Mapping character filter
editMapping character filter
editThe mapping character filter accepts a map of keys and values. Whenever it
encounters a string of characters that is the same as a key, it replaces them
with the value associated with that key.
Matching is greedy; the longest pattern matching at a given point wins. Replacements are allowed to be the empty string.
The mapping filter uses Lucene’s
MappingCharFilter.
Example
editThe following analyze API request uses the mapping filter
to convert Hindu-Arabic numerals (٠١٢٣٤٥٦٧٨٩) into their Arabic-Latin
equivalents (0123456789), changing the text My license plate is ٢٥٠١٥ to
My license plate is 25015.
response = client.indices.analyze(
body: {
tokenizer: 'keyword',
char_filter: [
{
type: 'mapping',
mappings: [
'٠ => 0',
'١ => 1',
'٢ => 2',
'٣ => 3',
'٤ => 4',
'٥ => 5',
'٦ => 6',
'٧ => 7',
'٨ => 8',
'٩ => 9'
]
}
],
text: 'My license plate is ٢٥٠١٥'
}
)
puts response
GET /_analyze
{
"tokenizer": "keyword",
"char_filter": [
{
"type": "mapping",
"mappings": [
"٠ => 0",
"١ => 1",
"٢ => 2",
"٣ => 3",
"٤ => 4",
"٥ => 5",
"٦ => 6",
"٧ => 7",
"٨ => 8",
"٩ => 9"
]
}
],
"text": "My license plate is ٢٥٠١٥"
}
The filter produces the following text:
[ My license plate is 25015 ]
Configurable parameters
edit-
mappings -
(Required*, array of strings) Array of mappings, with each element having the form
key => value.Either this or the
mappings_pathparameter must be specified. -
mappings_path -
(Required*, string) Path to a file containing
key => valuemappings.This path must be absolute or relative to the
configlocation, and the file must be UTF-8 encoded. Each mapping in the file must be separated by a line break.Either this or the
mappingsparameter must be specified.
Customize and add to an analyzer
editTo customize the mappings filter, duplicate it to create the basis for a new
custom character filter. You can modify the filter using its configurable
parameters.
The following create index API request
configures a new custom analyzer using a custom
mappings filter, my_mappings_char_filter.
The my_mappings_char_filter filter replaces the :) and :( emoticons
with a text equivalent.
response = client.indices.create(
index: 'my-index-000001',
body: {
settings: {
analysis: {
analyzer: {
my_analyzer: {
tokenizer: 'standard',
char_filter: [
'my_mappings_char_filter'
]
}
},
char_filter: {
my_mappings_char_filter: {
type: 'mapping',
mappings: [
':) => _happy_',
':( => _sad_'
]
}
}
}
}
}
)
puts response
PUT /my-index-000001
{
"settings": {
"analysis": {
"analyzer": {
"my_analyzer": {
"tokenizer": "standard",
"char_filter": [
"my_mappings_char_filter"
]
}
},
"char_filter": {
"my_mappings_char_filter": {
"type": "mapping",
"mappings": [
":) => _happy_",
":( => _sad_"
]
}
}
}
}
}
The following analyze API request uses the custom
my_mappings_char_filter to replace :( with _sad_ in
the text I'm delighted about it :(.
GET /my-index-000001/_analyze
{
"tokenizer": "keyword",
"char_filter": [ "my_mappings_char_filter" ],
"text": "I'm delighted about it :("
}
The filter produces the following text:
[ I'm delighted about it _sad_ ]