[CODEC-249] Fix Incorrect transform of CH digraph according Metaphone basic rules

[Shalujha0907]

Incorrect transformation of the CH digraph in the Metaphone algorithm

This PR fixes the above issue.

Previously, we checked whether the character following CH was a vowel; if so, we appended the character "K". I have removed that check so that when "CH" is encountered, the algorithm simply appends "X".

[garydgregory]

Hello @Shalujha0907

Thank you for the PR!

I see this PR changes the expectation for CHARACTER from KRKT to XRKT, but isn't the correct value XRKTR?

[jhaShalu]

Hello @Shalujha0907

Thank you for the PR!

I see this PR changes the expectation for CHARACTER from KRKT to XRKT, but isn't the correct value XRKTR?

Hello @garydgregory

Thanks for the review and for calling this out.

XRKTR is the full phonetic sequence, but Metaphone in Commons Codec uses a default maxCodeLen of 4, so the returned value is truncated to XRKT.
That’s why this test expects XRKT (and not XRKTR) unless setMaxCodeLen(...) is increased.

Please let me know if you think we should also add a separate assertion with a higher maxCodeLen to make that behavior explicit.

[garydgregory]

Hi @Shalujha0907
Ah, yes, you're right 👍
Would you please mind adding a new test for this specific 5 char use case?
Ty!

[jhaShalu]

5 cha

Hi @Shalujha0907 Ah, yes, you're right 👍 Would you please mind adding a new test for this specific 5 char use case? Ty!

Hi @garydgregory
Pushed the latest change.

Thank you!

[garydgregory]

Jira ticket is https://issues.apache.org/jira/browse/CODEC-249