I created a synonym filter named "synonym_filter"
"synonym_filter": {
"type": "synonym",
"synonyms": [
"adidas, xyz, abc",
"nike, rofl, lol"
]
}
and analyzer named "synonyms"
"synonyms": {
"filter": [
"lowercase",
"synonym_filter"
],
"tokenizer": "standard"
},
according to the elasticsearch documentation. The complete settings now look like this.
GET /test6_de_idx1/_settings
"test6_de_idx1": {
"settings": {
"index": {
"creation_date": "1471372087742",
"analysis": {
"filter": {
"edge_ngram_back": {
"min_gram": "2",
"side": "back",
"type": "edgeNGram",
"max_gram": "10"
},
"edge_ngram_front": {
"min_gram": "2",
"side": "front",
"type": "edgeNGram",
"max_gram": "10"
},
"synonym_filter": {
"type": "synonym",
"synonyms": [
"adidas, xyz, abc",
"nike, rofl, lol"
]
},
"stop": {
"type": "stop",
"stopwords": "_german_"
},
"strip_hyphens": {
"pattern": "-",
"type": "pattern_replace",
"replacement": ""
},
"length": {
"type": "length",
"min": "2"
},
"strip_spaces": {
"pattern": "\\s",
"type": "pattern_replace",
"replacement": ""
},
"snowball": {
"type": "snowball",
"language": "German"
},
"strip_dots": {
"pattern": "\\.",
"type": "pattern_replace",
"replacement": ""
}
},
"analyzer": {
"std": {
"filter": [
"standard",
"elision",
"asciifolding",
"lowercase",
"stop",
"length"
],
"char_filter": "html_strip",
"tokenizer": "standard"
},
"synonyms": {
"filter": [
"lowercase",
"synonym_filter"
],
"tokenizer": "standard"
},
"keyword_suffix": {
"filter": [
"asciifolding",
"lowercase",
"strip_spaces",
"strip_dots",
"strip_hyphens",
"edge_ngram_back"
],
"tokenizer": "keyword"
},
"text_suffix": {
"filter": [
"standard",
"elision",
"asciifolding",
"lowercase",
"stop",
"edge_ngram_back"
],
"char_filter": "html_strip",
"tokenizer": "standard"
},
"language": {
"filter": [
"standard",
"elision",
"asciifolding",
"lowercase",
"stop",
"snowball",
"length"
],
"char_filter": "html_strip",
"type": "custom",
"tokenizer": "standard"
},
"keyword": {
"filter": [
"asciifolding",
"lowercase",
"strip_spaces",
"strip_dots",
"strip_hyphens"
],
"tokenizer": "keyword"
},
"keyword_prefix": {
"filter": [
"asciifolding",
"lowercase",
"strip_spaces",
"strip_dots",
"strip_hyphens",
"edge_ngram_front"
],
"tokenizer": "keyword"
},
"text_prefix": {
"filter": [
"standard",
"elision",
"asciifolding",
"lowercase",
"stop",
"edge_ngram_front"
],
"char_filter": "html_strip",
"tokenizer": "standard"
}
}
},
"number_of_shards": "1",
"number_of_replicas": "0",
"uuid": "sAiM27R2QOKLj9wjrwoAqw",
"version": {
"created": "2030399"
}
}
}
}
Testing the analyzer with
GET /test6_de_idx1/_analyze
{
"analyzer" : "synonyms",
"text" : "xyz is the English queen"
}
shows that the analyzer is working:
"tokens": [
{
"token": "xyz",
"start_offset": 0,
"end_offset": 3,
"type": "<ALPHANUM>",
"position": 0
},
{
"token": "adidas",
"start_offset": 0,
"end_offset": 3,
"type": "SYNONYM",
"position": 0
},
{
"token": "abc",
"start_offset": 0,
"end_offset": 3,
"type": "SYNONYM",
"position": 0
},
{
"token": "is",
"start_offset": 4,
"end_offset": 6,
"type": "<ALPHANUM>",
"position": 1
},
{
"token": "the",
"start_offset": 7,
"end_offset": 10,
"type": "<ALPHANUM>",
"position": 2
},
{
"token": "english",
"start_offset": 11,
"end_offset": 18,
"type": "<ALPHANUM>",
"position": 3
},
{
"token": "queen",
"start_offset": 19,
"end_offset": 24,
"type": "<ALPHANUM>",
"position": 4
}
]
But when i do a test search with a synonym I get 0 results with adidas, which is used in data i get results.
GET /test6_de_idx1/_search?q=xyz&size=5
"took": 1,
"timed_out": false,
"_shards": {
"total": 1,
"successful": 1,
"failed": 0
},
"hits": {
"total": 0,
"max_score": null,
"hits": []
}
What could please be the problem?
EDIT:
Thanks for the quick reply! A hit looks like this for search term "adidas":
"hits": [
{
"_index": "test6_de_idx1",
"_type": "product",
"_id": "59279",
"_score": 1.0859994,
"_source": {
"id": 59279,
"sku": "0002-10780",
"type_id": "configurable",
"brand": "Adidas",
"color": "Mehrfarbig",
"manufacturer": "Adidas",
"material": "Polyester",
"model": "Damen",
"producttype": "Tank",
"status": 1,
"tax_class_id": 1,
"visibility": 4,
"price": 24.99,
"weight": 0,
"image": "http://sportokay.dev/skin/frontend/default/default/images/catalog/product/placeholder/image.jpg",
"name": "Adidas Keyhole Tank Damen Fitnessshirt",
"description": "Das Adidas Keyhole Tank Damen Fitnessshirt ist ein leichtes, weiches Fitness T-Shirt aus Adidas Climalite Material, welches optimales Feuchtigkeitsmanagement ermöglicht.\r\n- 100 % Polyester\r\n- Adidas Climalite\r\n- Lockere Passform\r\n- Schlitz am Rücken",
"short_description": "Adidas Keyhole Tank Damen Fitnessshirt",
"_categories": [
"Damen",
"Alle",
"Fitness",
"Fitnessbekleidung",
"Shirts"
],
"_prices": {
"price": 24.99,
"final_price": 24.99,
"minimal_price": 24.99,
"min_price": 24.99,
"max_price": 24.99,
"tier_price": null
},
"_url": "http://xxx.dev/at_de/adidas-keyhole-tank-damen-fitnessshirt.html"
}
}
GET /test6_de_idx1/_search?q=brand:xyz does not give results although xyz is synonym to adidas
EDIT2:
This is the mapping currently in use:
"test6_de_idx1": {
"mappings": {
"product": {
"_all": {
"analyzer": "std"
},
"properties": {
"_categories": {
"type": "string",
"analyzer": "language",
"include_in_all": true
},
"_parent_ids": {
"type": "integer",
"index": "no",
"store": true
},
"_prices": {
"properties": {
"final_price": {
"type": "double"
},
"max_price": {
"type": "double"
},
"min_price": {
"type": "double"
},
"minimal_price": {
"type": "double"
},
"price": {
"type": "double"
}
}
},
"_url": {
"type": "string",
"index": "no",
"store": true
},
"adjustment": {
"type": "string",
"norms": {
"enabled": false
},
"index_options": "docs",
"fields": {
"std": {
"type": "string",
"norms": {
"enabled": false
},
"index_options": "docs",
"analyzer": "std"
}
},
"analyzer": "language",
"include_in_all": true
},
"adjustmentrange": {
"type": "string",
"norms": {
"enabled": false
},
"index_options": "docs",
"fields": {
"std": {
"type": "string",
"norms": {
"enabled": false
},
"index_options": "docs",
"analyzer": "std"
}
},
"analyzer": "language",
"include_in_all": true
},
"antennas": {
"type": "string",
"norms": {
"enabled": false
},
"index_options": "docs",
"fields": {
"std": {
"type": "string",
"norms": {
"enabled": false
},
"index_options": "docs",
"analyzer": "std"
}
},
"analyzer": "language",
"include_in_all": true
},
"backlength": {
"type": "string",
"norms": {
"enabled": false
},
"index_options": "docs",
"fields": {
"std": {
"type": "string",
"norms": {
"enabled": false
},
"index_options": "docs",
"analyzer": "std"
}
},
"analyzer": "language",
"include_in_all": true
},
"bike_colour": {
"type": "string",
"fields": {
"prefix": {
"type": "string",
"analyzer": "text_prefix",
"search_analyzer": "std"
},
"std": {
"type": "string",
"analyzer": "std"
},
"suffix": {
"type": "string",
"analyzer": "text_suffix",
"search_analyzer": "std"
}
},
"analyzer": "language",
"include_in_all": true
},....
"brand": {
"type": "string",
"norms": {
"enabled": false
},
"index_options": "docs",
"fields": {
"std": {
"type": "string",
"norms": {
"enabled": false
},
"index_options": "docs",
"analyzer": "std"
}
},
"analyzer": "language",
"include_in_all": true
},
EDIT3:
I added the "synonyms" Analyzer to the mapping properties field but it still does not work. Did i do it correct?
"brand": {
"type": "string",
"norms": {
"enabled": false
},
"index_options": "docs",
"fields": {
"std": {
"type": "string",
"norms": {
"enabled": false
},
"index_options": "docs",
"analyzer": "std"
},
"synonyms": {
"type": "string",
"analyzer": "synonyms"
}
},
"analyzer": "language",
"include_in_all": true
},
