# Word\_delimiter Filter et position

**URL:** <https://discuss.elastic.co/t/word-delimiter-filter-et-position/89071>\
**Category:** Discussions en français\
**Created:** [June 12, 2017, 3:19pm UTC](https://discuss.elastic.co/t/word-delimiter-filter-et-position/89071 "2017-06-12T15:19:13Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![maryline](https://avatars.discourse-cdn.com/v4/letter/m/6de8d8/32.png) [@maryline](https://discuss.elastic.co/u/maryline)\
**Post date:** [June 12, 2017, 3:19pm UTC](https://discuss.elastic.co/t/word-delimiter-filter-et-position/89071/1 "2017-06-12T15:19:13Z")

</div>

Bonjour,  
Je suis passe de ES. 1.7 à la version 5.4.1  
et J'ai un "probleme" avec un word\_delimiter filter.  
Est il possible de ne pas tenir compte de la position des token.  
Je vaux pouvoir faire des recherche sur des noms composés  
Exemple :

```auto
DELETE my_index
PUT my_index
{
  "settings": {
    "analysis": {
      "analyzer": {
        "my_custom_analyzer": {
          "type": "custom",
          "tokenizer": "keyword",
          "char_filter": [
            "html_strip"
          ],
          "filter": [
               	"trim",
    		"lowercase",
		"asciifolding",				
    		"worddelimiter_test"
          ]
        }
  	    },
    	"filter": {
		"worddelimiter_test": {
			"type": "word_delimiter",
			"split_on_numerics": false,
			"split_on_case_change": false,
			"generate_word_parts": true,
			"generate_number_parts": false,
			"catenate_all": true,
			"preserve_original": true
				}
			}
    		}
  },
  "mappings": {
    	"personne": {
			"_all": {
				"enabled": false
			},
			"properties": {
				"nom": {
					"type": "string",
					"analyzer": "my_custom_analyzer",
					"omit_norms": "true",
					"index_options": "docs",
					"fielddata": "true"
				}
			}
    	}
  }
	
}

```

Je poste 1 documents

```auto
POST my_index/personne/
{ "nom": "noma"}

```

ET la recherche d'un nom composé approché a une lettre prêt :sdfsdf-nomb

```auto
POST my_index/personne/_search
{
    "query" :{
        "bool": {
            "must" : [{
                "match" :{
                    "nom" : {
                        "query": "sdfsdf-nomb",
                         "analyzer" : "my_custom_analyzer",
                 	  "fuzziness" : "2"
                }}}
                ]
        }
}}

```

le document noma est trouvé  
mais si la recherche se faite sur nomb-sdfsdf le document n'est pas trouvé.

```auto
POST my_index/personne/_search
{
    "query" :{
        "bool": {
            "must" : [{
                "match" :{
                    "nom" : {
                        "query": "nomb-sdfsdf",
                         "analyzer" : "my_custom_analyzer",
                   "fuzziness" : "2"
                }}}
                ]
        }
}}

```

Je ne comprends pas pourquoi ?

---

<div class="post-metadata">

**Author:** ![dadoonet](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/dadoonet/32/137187_2.png) [@dadoonet](https://discuss.elastic.co/u/dadoonet)\
**Post date:** [June 30, 2017, 2:47pm UTC](https://discuss.elastic.co/t/word-delimiter-filter-et-position/89071/2 "2017-06-30T14:47:11Z")

</div>

Please format your code using `</>` icon as explained in [this guide](https://discuss.elastic.co/t/about-the-elasticsearch-category/21). It will make your post more readable.

Or use markdown style like:

````
```
CODE
```

````

Utilise l'API `_analyze` pour comprendre ce que deviennent tes chaines de caractères lorsqu'elles passent dans un analyseur. Tu devrais trouver peut-être la réponse.

---

<div class="post-metadata">

**Author:** ![system](https://us1.discourse-cdn.com/elastic/original/3X/1/a/1ac57faf039f6b580b3f104ef42a2a89e41014de.png) [@system](https://discuss.elastic.co/u/system)\
**Post date:** [July 28, 2017, 2:47pm UTC](https://discuss.elastic.co/t/word-delimiter-filter-et-position/89071/3 "2017-07-28T14:47:24Z")

</div>

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.