elasticsearch.yml

# multilingual analysis for title search taken from :
# http://jprante.github.com/lessons/2012/05/16/multilingual-analysis-for-title-search.html
# or alternatively here (where the used plugin comes from, is less descriptive but has more technical details)
# https://github.com/yakaz/elasticsearch-analysis-combo
analysis:
  filter:
    germansnow:
      type: snowball
      language: German2
    ngram_filter:
      type: nGram
      max_gram: 10
      min_gram: 3
    url_stop:
      type: "stop"
      stopwords: ["http", "https"]
  analyzer:
    stripped_standard:
      type: custom
      tokenizer: standard
      filter:
      - standard
      - lowercase
      - stop
      char_filter:
      - html_strip
    url:
      type: custom
      tokenizer: lowercase
      filter: ["stop", "url_stop"]
    ngram_analyzer:
      type: custom
      tokenizer: standard
      filter:
      - ngram_filter
      char_filter:
      - html_strip
    german_text:
      type: custom
      tokenizer: standard
      filter:
      - germansnow
      char_filter:
      - html_strip
    default_index:
      type: combo
      sub_analyzers:
      - stripped_standard
      - german_text
      - ngram_analyzer
    default_search:
      type: combo
      sub_analyzers:
      - stripped_standard
      - german_text