Closed moodmosaic closed 6 years ago
/cc @vasily-kirichenko
I'm not that excited by the results, but they look better than what we currently have(?)
athanks, but it still generated same results ([<>]) in 5 of 10, or even 6 of 10 cases...
That's true. (That's also the reason I didn't merge the pull request.)
The difference here is that we're trying to increase the distinct elements count; contrary to #26 where the distinct elements were always about 2 to 3; in this pull request it's 5 to 8 (less [<>]
).
Xeger doesn't have the properties of a true PRNG (it's not SplitMix, for example)―if there's a way to fix this inside Xeger, I can't remember, but it'd take me a while to find the time to investigate further. 😢
@moodmosaic I see. Thanks, I think this PR is an improvement anyway.
I think so, too. However, I opened https://github.com/moodmosaic/Fare/issues/28 because I still think we can do better if we rebase Xeger.cs to the latest one from Java.
About to merge this and release a new NuGet package. @vasily-kirichenko, thanks for the feedback 💯
@vasily-kirichenko, I just pushed version 1.0.4 on NuGet.
Pushing Fare.1.0.4.nupkg to 'https://www.nuget.org/api/v2/package'...
PUT https://www.nuget.org/api/v2/package/
Created https://www.nuget.org/api/v2/package/ 13189ms
Your package was pushed.
Related to https://github.com/moodmosaic/Fare/issues/26. Based on https://github.com/bluezio/xeger/blob/3098732bc920194a5bf5378310d24dc32f1e6cf2/src/main/java/nl/flotsam/xeger/Xeger.java#L106-L110