Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Repo info
Activity
    jenisys
    @jenisys
    I have already told you my preferred solution.
    senthil kumar
    @senthilkumar369
    hI @jenisys , how to use "@table_generator.from_file" in my feature file. i have created a json file which has table column names . i want to pass it to my test script
    @at_028
    Scenario:Validate Parameter can be written with L1 key
    When TC_028 Authenticate with L1 KEY
    Then should be able to write Parameter
    @table_generator.from_file=L1_Parameters.json
    | Param_Name |MemoryAddress|Expected|
    senthil kumar
    @senthilkumar369
    i tried the above but not passing the values to the steps
    jenisys
    @jenisys
    @table_generator.from_file“ is an idea how this functionality can be provided. It still needs to be implemented.
    senthil kumar
    @senthilkumar369
    oh ok
    Vincent Bauchart
    @vbauchart
    Hi all, I need a so old PR from 2017 : behave/behave#591
    this PR is so small, is there any chance to merge it?
    We are using gitlab-CI and the colored output will help so much to read behave reports
    Weidong Sun
    @xiaoxianma
    I just open a new PR, do I need to add reviewers or it works as it is? behave/behave#774
    Milosz Mazur
    @_subtle_gitlab
    Hi there, any idea when can we expect 1.2.7 (or the next release, in any case)? I checked the milestones over at https://github.com/behave/behave/milestones, but it seems they're not up to date
    Erwin
    @ErwinBeen
    Hi guys, does any of you use a screenshot comparison library that works nicely with behave and python? I'm checking if in some steps I can compare certain elements on the page as well as full page comparisons
    Eduardo Mendes
    @dunossauro
    Hello everybody.
    Eduardo Mendes
    @dunossauro

    I'm think about extentions. Like pytest has with pluggy. Many projects has insane envrionment files.

    @jenisys, what we think about implement a plugin manager for hooks?

    I try explain my idea on this project https://github.com/dunossauro/hook_plug.

    Something like install plugins in hooks and inside de plugins you can implamente modules and then put context variables and execute routines in hooks

    Peter Bittner
    @bittner

    @ErwinBeen There is a "Visual Testing" section in the Practical Tips on Testing chapter of the documentation. This is not a direct answer to your question, but at least you can find some information there that may help you to dig further.

    Extending the docs, once you find something directly useful, is always welcome!

    Emiliano Heyns
    @retorquere
    How can I get a stack trace from failures in my hooks? Right now I'm getting HOOK-ERROR in before_all: AttributeError: 'Config' object has no attribute 'items' but no stack trace.
    jenisys
    @jenisys
    Use „—verbose „ option to get the stack-trace
    Emiliano Heyns
    @retorquere
    @jenisys thanks!
    Emiliano Heyns
    @retorquere
    I'm getting "ContextMaskWarning: user code is masking context attribute zotero; see the tutorial for what this means", but what tutorial is it referring to here?
    Griffin Obeid
    @griffinobeid
    Hello, All! I am experiencing an issue with the Junit formatter in which my skipped messages are not being captured in the report. Is this a known issue?
    Griffin Obeid
    @griffinobeid
    To clarify more, what I want is the string from the reason parameter in this function: scenario.skip(reason='My reason for skipping this test...')
    Emiliano Heyns
    @retorquere
    I need to do some expensive setup for my tests that I want to run only once. I've seen https://stackoverflow.com/questions/34126474/running-certain-steps-once-before-a-scenario-outline-python-behave, but is that still the only way to do this?
    jenisys
    @jenisys

    Regarding expansive Setup: Splitup the Setup in two Parts, First Part injects the Context attribute (via fixture and fixture-tag in before_feature or before_all hook), second part performs the setup when the Context attribute is actually used via Lazy-init (via a function call or property or ...). Therefore, you pay only for the setup if it is actually used by a step implementation or Python code. Problem: Other parts (Scenarios, Steps, Fixtures, ...) may partly destroy/modify your setup after it was performed and it may be hard to detect this case.

    Linked example has a flaw: The step implementation should not always power-up the board but just ensure that the board is in powered-up state.

    jenisys
    @jenisys
    Regarding expensive setup (Part 2): You can always meddle with the Context Stack to inject the Context attribute and cleanup at a deeper Level than you currently are (Feature Level instead of current scenario Level). But there, you are up on your own.
    Eduardo Mendes
    @dunossauro
    Hello everybody, behave has good first issues to hacktoberfest?
    James Belchamber
    @JamesBelchamber
    Was there ever a behave plugin written for pylint?
    I've seen some discussion about the issues but I can't believe I'm the only one that doesn't just want to disable errors :sweat_smile:
    Eduardo Mendes
    @dunossauro
    @JamesBelchamber, what do you want to omit from the linter?
    James Belchamber
    @JamesBelchamber
    pylint errors on a couple of things:
    1) It cannot identify the components of the behave module (given, when, then) so trigger No name 'then' in module 'behave (for each)
    2) Since step_impl is repeatedly defined below the decorators it triggers function already defined in line X
    Eduardo Mendes
    @dunossauro
    @JamesBelchamber, why don't you name your functions? step_impl is a generic name. Fou your fisrt topic I'll check here.
    Eduardo Mendes
    @dunossauro
    @JamesBelchamber, look this __init__ file. https://github.com/behave/behave/blob/master/behave/__init__.py
    James Belchamber
    @JamesBelchamber
    What am I looking at?
    James Belchamber
    @JamesBelchamber
    On #2 I don't know why we'd name the functions? Doesn't the decorator cause them to be nested in the when/then/given functions anyway?
    (That's not an assertion, I don't know :sweat_smile:)
    PyLint is alerting because it doesn't understand the decorator, and so assumes I've made a mistake (defining the same function twice)
    jenisys
    @jenisys
    The step decorator registers the step implemention function in a step registry. Therefore, the step implementation function behaves like an anonymous function. But if you want to call a step function from another step function to reuse the implementation, it needs to have a unique function name within the module.
    Gaurav-t
    @Gaurav-t
    Hi All, I am new to django-behave, I got stuck at a point where I need to consider the a celery task which get triggered from a view. Right now I am using "celery_always_eager=True" which would run celery tasks at once just like a normal function, but this time I am dealing with a celery task which should get triggered at a definite time interval(say 1 minute). Is there a more appropriate way to integrate celery with behave? Thanks in Advance :)