gistfile1.txt

import synonyms from './synonyms'
// Keywords that should not be stemmed
const swedishKeywords = ['senior', 'island']

const indexConfig = {
  settings: {
    index: {
      'mapping.total_fields.limit': 2200,
      number_of_shards: 1,
      number_of_replicas: 2,
    },
    analysis: {
      normalizer: {
        downcased_and_folded: {
          type: 'custom',
          filter: ['lowercase', 'asciifolding'],
        },
        downcased: {
          type: 'custom',
          filter: ['lowercase'],
        },
        alpha_numeric: {
          type: 'custom',
          char_filter: ['alpha_numeric_filter'],
          filter: ['lowercase'],
        },
      },
      filter: {
        synonyms: {
          type: 'synonym',
          synonyms,
        },
        swedish_stop: {
          type: 'stop',
          stopwords: '_swedish_',
        },
        swedish_stemmer: {
          type: 'stemmer',
          language: 'swedish',
        },
        swedish_keywords: {
          type: 'keyword_marker',
          keywords: swedishKeywords,
        },
        '22_char_hyphenation_decompound': {
          type: 'hyphenation_decompounder',
          hyphenation_patterns_path: '/elasticsearch/config/analysis/se.xml',
          word_list_path: '/elasticsearch/config/analysis/swedish_words.txt',
          max_subword_size: 22,
          min_subword_size: 4,
        },
      },
      char_filter: {
        alpha_numeric_filter: {
          type: 'pattern_replace',
          pattern: '[^A-ZÅÄÖa-zåäö0-9 ]',
          replacement: '',
        },
      },
      analyzer: {
        custom_swedish: {
          tokenizer: 'standard',
          filter: [
            'lowercase',
            // 'swedish_stop',
            'swedish_keywords',
            'swedish_stemmer',
            'synonyms',
          ],
        },
        swedish_decompounder: {
          tokenizer: 'standard',
          filter: [
            'lowercase',
            // 'swedish_stop',
            'swedish_keywords',
            'swedish_stemmer',
            'synonyms',
            '22_char_hyphenation_decompound',
          ],
        },
      },
    },
  },
  mappings: {
    _doc: {
      dynamic_templates: [
        {
          bag_of_dates: {
            match: '*_time',
            mapping: { type: 'date' },
          },
        },
        {
          bag_of_sort_dates: {
            match: '*_date',
            mapping: { type: 'date' },
          },
        },
        {
          files: {
            match: 'files.*',
            mapping: { type: 'keyword' },
          },
        },
        {
          swedish: {
            path_match: '*_sv',
            mapping: {
              type: 'text',
              analyzer: 'swedish_decompounder',
              search_analyzer: 'custom_swedish',
              search_quote_analyzer: 'simple',
            },
          },
        },
        {
          normalized: {
            path_match: '*_normalized',
            mapping: {
              type: 'keyword',
              normalizer: 'downcased_and_folded',
            },
          },
        },
        {
          downcased: {
            path_match: '*_downcased',
            mapping: {
              type: 'keyword',
              normalizer: 'downcased',
            },
          },
        },
        {
          facets: {
            match_pattern: 'regex',
            match: '^(language_level)|(keywords)|(subject_tree)|(categories)$',
            mapping: {
              type: 'keyword',
              copy_to: ['{name}_normalized', '{name}_downcased', '{name}_sv'],
            },
          },
        },
        {
          titles: {
            match_pattern: 'regex',
            match: '(.*title)|(usp)$',
            mapping: {
              type: 'text',
              analyzer: 'swedish_decompounder',
              search_analyzer: 'custom_swedish',
              search_quote_analyzer: 'simple',
              copy_to: ['{name}_sort', 'all_titles'],
            },
          },
        },
        {
          sorts: {
            path_match: '*_sort',
            mapping: { type: 'keyword', normalizer: 'alpha_numeric' },
          },
        },
        {
          languages: {
            match_pattern: 'regex',
            match:
              '^(languages)|(languages_translations)|(subtitle_languages)$',
            mapping: {
              type: 'keyword',
              copy_to: [
                '{name}_normalized',
                '{name}_downcased',
                '{name}_sv',
                'all_languages_and_codes_normalized',
              ],
            },
          },
        },
      ],
      properties: {
        description: { type: 'text', analyzer: 'swedish' },
        episode_number: { type: 'integer' },
        number_of_episodes: { type: 'integer' },
        duration: { type: 'integer' },
        product_type: { type: 'keyword' },
        format: { type: 'keyword' },
        typical_age_range: { type: 'keyword' },
        has_teacher_resource: { type: 'boolean' },
        published_at: { type: 'date' },
        subject_tree_raw: {
          type: 'keyword',
          normalizer: 'downcased_and_folded',
        },
        participants: {
          type: 'nested',
          properties: {
            firstname: {
              type: 'keyword',
              copy_to: ['firstname_sv', 'fullname'],
            },
            lastname: {
              type: 'keyword',
              copy_to: ['lastname_sv', 'fullname'],
            },
            fullname: { type: 'keyword', copy_to: 'fullname_sv' },
            role: { type: 'keyword' },
            profession: { type: 'keyword' },
          },
        },
        main_genre: { type: 'keyword' },
        emotional_tags: { type: 'keyword' },
        modified: { type: 'date' },
        subtitle_languages: { type: 'keyword' },
        image: { type: 'object', enabled: false },
        age_ranges: {
          type: 'nested',
          properties: {
            from: { type: 'integer' },
            to: { type: 'integer' },
          },
        },
        accessible_platforms: {
          type: 'nested',
          dynamic: true,
        },
        platforms: { type: 'keyword' },
        broadcasts: {
          type: 'nested',
          dynamic: true,
          properties: {
            platform: { type: 'keyword' },
          },
        },
        files: {
          type: 'nested',
          dynamic: true,
        },
        pod_info: {
          type: 'object',
          properties: {
            is_downloadable: { type: 'boolean' },
            file: { type: 'keyword' },
            filename: { type: 'keyword' },
            format: { type: 'keyword' },
          },
        },
        sab: { type: 'keyword' },
        streaming_info: {
          type: 'object',
          enabled: false,
        },
        seo_description: {
          type: 'text',
          analyzer: 'swedish_decompounder',
          search_analyzer: 'custom_swedish',
          search_quote_analyzer: 'simple',
        },
        only_in_sweden: { type: 'boolean' },
        parental_lock: { type: 'boolean' },
        is_sign_laguage_interpreted: { type: 'boolean' },
        is_audio_described: { type: 'boolean' },
        easy_to_read: { type: 'boolean' },
        super_series_id: { type: 'integer' },
        series_id: { type: 'integer' },
        program_ids: { type: 'integer' },
        product_views: { type: 'integer' },
      },
    },
  },
}

export default indexConfig