Running a getAllIndices call for indices created using the default token_chars parameter throws an error

[kved96]

Java API client version

8.16.1

Java version

17

Elasticsearch Version

8.14.3

Problem description

I'm running into an issue after creating an index with an n_gram tokenizer with the token_chars field when using the below method to get all the indices in the cluster. The specific settings I'm using at create time are:

{
"index.analysis.analyzer.test_ngram_analyzer.tokenizer": "ngram_tokenizer",
"index.analysis.test_ngram_analyzer.filter" : "lowercase",
"index.analysis.test_ngram_analyzer.type": "custom", 
"index.analysis.tokenizer.ngram_tokenizer.type": "ngram",
"index.analysis.tokenizer.ngram_tokenizer.min_gram": 1,
"index.analysis.tokenizer.ngram_tokenizer.max_gram": 2

}

Then using the get indices endpoint like below:

esClient.indices()
                .get(r -> r.index("_all")
                        .allowNoIndices(true)
                        .expandWildcards(ExpandWildcard.Open, ExpandWildcard.Closed))
                .result()

Throws with:

status: 200, [es/indices.get] Failed to decode response","throwable2_message":"Error deserializing co.elastic.clients.elasticsearch._types.analysis.TokenizerDefinition: co.elastic.clients.util.MissingRequiredPropertyException: Missing required property 'NGramTokenizer.tokenChars' 

I would expect this to be patched by #877, but it doesn't look like it is.

[l-trotta]

Hello @kved96, I'm sorry but I really can't reproduce the issue, I'm using the client version 8.16.1, created an index in Kibana like this:

PUT ngram-test
{
  "settings": {
    "analysis": {
      "analyzer": {
        "test_ngram_analyzer": {
          "tokenizer": "ngram_tokenizer"
        }
      },
      "test_ngram_analyzer": {
        "filter" : "lowercase",
        "type": "custom"
      },
      "tokenizer": {
        "ngram_tokenizer": {
          "type": "ngram",
          "min_gram": 1,
          "max_gram": 2
        }
      }
    }
  }
}

and performing the call you provided simply returns the list of all indices settings.
I have a request: could you try getting that specific index, like this?

esClient.indices()
                .get(r -> r.index("ngram-test") // or your index name
                        .allowNoIndices(true)
                        .expandWildcards(ExpandWildcard.Open, ExpandWildcard.Closed))
                .result()

Since you're getting all indices it could be another index that's causing the problem, and not the newly created one?
Also I have to ask, since this is pretty weird, are you super sure the version of the client is 8.16.1 and it's not clashing with any other version?

Thank you for your patience.

[kved96]

Hey @l-trotta, I just re-ran my test and it looks like I had something in cache that was keeping my client version at 8.14.3, can confirm that this is fixed!