app_episodes-index-config.yaml

app_episodes-index-config.yaml

COPY (
select
  json_build_object(
    'episode_id', id,
    'title', REPLACE(
        REPLACE(title, '"', ''),
        '\n',''
      ),
    'subtitle', REPLACE(
        REPLACE(subtitle, '"', ''),
        '\n',''
      ),
    'description', REPLACE(
        REPLACE(description, '"', ''),
        '\n',''
      ),
    'publish_date', extract(epoch from publish_date)::bigint
  )
from episodes
) TO '/Volumes/Ram/app_episodes.json';


select
  json_build_object(
    'episode_id', id,
    'title', REGEXP_REPLACE(title, '[^\w\\n]+', '', 'n'),
    'subtitle', REGEXP_REPLACE(subtitle, '[^\w\\n]+', '', 'n'),
    'description', REGEXP_REPLACE(description, '[^\w\\n]+', '', 'n'),
    'publish_date', extract(epoch from publish_date)::bigint
  )
from episodes limit 10;

quickwit index ingest --index app_episodes --input-path app_episodes.json


version: 0.4
index_id: app_episodes
doc_mapping:
  field_mappings:
    - name: episode_id
      description: episode id
      type: u64
      stored: true
      indexed: true
      fast: true

    - name: title
      type: text
      tokenizer: default
      record: position
      stored: true

    - name: subtitle
      type: text
      tokenizer: default
      record: position
      stored: true

    - name: description
      type: text
      tokenizer: default
      record: position
      stored: true

    - name: publish_date
      type: datetime
      input_formats:
        - unix_timestamp
        - "%Y %m %d %H:%M:%S.%f"
      output_format: unix_timestamp_secs
      precision: seconds
      fast: true

search_settings:
  # If you do not specify fields in your query, those fields will be used.
  default_search_fields: [title, description]