French analyzer

The built-in french analyzer can be applied to a text field using the following command:

  1. PUT /french-index
  2. {
  3.   "mappings": {
  4.     "properties": {
  5.       "content": {
  6.         "type": "text",
  7.         "analyzer": "french"
  8.       }
  9.     }
  10.   }
  11. }

copy

Stem exclusion

You can use stem_exclusion with this language analyzer using the following command:

  1. PUT index_with_stem_exclusion_french_analyzer
  2. {
  3.   "settings": {
  4.     "analysis": {
  5.       "analyzer": {
  6.         "stem_exclusion_french_analyzer": {
  7.           "type": "french",
  8.           "stem_exclusion": ["autorité", "acceptation"]
  9.         }
  10.       }
  11.     }
  12.   }
  13. }

copy

French analyzer internals

The french analyzer is built using the following components:

  • Tokenizer: standard

  • Token filters:

    • elision (French)
    • lowercase
    • stop (French)
    • keyword
    • stemmer (French)

Custom French analyzer

You can create a custom French analyzer using the following command:

  1. PUT /french-index
  2. {
  3.   "settings": {
  4.     "analysis": {
  5.       "filter": {
  6.         "french_stop": {
  7.           "type": "stop",
  8.           "stopwords": "_french_"
  9.         },
  10.         "french_elision": {
  11.           "type":         "elision",
  12.           "articles_case": true,
  13.           "articles": [
  14.               "l", "m", "t", "qu", "n", "s",
  15.               "j", "d", "c", "jusqu", "quoiqu",
  16.               "lorsqu", "puisqu"
  17.             ]
  18.         },
  19.         "french_stemmer": {
  20.           "type": "stemmer",
  21.           "language": "light_french"
  22.         },
  23.         "french_keywords": {
  24.           "type": "keyword_marker",
  25.           "keywords": []
  26.         }
  27.       },
  28.       "analyzer": {
  29.         "french_analyzer": {
  30.           "type": "custom",
  31.           "tokenizer": "standard",
  32.           "filter": [
  33.             "french_elision",
  34.             "lowercase",
  35.             "french_stop",
  36.             "french_keywords",
  37.             "french_stemmer"
  38.           ]
  39.         }
  40.       }
  41.     }
  42.   },
  43.   "mappings": {
  44.     "properties": {
  45.       "content": {
  46.         "type": "text",
  47.         "analyzer": "french_analyzer"
  48.       }
  49.     }
  50.   }
  51. }

copy

Generated tokens

Use the following request to examine the tokens generated using the analyzer:

  1. POST /french-index/_analyze
  2. {
  3.   "field": "content",
  4.   "text": "Les étudiants étudient à Paris et dans les universités françaises. Leurs numéros sont 123456."
  5. }

copy

The response contains the generated tokens:

  1. {
  2.   "tokens": [
  3.     {"token": "etudiant","start_offset": 4,"end_offset": 13,"type": "<ALPHANUM>","position": 1},
  4.     {"token": "etudient","start_offset": 14,"end_offset": 22,"type": "<ALPHANUM>","position": 2},
  5.     {"token": "pari","start_offset": 25,"end_offset": 30,"type": "<ALPHANUM>","position": 4},
  6.     {"token": "universit","start_offset": 43,"end_offset": 54,"type": "<ALPHANUM>","position": 8},
  7.     {"token": "francais","start_offset": 55,"end_offset": 65,"type": "<ALPHANUM>","position": 9},
  8.     {"token": "numero","start_offset": 73,"end_offset": 80,"type": "<ALPHANUM>","position": 11},
  9.     {"token": "123456","start_offset": 86,"end_offset": 92,"type": "<NUM>","position": 13}
  10.   ]
  11. }