A parallel unit test runner, with poor documentation.
-rerunFailedCategories
should look at the RunStats.html file, and only run categories that failed- Only considers tests that ran and failed (with a positive error code), if they did not run, it will not run them
- Probably does not work with fixtures, (i.e. do not use along with
-uncategorizedInParallel
)
-namespace
will allow you to specify a root namespace to look under- It uses
.StartsWith(namespace)
so it will find test fixtures in all the types under that - If they have categories, and if those categories exist in other namespaces, all the fixtures with those categories will be run
- It uses
You can do: -categories@=IAMAFILEPATH.txt
to run only specific categories
(This does not support feature names)
Bug when using-uncategorizedInParallel
running some that have category names...- If a full feature is ignored, do not include it in the run
- I don't believe I can do this with categories though, becauase a category might have two features, where one is ignored and the other isn't
- Ability to say run by namespaces, or groups of fixtures... idk
- Make the general design of the runners use IOC, instead of hacked together as it is now
- Use
RunnerSettings
for more stuff - Find an alternative to the
ConfigureRun
method?
- Use
- Support fixtures being in multiple categories?
- Seperate RunStats.html files for retry ability and ordering (make a copy, or a combined version when retrying, so it always has the time taken for all categories)
- An option to -excludeIgnoredFeatures from the count and from starting
- Hit ENTER or some other key, and dump all the current status data (namely, what is running, what is queued)
- Option to output a script that will run similar to this, or describe the run
- Either an actual batch file, or just a custom script langauge that can be modified externally then loaded in to concord
- This would allow complete customization of the test runs