Schema definition

Schema definition file

[
    {
        "database": "<Postgres database name>",
        "index": "<Elasticsearch index name>",
        "setting": "<Elasticsearch setting>",
        "plugins": ["<Plugin A>", "<Plugin B>"...],
        "pipeline": "<pipeline>",
        "routing": "<routing>",
        "nodes": {
            "table": "<root table name>",
            "schema": "<schema name>",
            "columns": [
                "<column 1>",
                "<column 2>",
                ...
            ],
            "children": [
                {
                    "table": "<child table name>",
                    "columns": [
                        "<column 1>",
                        "<column 2>",
                        ...
                    ],
                    "label": "<document label name>",
                    "relationship": {
                        "variant": "object" | "scalar",
                        "type": "one_to_one" | "one_to_many",
                        "through_tables": [
                            "<through table name>"
                        ]
                    },
                    "children": [],
                    "transform": {
                        "rename": {
                            "<old column 1>": "<new column 1>",
                            "<old column 2>": "<new column 2>",
                            ...
                        },
                        "mapping": {
                            "<new column 1>": {"<data type>"},
                            "<new column 2>": {"<data type>"},
                            ...
                        },
                        "concat": {
                            "columns": ["column 1", "column 2" ...],
                            "destination": "<new column 1>",
                            "delimiter": "<char>"
                        }
                    }
                },
                ...
            ]
        }
    }
]

Document and node structure:

database

This is the database name

index

An optional Elasticsearch index (defaults to database name)

nodes

An object node describing the Elasticsearch document

setting

Elasticsearch setting configuration

{
    "setting": {
        "analysis": {
            "analyzer": {
                "ngram_analyzer": {
                    "filter": [
                        "lowercase"
                    ],
                    "type": "custom",
                    "tokenizer": "ngram_tokenizer"
                }
            },
            "tokenizer": {
                "ngram_tokenizer": {
                    "token_chars": [
                        "letter",
                        "digit",
                        "punctuation",
                        "symbol"
                    ],
                    "min_gram": "9",
                    "type": "nGram",
                    "max_gram": "10"
                }
            }
        }
    }
}

table

Node table name

schema

An optional Postgres table schema (defaults to public)

label

An optional node name in Elasticsearch (defaults to table name)

columns

An optional list of columns to display. This can be omitted in which case it selects all columns.

children

An optional list of child nodes if any. This has the same structure as a parent node.

relationship

Describes the relationship between parent and child.

• variant

  variant can be object or scalar

  • object

    {
        "name": "Oxford Press",
        "id": 1,
        "is_active": false
    }
    

  • scalar

    ["Haruki Murakami", "Philip Gabriel"]
    

• type

  type can be one_to_one or one_to_many depending on the relationship type between parent and child

• through_tables

  This is the intermediate table that connects the parent to the child

transform

This allows transforming some node properties. For now, the only operation supported is the rename transform.

• rename

  rename a node column

      "rename": {
          "<old column name 1>": "<new column name 1>",
          "<old column name 2>": "<new column name 2>",
      }
  

• mapping

  Specify Elasticsearch mapping

      "mapping": {
           "book_id": {
              "type": "long"
          },
          "book_isbn": {
              "type": "long",
              "fields":{
                  "ngram": {
                      "type": "text",
                      "analyzer": "ngram_analyzer",
                      "search_analyzer": "ngram_analyzer",
                      "fielddata": true
                  }
              }
          }
      }
  

pipeline

Optional injest pipeline

routing

Optional routing field

Info

Changing the schema effectively changes the structure of the document in Elasticsearch and this requires re-indexing Elasticsearch.

See the advanced section on re-indexing on how-to.