Add PersianStemmer

[raminmjj]

No description provided.

[NightOwl888]

Thanks for the PR.

You have done a great job following conventions and putting this together. However, PersianStemmer is not something that already exists in Lucene and we don't have an approval process set up for this sort of thing in .NET (yet).

Options

1. If you would like to see this released in Lucene.NET sooner rather than later, you can port this great contribution to Java and submit it to the Lucene repository on GitHub. See their Contributing guide for more info.
2. Alternatively, we can port this to Java and submit the contribution for you (when we get a chance to) and will include it in a future Lucene.NET 4.8.0 release, even though it didn't exist in Java at this version.

Either way, it would be great to keep this PR open and to report the status of the contribution to Lucene back here so we can track the progress and learn about how these sort of approvals work. We will merge it once the Lucene team makes the final approval.

[raminmjj]

Hi Shad,
I try to port this to Java.
Thank you.

[NightOwl888]

Great. Let us know if you need help with that. The Reader has been moved to a different place in Lucene analyzers, but other than that, I think it will be straightforward to do.

[raminmjj]

I made a pull request 3 months ago. And I'm waiting for them :)

[NightOwl888]

Thanks for updating this. There are just a few minor changes to do before this can be merged.

[raminmjj]

Hi Shad.
I implemented your suggestions.
Please review the changes.

[NightOwl888]

Sorry for the delay. There is one more requested change that has not been resolved, calling Arrays.Equals() to do the char[] comparison.

[NightOwl888]

@raminmjj - Looks like the Lucene team decided not to make the PersianStemmer part of the built-in PersianAnalyzer, but expect users to custom build an analyzer if they want to utilize it: apache/lucene#904

I will make the same changes to Lucene.NET in a new PR.

[mocobeta]

Hi,

Looks like the Lucene team decided not to make the PersianStemmer part of the built-in PersianAnalyzer, but expect users to custom build an analyzer if they want to utilize it: apache/lucene#904

We removed PersianStemmer from PersianAnalyzer only for 9x branch (current maintenance branch) to avoid breaking the backward compatibility in analysis in 9.x, but the change was not reverted from main and will be released in the next major release.