ElasticSearch boundary_chars not working

As of my knowledge cutoff in September 2021, Elasticsearch does not have a specific configuration parameter called boundary_chars. It's possible that there might be some confusion or misinformation regarding this setting.

If you are trying to set up custom boundary characters for tokenization in Elasticsearch, you should be looking at the char_filter settings. Elasticsearch allows you to define custom character filters to preprocess the text before it is tokenized into terms. However, there is no direct boundary_chars parameter to control tokenization behavior.

To achieve custom tokenization behavior, you can define your custom character filter in the analyzer settings and then use that analyzer for your specific fields. Here's an example of how to define a custom character filter and use it in an analyzer:

jsonCopy code
{
  "settings": {
    "analysis": {
      "char_filter": {
        "custom_filter": {
          "type": "mapping",
          "mappings": ["| => "]
        }
      },
      "analyzer": {
        "custom_analyzer": {
          "type": "custom",
          "tokenizer": "standard",
          "char_filter": ["custom_filter"]
        }
      }
    }
  },
  "mappings": {
    "properties": {
      "my_text_field": {
        "type": "text",
        "analyzer": "custom_analyzer"
      }
    }
  }
}

In this example, we have defined a custom character filter called custom_filter, which replaces the | character with a space. Then, we created a custom analyzer called custom_analyzer, which uses the standard tokenizer along with our custom character filter.

Please note that Elasticsearch features and settings may change over time, so always refer to the latest official Elasticsearch documentation for the most up-to-date information. If you are using a version of Elasticsearch beyond my knowledge cutoff date, there may have been changes or new features introduced.