Created
January 20, 2025 08:28
-
-
Save vitalibaranov/5f57379001808a14af927f0c916ef569 to your computer and use it in GitHub Desktop.
Create "persons" index with icu plugin person name handling
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| curl -XPUT "http://tms17.pyn.ru:7600/persons" -H 'Content-Type: application/json' -d' | |
| { | |
| "mappings": { | |
| "_source": { | |
| "enabled": false | |
| }, | |
| "properties": { | |
| "firstName": { | |
| "analyzer": "name_3gram_analyzer", | |
| "type": "text", | |
| "fields": { | |
| "latin": { | |
| "type": "text", | |
| "analyzer": "name_3gram_analyzer_translit" | |
| } | |
| } | |
| }, | |
| "lastName": { | |
| "analyzer": "name_3gram_analyzer", | |
| "type": "text", | |
| "fields": { | |
| "latin": { | |
| "type": "text", | |
| "analyzer": "name_3gram_analyzer_translit" | |
| } | |
| } | |
| }, | |
| "middleName": { | |
| "analyzer": "name_3gram_analyzer", | |
| "type": "text", | |
| "fields": { | |
| "latin": { | |
| "type": "text", | |
| "analyzer": "name_3gram_analyzer_translit" | |
| } | |
| } | |
| }, | |
| "fullName": { | |
| "type": "text", | |
| "analyzer": "lowercase_analyzer", | |
| "fields": { | |
| "latin": { | |
| "type": "text", | |
| "analyzer": "lowercase_analyzer_translit" | |
| } | |
| } | |
| }, | |
| "contacts": { | |
| "type": "keyword", | |
| "normalizer": "keyword_lowercase" | |
| }, | |
| "hhClientId": { | |
| "type": "integer" | |
| }, | |
| "cellPhone": { | |
| "analyzer": "contact_3gram_analyzer", | |
| "type": "text" | |
| }, | |
| "workPhone": { | |
| "analyzer": "contact_3gram_analyzer", | |
| "type": "text" | |
| }, | |
| "homePhone": { | |
| "analyzer": "contact_3gram_analyzer", | |
| "type": "text" | |
| }, | |
| "email": { | |
| "analyzer": "contact_3gram_analyzer", | |
| "type": "text" | |
| }, | |
| "consentStatus": { | |
| "type": "keyword" | |
| }, | |
| "sourceId": { | |
| "type": "integer" | |
| }, | |
| "vacancies": { | |
| "type": "integer" | |
| }, | |
| "tags": { | |
| "type": "integer" | |
| }, | |
| "mergeable": { | |
| "type": "boolean" | |
| }, | |
| "creationTime": { | |
| "type": "date" | |
| }, | |
| "editingTime": { | |
| "type": "date" | |
| }, | |
| "resumes": { | |
| "analyzer": "ru_full_text", | |
| "type": "text" | |
| }, | |
| "comments": { | |
| "analyzer": "ru_full_text", | |
| "type": "text" | |
| }, | |
| "hhComments": { | |
| "analyzer": "ru_full_text", | |
| "type": "text" | |
| }, | |
| "workflowStatuses": { | |
| "type": "keyword" | |
| }, | |
| "currentWorkflowStatuses": { | |
| "type": "keyword" | |
| }, | |
| "relocationType": { | |
| "type": "keyword" | |
| }, | |
| "area": { | |
| "type": "keyword" | |
| }, | |
| "relocationAreas": { | |
| "type": "keyword" | |
| }, | |
| "birthDay": { | |
| "type": "date" | |
| }, | |
| "discardReasons": { | |
| "type": "integer" | |
| }, | |
| "salary": { | |
| "properties": { | |
| "currency": { | |
| "type": "keyword" | |
| }, | |
| "amount": { | |
| "type": "integer" | |
| } | |
| } | |
| } | |
| } | |
| }, | |
| "settings": { | |
| "number_of_shards": "9", | |
| "number_of_replicas": "2", | |
| "index": { | |
| "search": { | |
| "slowlog": { | |
| "level": "info", | |
| "threshold": { | |
| "fetch": { | |
| "warn": "1s", | |
| "trace": "-1", | |
| "debug": "-1", | |
| "info": "400ms" | |
| }, | |
| "query": { | |
| "warn": "3s", | |
| "trace": "-1", | |
| "debug": "-1", | |
| "info": "2s" | |
| } | |
| } | |
| } | |
| }, | |
| "analysis": { | |
| "normalizer": { | |
| "keyword_lowercase": { | |
| "type": "custom", | |
| "filter": [ | |
| "lowercase" | |
| ] | |
| } | |
| }, | |
| "analyzer": { | |
| "lowercase_analyzer": { | |
| "type": "custom", | |
| "filter": [ | |
| "lowercase", | |
| "icu_transform_to_russian" | |
| ], | |
| "tokenizer": "standard" | |
| }, | |
| "lowercase_analyzer_translit": { | |
| "type": "custom", | |
| "filter": [ | |
| "lowercase", | |
| "icu_transform_to_russian" | |
| ], | |
| "tokenizer": "standard" | |
| }, | |
| "contact_3gram_analyzer": { | |
| "type": "custom", | |
| "filter": [ | |
| "lowercase", | |
| "icu_transform_to_russian" | |
| ], | |
| "tokenizer": "contact_3gram" | |
| }, | |
| "name_3gram_analyzer": { | |
| "type": "custom", | |
| "char_filter": "mapping_filter", | |
| "filter": [ | |
| "lowercase", | |
| "icu_transform_to_russian" | |
| ], | |
| "tokenizer": "name_3gram" | |
| }, | |
| "name_3gram_analyzer_translit": { | |
| "type": "custom", | |
| "char_filter": "mapping_filter", | |
| "filter": [ | |
| "lowercase", | |
| "icu_transform_to_latin" | |
| ], | |
| "tokenizer": "name_3gram" | |
| }, | |
| "ru_full_text": { | |
| "type": "custom", | |
| "char_filter": [ | |
| "html_strip", | |
| "mapping_filter" | |
| ], | |
| "filter": [ | |
| "lowercase", | |
| "russian_stop", | |
| "russian_stemmer" | |
| ], | |
| "tokenizer": "standard" | |
| } | |
| }, | |
| "char_filter": { | |
| "mapping_filter": { | |
| "mappings": [ | |
| "ё => е" | |
| ], | |
| "type": "mapping" | |
| } | |
| }, | |
| "filter": { | |
| "russian_stemmer": { | |
| "language": "russian", | |
| "type": "stemmer" | |
| }, | |
| "russian_stop": { | |
| "stopwords": "_russian_", | |
| "type": "stop" | |
| }, | |
| "icu_transform_to_latin": { | |
| "type": "icu_transform", | |
| "id": "Russian-Latin/BGN" | |
| }, | |
| "icu_transform_to_russian": { | |
| "type": "icu_transform", | |
| "id": "Latin-Russian/BGN" | |
| } | |
| }, | |
| "tokenizer": { | |
| "contact_3gram": { | |
| "max_gram": 3, | |
| "min_gram": 3, | |
| "token_chars": [], | |
| "type": "ngram" | |
| }, | |
| "name_3gram": { | |
| "max_gram": 3, | |
| "min_gram": 3, | |
| "token_chars": [ | |
| "letter", | |
| "digit" | |
| ], | |
| "type": "ngram" | |
| } | |
| } | |
| } | |
| } | |
| } | |
| }' |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment