NightOwl888 commented on issue #257: URL: https://github.com/apache/lucenenet/issues/257#issuecomment-799350848
I'm changing the priority of this because there are at least 3 different places where centralizing random generators would be useful: - [Lucene.Net.TestFramework](https://github.com/apache/lucenenet/tree/2731bfd4443cb12af664c03d0683b938884f1448/src/Lucene.Net.TestFramework/Support/Randomized/Generators) - [Morfologik.TestFramework](https://github.com/NightOwl888/Morfologik.Stemming/tree/c8c5492c391a736277015bc0fc6567eab347638b/src/Morfologik.TestFramework/RandomizedTesting/Generators) - Planned Benchmark Projects (See #310 and https://github.com/apache/lucenenet/pull/349#issuecomment-698747477) Given the fact that these generators have evolved over time and don't likely have any breaking backward changes, porting the latest released version of the [generators](https://github.com/randomizedtesting/randomizedtesting/tree/release/2.7.8/randomized-runner/src/main/java/com/carrotsearch/randomizedtesting/generators) and their [associated tests](https://github.com/randomizedtesting/randomizedtesting/tree/release/2.7.8/randomized-runner/src/test/java/com/carrotsearch/randomizedtesting/generators) would probably be the right approach. Most likely any differences in behavior are bug fixes and important design changes that would be better to have than to not have. In the case of benchmarks, using random generators is useful because it allows us to create repeatable benchmarks with large amounts of data without having to store the data in our repo. ---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected]
