I'm looking over various Ruby gems that are tested using Cucumber, and I see this kind of line in the feature files:
Given a file named "myfile.txt":
and I can see it successfully run, but I can't find the step definition, which makes me think it's a "core" step that's defined by Cucumber rather than my code, only I can't find the docs of code for those "core" steps.
Cucumber does not provide any step definition on its own. All step definitions has to written by us.
If you are using Eclipse, you can install this plugin https://github.com/matthewpietal/Eclipse-Plugin-for-Cucumber
Jump to defintion: Click on the keyword (here it is "Given"), hit F3 to jump to the Java code defined for that rule
It's defined in the "Aruba" gem.
https://github.com/cucumber/aruba/blob/master/lib/aruba/cucumber.rb
Given /^a file named "([^"]*)" with:$/ do |file_name, file_content|
write_file(file_name, file_content)
end
Related
Is there any way to extract list of steps used in a Cucumber .feature file? The steps should be in the format they're defined in the step definition #When / #Given / #Then annotations.
Use runner class to generate steps with annotations then later fill with java code topass your unit test cases ,Follow this reference link step by step procedure http://www.automationtestinghub.com/cucumber-step-definition/
We can generate the list of steps by either using the runner class or using the tidy gherkin plugin which is available as extension in chrome
I am currently attempting to use the go_remote_library target??, package??, plugin?? in Pants. Real simple question, here:
If in my code I have the import listed as:
import(
"github.com/golang/groupcache"
)
is it valid for me to specify a name of simply "groupcache" instead of the full import path? Here is what my BUILD file looks like:
go_remote_library(name="groupcache",
rev="d781998583680cda80cf61e0b37dd0cd8da2eb52"
)
Am I doing this right? As a side note, is there a Pants target that I can use to test that my BUILD file is valid? Thanks!
You are doing it right. All of the go targets - go_remote_library in this case, but also go_library and go_binary - currently take a name parameter and it must be the name of the directory the BUILD file lives in. The next release of pants (0.0.44) should remove the name parameter taking the choice away from you.
The 1st line of defense is the BUILD Dictionary.
For go_remote_library you'll find this doc.
As to testing, the simplest test is checking syntax, and for that this does the trick:
./pants list path/to/BUILD:
Note the trailing colon attached to the path
This says "List all the targets defined in path/to/BUILD. Here the : means all - its equivalent to the * wildcard in bourne shells for pants targets in BUILD files.
If you want to check more targets all at once you could say:
./pants list ::
Here the recursive glob is used - equivalent to ** in zsh, and so this asks pants to list all the targets in the repo.
If the syntax checks out, you may still have more subtle issues, like defining a go_remote_library that does not point to a valid github project. These issues will only show up when you try to do more than act on the target's metadata like list and depmap goals do. For a go_remote_library, the simplest way to exercise it is to try and resolve the library:
./pants resolve 3rdparty/go/github.com/bitly/go-simplejson2
If you have this BUILD file contents at that path:
go_remote_library(name='go-simplejson2')
Running the resolve will fail since no such github repo exists.
You can do a similar higher-level check with go_library and go_binary targets, instead running ./pants compile .... This will smoke out whether you're missing any required go_remote_library BUILD files or dependencies.
I have IntelliJ 12 and some groovy code (along with a pile of java code) in a project.
In intelliJ, i can see class A's import of some groovy code, and i have also included the library that has that code.
However, while the package itself is in one colour (for the import), the actual class being imported is in red, which implies an issue of some sort. Hovering the mouse over it reveals no issue though.
When i run a "make" or a "rebuild project" is where the problems start - i get
Groovyc: unable to resolve class com.blah.blah.blah.A
How can i resolve this?
Currently, my project setup is like so:
Under "Libraries" in (Project Structure -> Project Settings -> Libraries) I have:
the jar file with all the groovy code
the src jar file with all the groovy code
In the "Modules" section i have the - well, i don't know what to call it, the column isn't labelled - the library name from the libraries section associated with the src and class files, and the little "export" button beside it is ticked.
Incidentally, opening the class in intelliJ never shows the source code, which given the source is included struck me as weird.
Is there anything else I should need to do?
I've worked this one out, but if anybody knows why groovy cannot be in the "Resource Patterns" list and wants an upvote, do chime in
Oh, right.
I removed the !?*.groovy entry from the list of, um, entries in the File : Settings -> Compiler -> Resource Patterns thingy.
It doesn't seem to matter if "use external build" is on or off for this, but the !?*.groovy; entry cannot be there.
I wonder if anybody knows why?
I had the same problem and had to Add Framework Support and add Groovy to the project to get round this problem.
I created the project using gradle.
I just got your question in my Google results as I had a similar issue. My problem was that I was able to get the groovy code in my IntelliJ 12 project to compile ok, but it wasn't getting wired in properly when I tried to run unit tests within the IDE.
After some investigation, I uncovered that groovy and logback libraries were all set up in the project to be available in the runtime stage of the Maven build of the project, but that resulted in them not being available in the test stage. To fix this, I ended up manually updating the groovy-all and the logback libraries scope from runtime to provided under File->Project Structure->Modules->Dependencies. This allowed me to both compile and test within the IDE while including the Groovy modules as well as the Java modules.
Perhaps you had something similar going on in your project?
Six years later, I also just got this question near the top of my search results.
In my project my Unable to load class 'groovy.text.SimpleTemplateEngine' problem was actually due to a codenarc issue. I was able to resolve the issue by adding the following to build.gradle:
// codenarc version issue work-around
configurations.codenarc {
resolutionStrategy.eachDependency { DependencyResolveDetails d ->
if (d.requested.group == 'org.codehaus.groovy') {
d.useVersion '2.4.7'
}
}
}
Today i decided to see up Groovy source code and to built up my programming muscles in Groovy. I downloaded the Groovy Source code 1.8 from this link. But how do i proceed? In sense which folder first i have to see, so that i can understand better how groovy works(because there are many folders like benchmarks,bootstraps, src etc). May be this seems to be stupid question but i wanna to ask it.
Correct me if am worng.
The source code is inside src/main.
The unit tests all live inside src/test.
I found a good place to start looking was inside the huge class:
src/main/org/codehaus/groovy/runtime/DefaultGroovyMethods.java
This is where a lot of the extra Groovy methods are defined, so you can pick your favourite function (such as String.capitalize for example) and find the definition of that method (around line 9561, but that might be different in the version of the code you have downloaded)
You should then be able to (for example) change how something works, and check that the unit tests still function by calling
ant test
from the root folder, then you should see it build and test reports should be created and placed in the target folder
I tend to use a combination of find and grep to locate the area in the source that I am interested in, then slowly expand out from that class as I find other things that relate to it...
Hope this helps...it's a bit of a big question to try and cover
Is there a way to transfer C++ preprocessor definitions into a custom pre-link step procedure call as a command-line parameter or export them into a file any other way?
Example:
Let's say, I have a c++ project, and in it's Debug configuration I put a preprocessor definition like MAKUMBA_OBA=0x13
Then I add custom pre-link step which executes some javascript like
sarahjessicaparker.js /to tomsrhinoplasty $(MAKUMBA_OBA)
It would be great, if it just worked, but I never get a third parameter in my js. So the question is: how to pass a preprocessor definition to s script?
There is no way to do such a thing. The only acceptable solution is to analyze the build profile name and act accordingly.