data structure schema-OpenSearch(Open Search)-阿里云帮助中心

A schema defines the data structure of an OpenSearch application. It specifies which tables store your data, how fields are typed within those tables, and which fields are indexed for full-text search or filtering.

All configuration lives under two top-level keys: tables (your data) and indexes (how that data is searchable).

Schema structure

Field	Type	Description
`tables`	Object	The tables in the application. Each key under `tables` is a table name.
`tables.*`	Object	The definition of a table. See Table.
`indexes`	Object	The index configuration for the application. See Index.

Example

{
    "tables": {
        "main": {
            "primaryTable": true,
            "name": "main",
            "fields": {
                "id": {
                    "name": "id",
                    "type": "LITERAL",
                    "primaryKey": true
                },
                "title": {
                    "name": "title",
                    "type": "TEXT",
                    "primaryKey": false
                },
                "buy": {
                    "name": "buy",
                    "type": "INT",
                    "primaryKey": false
                },
                "cate_id": {
                    "name": "cate_id",
                    "type": "INT",
                    "primaryKey": false
                },
                "cate_name": {
                    "name": "cate_name",
                    "type": "LITERAL",
                    "primaryKey": false
                }
            }
        }
    },
    "indexes": {
        "searchFields": {
            "id": {
                "fields": [
                    "id"
                ]
            },
            "default": {
                "fields": [
                    "title"
                ],
                "analyzer": "chn_standard"
            },
            "cate_name": {
                "fields": [
                    "cate_name"
                ]
            }
        },
        "filterFields": [
            "id",
            "buy",
            "cate_id",
            "cate_name"
        ]
    }
}

Table

A table maps to a data source in your application. One table must be designated as the primary table.

Field	Type	Description
`name`	String	The name of the table.
`primaryTable`	Boolean	Indicates whether the table is the primary table.
`fields`	Object	The fields in the table. Each key under `fields` is a field name.
`fields.*`	Object	The definition of a field. See Field.

Example

{
    "primaryTable": true,
    "name": "main",
    "fields": {
        "id": {
            "name": "id",
            "type": "LITERAL",
            "primaryKey": true
        },
        "title": {
            "name": "title",
            "type": "TEXT",
            "primaryKey": false
        },
        "buy": {
            "name": "buy",
            "type": "INT",
            "primaryKey": false
        },
        "cate_id": {
            "name": "cate_id",
            "type": "INT",
            "primaryKey": false
        },
        "cate_name": {
            "name": "cate_name",
            "type": "LITERAL",
            "primaryKey": false
        }
    }
}

Field

A field represents a column in your table. Its type determines how the data is stored and which index operations are supported.

Field	Type	Description
`name`	String	The name of the field.
`type`	String	The data type of the field. For supported types, see Schema of tables for Industry Algorithm Edition.
`primaryKey`	Boolean	Indicates whether the field is a primary key field.
`joinWith`	Array	The tables that are joined to external tables.

Example

{
    "type": "INT",
    "name": "id",
    "primaryKey": true,
    "joinWith": ["other_table"]
}

Index

The indexes object controls two types of indexing:

`searchFields`: Fields indexed for full-text search. Use these when you need to search across text content.
`filterFields`: Fields indexed as attributes for filtering and sorting. Use these for numeric comparisons, category filtering, and range queries.

Field	Type	Description
`searchFields`	Object	Index fields for full-text search. Each key is an index field name.
`searchFields.*`	Object	The definition of a search index field. See Search field.
`filterFields`	Array	Field names to index as attribute (filter) fields.

Example

{
    "searchFields": {
        "default": {
            "fields": [
                "title"
            ],
            "analyzer": "chn_standard"
        },
        "id": {
            "fields": [
                "id"
            ]
        }
    },
    "filterFields": [
        "id"
    ]
}

Search field

A search field defines how one or more table fields are combined into a single full-text index.

Field	Type	Description
`fields`	Array	The list of index fields.
`analyzer`	String	The name of the analyzer, such as a custom analyzer or built-in analyzer. For custom analyzers, see Custom analyzers. For built-in options, see Built-in analyzers.

Example

{
    "fields": ["title"],
    "analyzer": "chn_standard"
}

Built-in analyzers

Choose an analyzer based on the language and search behavior you need.

Analyzer	Use when
`chn_standard`	Searching Chinese text in general scenarios (e-commerce, news, documentation).
`simple`	Searching content in specific scenarios using basic tokenization.
`chn_single`	Searching by individual Chinese characters, such as for character-level matching.
`eng_standard`	Searching English text with stemming — `running` matches `run`, `runs`, and `ran`.
`eng_nostem`	Searching English text while retaining the roots of words.
`fuzzy`	Tolerating minor spelling variations and typos in search queries.
`keyword`	Matching the exact string without any tokenization — suitable for IDs, codes, and tags.
`chn_ecommerce`	Searching product names, SKUs, and descriptions in Chinese e-commerce contexts.
`chn_film`	Searching Chinese titles, cast names, and descriptions in video or media contexts.
`chn_scene_name`	Recognizing and matching Chinese person names.
`chn_scene_org`	Recognizing and matching Chinese organization names.
`first_letter`	Searching by the first letter of each pinyin syllable (simple pinyin spelling).
`full_pinyin`	Searching by the full pinyin spelling of Chinese characters.
`numeric`	Performing range searches on numeric values.
`geo`	Performing range searches based on geographical location.
`chn_it_content`	Searching Chinese technical content, such as IT documentation and product descriptions.