Feature Requests

  1. Custom Engine Modes - Dynamic project name regex by changed project name

    Hello,

    we have the following naming conversion:
    Company.ComponentA.Foo.dll
    Company.ComponentA.Foo_uTest.dll

    I want the relevant unit test dlls to be selected based on the changed product projects.

    As an addition the mapping should be flexible and support:
    1. Just the unit test target of the changed product target should be used
    2. All unit test targets of a changed component should be used.

    Idea:
    1. Selection Pattern for the changed project
    '^(\w+[.]\w+[.]).*$' // Captures "Company.ComponentA."

    Then a string.Format line to create the regex pattern for the project name selection:

    '^{0}.*_uTest$' which would lead to "Company.ComponentA..*_uTest"

    => All unit tests of a changed…

    1 vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  2. Custom Engine Modes - Project name matches regex

    Hello,
    I want to use regex to match project names because in our projects there is a naming rule:

    <Product-Assembly> has a corresponding unit test target with the name:
    <Product-Assembly>_uTest

    Therefore I want to define a rule like:
    Project name matches regex '^.*_uTest$'

    With best regards,
    stefan

    1 vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  3. Allow mouse-click navigation on stack trace even when instrumentation is off

    When instrumentation is ON you can navigate from error stack trace to the code with a single mouse-click (very similar to ReSharper stack trace explorer).
    When instrumentation is OFF then mouse-click functionality stops working because NCrunch-specific paths are used in this case. ReSharper Stack Trace Explorer still works fine and is able to handle stack traces from NCrunch correctly.
    Original issue reports is here: http://forum.ncrunch.net/yaf_postst1513_Error-StackTrace-is-not-navigatable-when-instrumentation-is-off.aspx

    2 votes
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  4. Add progress bar to Metrics tree view

    I would find the metrics tree view more useful if the code coverage % column had a progress bar in the background. The numbers are great of course, but a progress bar would be much easier to glance at, especially when looking at a large number of rows.

    DotCover has a view that shows what I mean well: https://www.jetbrains.com/dotcover/features/img/dotcover-app.png

    3 votes
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  5. Add an attribute like [Isolated] which applies to each member of a fixture rather than the fixture as a whole

    The [Isolated] attribute currently only provides isolation for the fixture, not each test within it. Where tests need to be isolated, they need to be individually annotated.

    It would be nice to have an attribute like [IsolateEachMember] which could apply to each test in a fixture.

    4 votes
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  6. Install the service as "Automatic (Delayed Start)"

    Really small change, but can the Grid Node server install with the service option of Delayed Start rather than Automatic.

    The only difference is that there will be roughly a 2 minute delay before NCrunch gets started after a server restart, but it means that the server will have had chance to settle down before NCrunch starts running builds and tests.

    2 votes
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    4 comments  ·  Flag idea as inappropriate…  ·  Admin →
  7. Test result icon next to method name

    In the text editor when all the test methods are collapsed, we cannot see which test have failed.

    If an icon next to the method declaration was also displayed we could know which one needs to be expanded and work on without the need to look at the overall NCrunch test window.

    4 votes
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    1 comment  ·  Flag idea as inappropriate…  ·  Admin →
  8. Implement Ctrl+A shortcut in NCrunch configuration grid

    For the moment it's possible to select more than one project but only holding Ctrl key and clicking on each project.

    3 votes
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  9. Separate 'successfully completed' and 'failed tasks' in the Processing Queue window

    The NCrunch Processing Queue window currently has 3 toggles:
    - Pending tasks
    - Tasks being processed
    - Completed tasks

    The last of these has both the successfully completed and the failed tasks. However, I find that confusing and inconsistent:

    - The icon on the toggle button is a checkmark. The icon for successful tasks is a checkmark, while the icon for a failed task is a red "X".
    - The NCrunch Tests window also has successful and failed "tasks", in this case the tests. However here there are two toggles, one for successful and one for failed tests.

    3 votes
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    1 comment  ·  Flag idea as inappropriate…  ·  Admin →
  10. Support Windows Runtime Components

    Support C++ Windows Runtime Components. These are used in Windows Store projects. One of the cool things about WinRT is that you can mix and match languages. I'm not talking about running tests against C++, just allow referencing them. Currently even referencing a component breaks NCrunch compilation for the whole solution.

    4 votes
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  11. Add filter in "test popup" window

    So I navigate code and possible spot an issue, I click on line to bring up the "tests covering this line" and I am interested in a test containing a string e.g. "NoOfQueries" but I am a bit overwelmed with the no of tests for this code so I cannot find it.

    Current workaround is to pin all tests covering it, and then add a text filter in the Tests window (and then unpin all tests which kindof ruins my current "pinned tests scope").

    1 vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  12. cover

    When collapsing a method have another icon next to it indicating the coverage.

    It can have 5 colors:
    Black - no cover
    Green - 100% covering tests. All tests pass
    Red - 100% covering tests with at least one failing test
    Blue - Contains some passing tests but also there are lines of code that are not covered by tests
    Orange - Contains some failing tests but also there are lines of code that are not covered by tests

    2 votes
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  13. Allow a test project to be excluded from creating inline code coverage

    Inline code coverage is great for unit tests, but when integration tests are also instrumented we lose the granularity of coverage.

    N.B Our integration tests are for testing the interaction of our classes either with each other, or 3rd party libraries. So running them in ncrunch and getting test failure notifications straight away is useful for us.

    1 vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    1 comment  ·  Flag idea as inappropriate…  ·  Admin →
  14. 1 vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  15. 1 vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  16. Support more than one workspace base path

    I'm using a RAM-drive as my workspace base path and it often gets full (see complementary feature request here: https://ncrunch.uservoice.com/forums/245203-feature-requests/suggestions/6673500-cleanup-of-workspace). It would be nice to have a fallback solution so that when the target drive is full, it doesn't block the whole building process. This fallback solution could be the %temp%. I guess the best would be to try all workspaces in order.

    So for example, in my case the workspace base path would be:
    d:\;%temp%

    2 votes
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  17. Create a 'Kill Currently Running Tests and Execute Selected Tests Immediately' button

    The existing 'Run selected test(s) in new process' button just gives the selected tests high priority. This is very irritating when longer running tests are clogging the queue. I often go to the Processing Queue to kill the tests I don't care about so that my test will start sooner. Worse, sometimes the test I want runs together with long tests I don't want, so I can't kill just the long tests without removing my requested test as well. Just run my test, right now! The others can wait.

    7 votes
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  18. Support IntelliTrace when debugging tests

    At the moment I am unable to make use of IntelliTrace when debugging tests using the NCrunch test runner.

    I use IntelliTrace frequently while debugging, and find it especially useful when I want to look into ADO.Net calls from ORMs.

    4 votes
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  19. Cleanup of workspace

    For faster unit-testing, I am using a RAM-drive as workspace, with 4GB space.

    NCrunch keeps all old build artifacts around until they are no longer needed. In practice, this means that as I am working, the workspace keeps on filling up, since a change in one place which doesn't impact something else means both versions are kept in the workspace.

    As a result of this, the 4GB quickly get gobbled up, forcing me to cleanup by deactivating/activating NCrunch (which means everything gets rebuilt, even though this isn't necessary) or manually trying to delete as much as possible in the filesystem.

    8 votes
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  20. Option to Build and Start Running Test Only on File Saves

    It would helpful to have an option where NCrunch does not start building the project and running tests until after I save the file. We've had some issues with this in a project I was on that had several thousand unit tests. This would be similar to the way Karma's auto-file watcher works for JavaScript unit testing.

    I'm aware of the sliding building delay - and it can help - but I feel it'd be better to just provide an option to not start processing until I save a file.

    43 votes
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    6 comments  ·  Flag idea as inappropriate…  ·  Admin →
  • Don't see your idea?

Feature Requests

Categories

Feedback and Knowledge Base