Query URL Parameters #

The query URL parameters can be used in many places (e.g., HTTP API endpoints, internal service calls, debug consoles). We use them to unify how queries are received and processed across different components of the system. This provides a powerful, composable, and human-readable way to construct both full-text search and structured filters, while also supporting advanced features like fuzziness, field selection, and pagination.

🔧 Query URL Parameters #

These URL parameters are used to construct a rich and dynamic search query.

Name	Type	Description	Example
`query`	`string`	The main query string. Supports field boosting (`field^boost:value`).	`query=title^2:search engine`
`filter`	`string[]`	One or more filter clauses. Can be negated with `-` or `!`. Supports:	`filter=status:active`, `filter=-exists(deleted_at)`
		- `field=value`, `field!=value`, `field>=x`, `field<y`	`filter=age>=18`, `filter=tag!=archived`
		- `exists(field)`	`filter=exists(status)`
`sort`	`string`	Sort rules separated by comma. Each rule is `field[:asc\|desc]`.	`sort=published_at:desc,_score`
`from`	`int`	Pagination offset.	`from=20`
`size`	`int`	Number of results to return.	`size=10`
`fuzziness`	`int`	Fuzziness level for the query (0–5).	`fuzziness=3`
`default_operator`	`string`	Operator between terms if not specified (`AND` or `OR`).	`default_operator=AND`
`default_fields`	`string`	Comma-separated list of fields used as fallback for both query and filter.	`default_fields=title,description`
`default_query_fields`	`string`	Comma-separated list of fields used only for full-text search.	`default_query_fields=title,body`
`default_filter_fields`	`string`	Comma-separated list of fields used only for filters.	`default_filter_fields=status,tag`
`_source_includes`	`string`	Comma-separated fields to include in `_source`.	`_source_includes=title,author`
`_source_excludes`	`string`	Comma-separated fields to exclude from `_source`.	`_source_excludes=internal_notes,raw_data`

🧠 Filter Syntax Summary #

Syntax	Meaning	Example
`field=value`	Term query	`status=active`
`field!=value`	Negated term	`status!=deleted`
`field>=value`	Range query (greater than or equal)	`views>=1000`
`field<value`	Range query (less than)	`age<30`
`any(value1,value2)`	Terms filter (any term can be match)	`any(tag1,tag2,tag3)`
`exists(field)`	Field existence check	`exists(tags)`
`-filterExpr` / `!filterExpr`	Negate any filter expression	`!exists(deleted_at)`

🚀 Example #

Simple Search #

GET /search?query=go+language&default_fields=title,description

Filtered and Sorted #

GET /search?query=distributed+search&filter=status=active&sort=_score:desc,created_at:desc

Advanced with Fuzziness and Range #

GET /search?query=search&fuzziness=3&filter=age>=18&filter=!exists(deleted_at)&size=20

🧩 Notes #

Filters are ANDed together by default.
Negated filters use either - or ! prefix.
If query is not provided, only filters will be applied.
Supports combining query string logic and JSON body input (BuildQueryDSLOnTopOfDSL handles merging).
Values in filters are auto-casted to int, bool, or string.

Example usage:

	q := orm.NewQuery().Must(
		orm.ShouldQuery(
			orm.MatchQuery("lang", "en"),
			orm.MatchQuery("lang", "zh"),
		),
		orm.MustNotQuery(
			orm.TermQuery("deleted", true),
		),
	).Size(10).SortBy(
		orm.Sort{Field: "score", SortType: orm.DESC},
	)