Analyzer.Kind to support all languages for analyzers

[brentam]

Is your feature request related to a problem?

When trying to build a client using json files, when deserialising the settings, I get the error Missing required property 'Builder.<variant kind>'. This is an example of a settings that will fail: { "index": { "analysis": { "analyzer": { "english_analyzer": { "type": "english", "char_filter": [ "html_strip" ], "tokenizer": "standard" } } } } }

The problem is that Analyzer.Kind does not support most languages. Actually only Dutch is listed in the enum. Analyzer source code

The problem is that

Not able to build the client using analyzers type for languages other than Dutch

What solution would you like?

Add all other languages listed here

What alternatives have you considered?

Using a different client.

Do you have any additional context?

All the necessary info is already mentioned above. here is a sample test that will fail with org.opensearch.client.util.MissingRequiredPropertyException: Missing required property Builder.\<variant kind>

   @Test
    public void testParsingAnalyzer() {

        JsonpMapper mapper = new JsonbJsonpMapper();
        String json = "{ \"index\": { \"analysis\": { \"analyzer\": { \"some_analyzer\": { \"type\": \"english\", \"char_filter\": [ \"html_strip\" ], \"tokenizer\": \"standard\" } } } } } ";

        JsonParser parser = mapper.jsonProvider().createParser(new StringReader(json));
        IndexSettings indexSettings = IndexSettings._DESERIALIZER.deserialize(parser, mapper);
        assertTrue(true);
    }

[dblock]

@brentam Want to try to contribute a fix for it? Feels like at least you should be able to specify an arbitrary analyzer without constantly having to expand the enumeration.

[brentam]

@dblock hi, I will try to contribute when I have some time, I probably will add all the missing languages. That would be enough for this case.

[brentam]

Created PR: https://github.com/opensearch-project/opensearch-java/pull/779 to fix this