gistfile1.sh

#!/bin/bash

# Version info from elasticsearch:9200
#{
# "ok" : true,
# "status" : 200,
# "name" : "Tiger Shark",
# "version" : {
#   "number" : "0.90.2",
#   "snapshot_build" : false,
#   "lucene_version" : "4.3.1"
# },
# "tagline" : "You Know, for Search"
#}

ES='http://elasticsearch:9200'
INDEX='my_new_index'
MAPPING='my_mapping'

# create index
curl -XDELETE $ES/$INDEX
curl -XPUT $ES/$INDEX/ -d '{
        "settings": {
            "index": {
                "analysis": {
                    "analyzer": {
                        "my_custom_analyzer": {
                            "type": "custom",
                            "tokenizer": "keyword",
                            "filter": ["asciifolding", "lowercase"]
                        }
                    }
                }
            }
        }
    }'

# create mapping
curl -XPUT $ES/$INDEX/$MAPPING/_mapping -d '{
        "properties": {
            "guid": {"type": "string", "index": "no"},
            "name": {
                "type": "string",
                "index": "analyzed",
                "analyzer": "my_custom_analyzer"
            }
        }
}'

# put some data
curl -XPUT $ES/$INDEX/$MAPPING/1 -d '{
    "guid": "some_guid",
    "name": "politica"
}'

curl -XPUT $ES/$INDEX/$MAPPING/2 -d '{
    "guid": "another_guid",
    "name": "política"
}'

# get data
# this matches only "politica" and not "política"
curl -XPOST $ES/$INDEX/$MAPPING/_search?pretty=true -d '{
    "query": {
        "prefix": {
            "name" : {
                "prefix": "poli"
            }
        }
    }
}'

# Both the following queries show this result:
# (which proves my analyzer is working):
#{
#  "tokens" : [ {
#    "token" : "politica",
#    "start_offset" : 0,
#    "end_offset" : 8,
#    "type" : "word",
#    "position" : 1
#  } ]
#}
curl -XGET 'elasticsearch:9200/my_new_index/_analyze?analyzer=my_custom_analyzer' -d 'política'
curl -XGET 'elasticsearch:9200/my_new_index/_analyze?analyzer=my_custom_analyzer' -d 'politica'

#But when I run
curl -XGET 'elasticsearch:9200/my_new_index/_analyze?field=my_mapping.name&pretty=true' -d 'política'
# It doesn't work (the accented character is still there)
#{
#  "tokens" : [ {
#    "token" : "política",
#    "start_offset" : 0,
#    "end_offset" : 8,
#    "type" : "<ALPHANUM>",
#    "position" : 1
#  } ]
#}