elasticsearch query string probplem with lowercase and underscore - elasticsearch

Question

I create a custom analyzer in my Elasticsearch, I want to seperate from only white space of my words in my defined field, is "my_field". And my search needs to be case insensitive, for this feature, I used lowercase filter.
PUT my_index
{
"settings": {
"analysis": {
"analyzer": {
"my_custom_analyzer": {
"type": "custom",
"tokenizer": "whitespace",
"filter": [
"lowercase",
"trim"
]
}
}
}
}
},
"mappings" : {
"my_type" : {
"properties" : {
"my_field" : {
"type" : "string",
"analyzer" : "my_custom_analyzer"
}
}
}
}
After that this creation, I analyze my sample data:
POST my_index/_analyze
{
"analyzer": "my_custom_analyzer",
"text": "my_Sample_TEXT"
}
and the output is
{
"tokens": [
{
"token": "my_sample_text",
"start_offset": 0,
"end_offset": 14,
"type": "word",
"position": 0
}
]
}
I have many data in my documents and in "my_field" that contains "my_Sample_TEXT" but when I search for this text using query string, result returns 0:
GET my_index/_search
{
"query": {
"query_string" : {
"default_field" : "my_type",
"query" : "*my_sample_text*",
"analyzer" : "my_custom_analyzer",
"enable_position_increments": true,
"default_operator": "AND"
}
}
}
My result is:
{
"took": 9,
"timed_out": false,
"_shards": {
"total": 5,
"successful": 5,
"failed": 0
},
"hits": {
"total": 0,
"max_score": null,
"hits": []
}
}
I found that, this problem happens when my text has underscore and uppercase text, can anyone help me to fix this problem?

Answer 1

Could you try to change your mapping part of "my_filed" as below:
"my_field" : {
"type" : "string",
"analyzer" : "my_custom_analyzer"
"search_analyzer": "my_custom_analyzer"
}
Because, ES use standart analyzer when you do not set any analyzer. And standart analyzer can create multiple tokens from your underscored text.