Persian analyzer

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

  1. PUT /persian-index
  2. {
  3.   "mappings": {
  4.     "properties": {
  5.       "content": {
  6.         "type": "text",
  7.         "analyzer": "persian"
  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_persian_analyzer
  2. {
  3.   "settings": {
  4.     "analysis": {
  5.       "analyzer": {
  6.         "stem_exclusion_persian_analyzer": {
  7.           "type": "persian",
  8.           "stem_exclusion": ["حکومت", "تأیید"]
  9.         }
  10.       }
  11.     }
  12.   }
  13. }

copy

Persian analyzer internals

The persian analyzer is built using the following components:

  • Tokenizer: standard

  • Char filter: mapping

  • Token filters:

    • lowercase
    • decimal_digit
    • normalization (Arabic)
    • normalization (Persian)
    • keyword
    • stemmer (Norwegian)

Custom Persian analyzer

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

  1. PUT /persian-index
  2. {
  3.   "settings": {
  4.     "analysis": {
  5.       "filter": {
  6.         "persian_stop": {
  7.           "type": "stop",
  8.           "stopwords": "_persian_"
  9.         },
  10.         "persian_keywords": {
  11.           "type": "keyword_marker",
  12.           "keywords": []
  13.         }
  14.       },
  15.       "char_filter": {
  16.         "null_width_replace_with_space": {
  17.             "type":       "mapping",
  18.             "mappings": [ "\\u200C=>\\u0020"] 
  19.         }
  20.       },
  21.       "analyzer": {
  22.         "persian_analyzer": {
  23.           "type": "custom",
  24.           "tokenizer": "standard",
  25.           "char_filter": [ "null_width_replace_with_space" ],
  26.           "filter": [
  27.             "lowercase",
  28.             "decimal_digit",
  29.             "arabic_normalization",
  30.             "persian_normalization",
  31.             "persian_stop"
  32.           ]
  33.         }
  34.       }
  35.     }
  36.   },
  37.   "mappings": {
  38.     "properties": {
  39.       "content": {
  40.         "type": "text",
  41.         "analyzer": "persian_analyzer"
  42.       }
  43.     }
  44.   }
  45. }

copy

Generated tokens

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

  1. POST /persian-index/_analyze
  2. {
  3.   "field": "content",
  4.   "text": "دانشجویان در دانشگاه‌های ایرانی تحصیل می‌کنند. شماره‌های آن‌ها ۱۲۳۴۵۶ است."
  5. }

copy

The response contains the generated tokens:

  1. {
  2.   "tokens": [
  3.     {"token": "دانشجويان","start_offset": 0,"end_offset": 9,"type": "<ALPHANUM>","position": 0},
  4.     {"token": "دانشگاه","start_offset": 13,"end_offset": 20,"type": "<ALPHANUM>","position": 2},
  5.     {"token": "ايراني","start_offset": 25,"end_offset": 31,"type": "<ALPHANUM>","position": 4},
  6.     {"token": "تحصيل","start_offset": 32,"end_offset": 37,"type": "<ALPHANUM>","position": 5},
  7.     {"token": "شماره","start_offset": 47,"end_offset": 52,"type": "<ALPHANUM>","position": 8},
  8.     {"token": "123456","start_offset": 63,"end_offset": 69,"type": "<NUM>","position": 12}
  9.   ]
  10. }