical) reasons you don't want to use the X-Domain loader (CDNs use this type of loader), you can do a custom build (well-described in many places). After the build you copy only relevant files to your server. No need to copy all 2000+ tests, demos, unused DojoX projects, Dijits and so on.
During the build you will create a single minified file (or a few layers), which will include all Dojo JavaScript code you use. If you use Dojo widgets, their templates will be already inlined, so you do not incur hits for them. As part of the build CSS files are combined together and minified too. So literally in most cases you will have just two files: a Dojo layer, which includes everything + your custom code, and a CSS file. In more complex cases you may have more files, but usually we are talking about handful.
How to make sure that everything is in the build? Fire up your favorite network analyzer (Live HTTP Headers, Firebug, Fiddler2, or Charles Proxy would do fine) and see if you hit any files outside of your build. If you do — include them in the build, or try to figure out why they are requested, and eliminate these requests (some localization-related calls are fine).
Personally I would start with the CDN option — works well, no hassle, hosted by somebody else with fat pipes.
To address your first question, use the full source version locally for development, so that you can get clearer debug info which points to a legible line in source, rather than the single line the minified version is reduced to. Use the CDN for production.