[
https://issues.apache.org/jira/browse/CALCITE-7300?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18041036#comment-18041036
]
Oleg Alekseev edited comment on CALCITE-7300 at 11/27/25 11:17 AM:
-------------------------------------------------------------------
For now I am still evaluating this functionality and possible design options,
so it is not implemented yet.
## REST API Filter Specification Variants
Different REST APIs use different structures for expressing filters. The
adapter automatically detects the supported format from the OpenAPI
specification. Below are the supported variants:
h3. Variant 1: DNF with "where" (AND) + "or" (OR groups)
*Description:* REST API accepts:
* *where*: array of filter criteria joined by AND (common conditions from first
DNF group)
* *or*: array of arrays, where each inner array contains criteria joined by
AND, and outer array represents OR
*OpenAPI specification:*
{code:language=yaml}
requestBody:
content:
application/json:
schema:
properties:
where:
type: array
items:
$ref: '#/components/schemas/FilterCriterion'
description: List of AND filter criteria
or:
type: array
items:
type: array
items:
$ref: '#/components/schemas/FilterCriterion'
description: DNF - List of OR groups
{code}
*Example scenario:*
REST API contains the following users:
{code:language=json}
[
\{ "id": 1, "name": "Bob", "lastName": "Smith", "age": 25, "department": "IT"
},
\{ "id": 2, "name": "Alice", "lastName": "Johnson", "age": 22, "department":
"HR" },
\{ "id": 3, "name": "Bob", "lastName": "Brown", "age": 19, "department": "IT"
},
\{ "id": 4, "name": "Martin", "lastName": "Smith", "age": 35, "department":
"Sales" },
\{ "id": 5, "name": "Alice", "lastName": "Smith", "age": 28, "department":
"IT" },
\{ "id": 6, "name": "John", "lastName": "Doe", "age": 30, "department": "HR"
},
\{ "id": 7, "name": "Sarah", "lastName": "Smith", "age": 26, "department":
"IT" }
]
{code}
*SQL Query with nested conditions:*
{code:language=sql}
SELECT * FROM users
WHERE (
(name = 'Bob' AND age >= 21)
OR
(name = 'Alice' AND (department = 'IT' OR age >= 25))
)
AND
(lastName = 'Smith' OR (department = 'IT' AND age < 30))
{code}
*DNF Expansion:*
{code}
(name = 'Bob' AND age >= 21 AND lastName = 'Smith')
OR (name = 'Bob' AND age >= 21 AND department = 'IT' AND age < 30)
OR (name = 'Alice' AND department = 'IT' AND lastName = 'Smith')
OR (name = 'Alice' AND department = 'IT' AND department = 'IT' AND age < 30)
OR (name = 'Alice' AND age >= 25 AND lastName = 'Smith')
OR (name = 'Alice' AND age >= 25 AND department = 'IT' AND age < 30)
{code}
*Generated REST request body:*
{code:language=json}
{
"page": 0,
"limit": 100,
"where": [
\{ "name": "name", "operator": "=", "value": "Bob" },
\{ "name": "age", "operator": ">=", "value": 21 },
\{ "name": "lastName", "operator": "=", "value": "Smith" }
],
"or": [
[
\{ "name": "name", "operator": "=", "value": "Bob" },
\{ "name": "age", "operator": ">=", "value": 21 },
\{ "name": "department", "operator": "=", "value": "IT" },
\{ "name": "age", "operator": "<", "value": 30 }
],
[
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "department", "operator": "=", "value": "IT" },
\{ "name": "lastName", "operator": "=", "value": "Smith" }
],
[
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "department", "operator": "=", "value": "IT" },
\{ "name": "age", "operator": "<", "value": 30 }
],
[
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "age", "operator": ">=", "value": 25 },
\{ "name": "lastName", "operator": "=", "value": "Smith" }
],
[
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "age", "operator": ">=", "value": 25 },
\{ "name": "department", "operator": "=", "value": "IT" },
\{ "name": "age", "operator": "<", "value": 30 }
]
]
}
{code}
*Result:* Returns users with id: 1 (Bob Smith, 25, IT) and 5 (Alice Smith, 28,
IT)
----
h3. Variant 2: DNF with "or" only (no "where")
*Description:* REST API accepts:
* *or*: array of arrays (DNF structure), no separate "where" field
*OpenAPI specification:*
{code:language=yaml}
requestBody:
content:
application/json:
schema:
properties:
or:
type: array
items:
type: array
items:
$ref: '#/components/schemas/FilterCriterion'
description: DNF - List of OR groups (each group is ANDed)
{code}
*Example scenario:*
REST API contains the following users:
{code:language=json}
[
\{ "id": 1, "name": "Bob", "lastName": "Smith", "age": 25, "department": "IT"
},
\{ "id": 2, "name": "Alice", "lastName": "Johnson", "age": 22, "department":
"HR" },
\{ "id": 3, "name": "Bob", "lastName": "Brown", "age": 19, "department": "IT"
},
\{ "id": 4, "name": "Martin", "lastName": "Smith", "age": 35, "department":
"Sales" },
\{ "id": 5, "name": "Alice", "lastName": "Smith", "age": 28, "department":
"IT" },
\{ "id": 6, "name": "John", "lastName": "Doe", "age": 30, "department": "HR"
},
\{ "id": 7, "name": "Sarah", "lastName": "Smith", "age": 26, "department":
"IT" }
]
{code}
*SQL Query with nested conditions:*
{code:language=sql}
SELECT * FROM users
WHERE (
(name = 'Bob' AND age >= 21)
OR
(name = 'Alice' AND (department = 'IT' OR age >= 25))
)
AND
(lastName = 'Smith' OR (department = 'IT' AND age < 30))
{code}
*DNF Expansion:*
{code}
(name = 'Bob' AND age >= 21 AND lastName = 'Smith')
OR (name = 'Bob' AND age >= 21 AND department = 'IT' AND age < 30)
OR (name = 'Alice' AND department = 'IT' AND lastName = 'Smith')
OR (name = 'Alice' AND department = 'IT' AND department = 'IT' AND age < 30)
OR (name = 'Alice' AND age >= 25 AND lastName = 'Smith')
OR (name = 'Alice' AND age >= 25 AND department = 'IT' AND age < 30)
{code}
*Generated REST request body:*
{code:language=json}
{
"page": 0,
"limit": 100,
"or": [
[
\{ "name": "name", "operator": "=", "value": "Bob" },
\{ "name": "age", "operator": ">=", "value": 21 },
\{ "name": "lastName", "operator": "=", "value": "Smith" }
],
[
\{ "name": "name", "operator": "=", "value": "Bob" },
\{ "name": "age", "operator": ">=", "value": 21 },
\{ "name": "department", "operator": "=", "value": "IT" },
\{ "name": "age", "operator": "<", "value": 30 }
],
[
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "department", "operator": "=", "value": "IT" },
\{ "name": "lastName", "operator": "=", "value": "Smith" }
],
[
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "department", "operator": "=", "value": "IT" },
\{ "name": "age", "operator": "<", "value": 30 }
],
[
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "age", "operator": ">=", "value": 25 },
\{ "name": "lastName", "operator": "=", "value": "Smith" }
],
[
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "age", "operator": ">=", "value": 25 },
\{ "name": "department", "operator": "=", "value": "IT" },
\{ "name": "age", "operator": "<", "value": 30 }
]
]
}
{code}
*Result:* Returns users with id: 1 (Bob Smith, 25, IT) and 5 (Alice Smith, 28,
IT)
----
h3. Variant 3: CNF with "where" (OR) + "and" (AND groups)
*Description:* REST API accepts:
* *where*: array of filter criteria joined by OR (conditions from first CNF
group)
* *and*: array of arrays, where each inner array contains criteria joined by
OR, and outer array represents AND
*OpenAPI specification:*
{code:language=yaml}
requestBody:
content:
application/json:
schema:
properties:
where:
type: array
items:
$ref: '#/components/schemas/FilterCriterion'
description: List of OR filter criteria
and:
type: array
items:
type: array
items:
$ref: '#/components/schemas/FilterCriterion'
description: CNF - List of AND groups (each group contains OR
criteria)
{code}
*Example scenario:*
REST API contains the following users:
{code:language=json}
[
\{ "id": 1, "name": "Bob", "lastName": "Smith", "age": 25, "department": "IT"
},
\{ "id": 2, "name": "Alice", "lastName": "Johnson", "age": 22, "department":
"HR" },
\{ "id": 3, "name": "Bob", "lastName": "Brown", "age": 19, "department": "IT"
},
\{ "id": 4, "name": "Martin", "lastName": "Smith", "age": 35, "department":
"Sales" },
\{ "id": 5, "name": "Alice", "lastName": "Smith", "age": 28, "department":
"IT" },
\{ "id": 6, "name": "John", "lastName": "Doe", "age": 30, "department": "HR"
},
\{ "id": 7, "name": "Sarah", "lastName": "Smith", "age": 26, "department":
"IT" }
]
{code}
*SQL Query with nested conditions:*
{code:language=sql}
SELECT * FROM users
WHERE (
(name = 'Bob' AND age >= 21)
OR
(name = 'Alice' AND (department = 'IT' OR age >= 25))
)
AND
(lastName = 'Smith' OR (department = 'IT' AND age < 30))
{code}
*CNF Expansion:*
{code}
(name = 'Bob' OR name = 'Alice')
AND (name = 'Bob' OR name = 'Alice' OR age >= 25)
AND (name = 'Bob' OR name = 'Alice' OR department = 'IT')
AND (name = 'Bob' OR department = 'IT' OR age >= 25)
AND (name = 'Alice' OR age >= 21)
AND (name = 'Alice' OR age >= 21 OR department = 'IT')
AND (age >= 21 OR department = 'IT' OR age >= 25)
AND (lastName = 'Smith' OR department = 'IT')
AND (lastName = 'Smith' OR age < 30)
{code}
*Generated REST request body:*
{code:language=json}
{
"page": 0,
"limit": 100,
"where": [
\{ "name": "name", "operator": "=", "value": "Bob" },
\{ "name": "name", "operator": "=", "value": "Alice" }
],
"and": [
[
\{ "name": "name", "operator": "=", "value": "Bob" },
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "age", "operator": ">=", "value": 25 }
],
[
\{ "name": "name", "operator": "=", "value": "Bob" },
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "department", "operator": "=", "value": "IT" }
],
[
\{ "name": "name", "operator": "=", "value": "Bob" },
\{ "name": "department", "operator": "=", "value": "IT" },
\{ "name": "age", "operator": ">=", "value": 25 }
],
[
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "age", "operator": ">=", "value": 21 }
],
[
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "age", "operator": ">=", "value": 21 },
\{ "name": "department", "operator": "=", "value": "IT" }
],
[
\{ "name": "age", "operator": ">=", "value": 21 },
\{ "name": "department", "operator": "=", "value": "IT" },
\{ "name": "age", "operator": ">=", "value": 25 }
],
[
\{ "name": "lastName", "operator": "=", "value": "Smith" },
\{ "name": "department", "operator": "=", "value": "IT" }
],
[
\{ "name": "lastName", "operator": "=", "value": "Smith" },
\{ "name": "age", "operator": "<", "value": 30 }
]
]
}
{code}
*Result:* Returns users with id: 1 (Bob Smith, 25, IT) and 5 (Alice Smith, 28,
IT)
----
h3. Variant 4: CNF with "and" (AND groups with OR inside)
*Description:* REST API accepts:
* *and*: array of arrays, where each inner array contains criteria joined by
OR, and outer array represents AND
*OpenAPI specification:*
{code:language=yaml}
requestBody:
content:
application/json:
schema:
properties:
and:
type: array
items:
type: array
items:
$ref: '#/components/schemas/FilterCriterion'
description: CNF - List of AND groups (each group is a list of OR
criteria)
{code}
*Example scenario:*
REST API contains the following users:
{code:language=json}
[
\{ "id": 1, "name": "Bob", "lastName": "Smith", "age": 25, "department": "IT"
},
\{ "id": 2, "name": "Alice", "lastName": "Johnson", "age": 22, "department":
"HR" },
\{ "id": 3, "name": "Bob", "lastName": "Brown", "age": 19, "department": "IT"
},
\{ "id": 4, "name": "Martin", "lastName": "Smith", "age": 35, "department":
"Sales" },
\{ "id": 5, "name": "Alice", "lastName": "Smith", "age": 28, "department":
"IT" },
\{ "id": 6, "name": "John", "lastName": "Doe", "age": 30, "department": "HR"
},
\{ "id": 7, "name": "Sarah", "lastName": "Smith", "age": 26, "department":
"IT" }
]
{code}
*SQL Query with nested conditions:*
{code:language=sql}
SELECT * FROM users
WHERE (
(name = 'Bob' AND age >= 21)
OR
(name = 'Alice' AND (department = 'IT' OR age >= 25))
)
AND
(lastName = 'Smith' OR (department = 'IT' AND age < 30))
{code}
*CNF Expansion:*
{code}
(name = 'Bob' OR name = 'Alice')
AND (name = 'Bob' OR name = 'Alice' OR age >= 25)
AND (name = 'Bob' OR name = 'Alice' OR department = 'IT')
AND (name = 'Bob' OR department = 'IT' OR age >= 25)
AND (name = 'Alice' OR age >= 21)
AND (name = 'Alice' OR age >= 21 OR department = 'IT')
AND (age >= 21 OR department = 'IT' OR age >= 25)
AND (lastName = 'Smith' OR department = 'IT')
AND (lastName = 'Smith' OR age < 30)
{code}
*Generated REST request body:*
{code:language=json}
{
"page": 0,
"limit": 100,
"and": [
[
\{ "name": "name", "operator": "=", "value": "Bob" },
\{ "name": "name", "operator": "=", "value": "Alice" }
],
[
\{ "name": "name", "operator": "=", "value": "Bob" },
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "age", "operator": ">=", "value": 25 }
],
[
\{ "name": "name", "operator": "=", "value": "Bob" },
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "department", "operator": "=", "value": "IT" }
],
[
\{ "name": "name", "operator": "=", "value": "Bob" },
\{ "name": "department", "operator": "=", "value": "IT" },
\{ "name": "age", "operator": ">=", "value": 25 }
],
[
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "age", "operator": ">=", "value": 21 }
],
[
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "age", "operator": ">=", "value": 21 },
\{ "name": "department", "operator": "=", "value": "IT" }
],
[
\{ "name": "age", "operator": ">=", "value": 21 },
\{ "name": "department", "operator": "=", "value": "IT" },
\{ "name": "age", "operator": ">=", "value": 25 }
],
[
\{ "name": "lastName", "operator": "=", "value": "Smith" },
\{ "name": "department", "operator": "=", "value": "IT" }
],
[
\{ "name": "lastName", "operator": "=", "value": "Smith" },
\{ "name": "age", "operator": "<", "value": 30 }
]
]
}
{code}
*Result:* Returns users with id: 1 (Bob Smith, 25, IT) and 5 (Alice Smith, 28,
IT)
----
h3. Variant 5: Ambiguous "filters" field (requires description hint)
*Description:* REST API uses generic field name like *filters* which doesn't
clearly indicate DNF or CNF structure. The OpenAPI description MUST explicitly
state "DNF" or "CNF".
h4. DNF Example
*OpenAPI specification:*
{code:language=yaml}
requestBody:
content:
application/json:
schema:
properties:
filters:
type: array
items:
type: array
items:
$ref: '#/components/schemas/FilterCriterion'
description: "DNF: Array of OR groups, each group contains ANDed
criteria"
{code}
*Example scenario:*
REST API contains the following users:
{code:language=json}
[
\{ "id": 1, "name": "Bob", "lastName": "Smith", "age": 25, "department": "IT"
},
\{ "id": 2, "name": "Alice", "lastName": "Johnson", "age": 22, "department":
"HR" },
\{ "id": 3, "name": "Bob", "lastName": "Brown", "age": 19, "department": "IT"
},
\{ "id": 4, "name": "Martin", "lastName": "Smith", "age": 35, "department":
"Sales" },
\{ "id": 5, "name": "Alice", "lastName": "Smith", "age": 28, "department":
"IT" },
\{ "id": 6, "name": "John", "lastName": "Doe", "age": 30, "department": "HR"
},
\{ "id": 7, "name": "Sarah", "lastName": "Smith", "age": 26, "department":
"IT" }
]
{code}
*SQL Query with nested conditions:*
{code:language=sql}
SELECT * FROM users
WHERE (
(name = 'Bob' AND age >= 21)
OR
(name = 'Alice' AND (department = 'IT' OR age >= 25))
)
AND
(lastName = 'Smith' OR (department = 'IT' AND age < 30))
{code}
*DNF Expansion:*
{code}
(name = 'Bob' AND age >= 21 AND lastName = 'Smith')
OR (name = 'Bob' AND age >= 21 AND department = 'IT' AND age < 30)
OR (name = 'Alice' AND department = 'IT' AND lastName = 'Smith')
OR (name = 'Alice' AND department = 'IT' AND age < 30)
OR (name = 'Alice' AND age >= 25 AND lastName = 'Smith')
OR (name = 'Alice' AND age >= 25 AND department = 'IT' AND age < 30)
{code}
*Generated REST request body:*
{code:language=json}
{
"page": 0,
"limit": 100,
"filters": [
[
\{ "name": "name", "operator": "=", "value": "Bob" },
\{ "name": "age", "operator": ">=", "value": 21 },
\{ "name": "lastName", "operator": "=", "value": "Smith" }
],
[
\{ "name": "name", "operator": "=", "value": "Bob" },
\{ "name": "age", "operator": ">=", "value": 21 },
\{ "name": "department", "operator": "=", "value": "IT" },
\{ "name": "age", "operator": "<", "value": 30 }
],
[
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "department", "operator": "=", "value": "IT" },
\{ "name": "lastName", "operator": "=", "value": "Smith" }
],
[
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "department", "operator": "=", "value": "IT" },
\{ "name": "age", "operator": "<", "value": 30 }
],
[
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "age", "operator": ">=", "value": 25 },
\{ "name": "lastName", "operator": "=", "value": "Smith" }
],
[
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "age", "operator": ">=", "value": 25 },
\{ "name": "department", "operator": "=", "value": "IT" },
\{ "name": "age", "operator": "<", "value": 30 }
]
]
}
{code}
*Result:* Returns users with id: 1 (Bob Smith, 25, IT) and 5 (Alice Smith, 28,
IT)
h4. CNF Example
*OpenAPI specification:*
{code:language=yaml}
requestBody:
content:
application/json:
schema:
properties:
filters:
type: array
items:
type: array
items:
$ref: '#/components/schemas/FilterCriterion'
description: "CNF: Array of AND groups, each group contains ORed
criteria"
{code}
*Example scenario:*
REST API contains the following users:
{code:language=json}
[
\{ "id": 1, "name": "Bob", "lastName": "Smith", "age": 25, "department": "IT"
},
\{ "id": 2, "name": "Alice", "lastName": "Johnson", "age": 22, "department":
"HR" },
\{ "id": 3, "name": "Bob", "lastName": "Brown", "age": 19, "department": "IT"
},
\{ "id": 4, "name": "Martin", "lastName": "Smith", "age": 35, "department":
"Sales" },
\{ "id": 5, "name": "Alice", "lastName": "Smith", "age": 28, "department":
"IT" },
\{ "id": 6, "name": "John", "lastName": "Doe", "age": 30, "department": "HR"
},
\{ "id": 7, "name": "Sarah", "lastName": "Smith", "age": 26, "department":
"IT" }
]
{code}
*SQL Query with nested conditions:*
{code:language=sql}
SELECT * FROM users
WHERE (
(name = 'Bob' AND age >= 21)
OR
(name = 'Alice' AND (department = 'IT' OR age >= 25))
)
AND
(lastName = 'Smith' OR (department = 'IT' AND age < 30))
{code}
*CNF Expansion:*
{code}
(name = 'Bob' OR name = 'Alice')
AND (name = 'Bob' OR name = 'Alice' OR age >= 25)
AND (name = 'Bob' OR name = 'Alice' OR department = 'IT')
AND (name = 'Bob' OR department = 'IT' OR age >= 25)
AND (name = 'Alice' OR age >= 21)
AND (name = 'Alice' OR age >= 21 OR department = 'IT')
AND (age >= 21 OR department = 'IT' OR age >= 25)
AND (lastName = 'Smith' OR department = 'IT')
AND (lastName = 'Smith' OR age < 30)
{code}
*Generated REST request body:*
{code:language=json}
{
"page": 0,
"limit": 100,
"filters": [
[
\{ "name": "name", "operator": "=", "value": "Bob" },
\{ "name": "name", "operator": "=", "value": "Alice" }
],
[
\{ "name": "name", "operator": "=", "value": "Bob" },
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "age", "operator": ">=", "value": 25 }
],
[
\{ "name": "name", "operator": "=", "value": "Bob" },
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "department", "operator": "=", "value": "IT" }
],
[
\{ "name": "name", "operator": "=", "value": "Bob" },
\{ "name": "department", "operator": "=", "value": "IT" },
\{ "name": "age", "operator": ">=", "value": 25 }
],
[
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "age", "operator": ">=", "value": 21 }
],
[
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "age", "operator": ">=", "value": 21 },
\{ "name": "department", "operator": "=", "value": "IT" }
],
[
\{ "name": "age", "operator": ">=", "value": 21 },
\{ "name": "department", "operator": "=", "value": "IT" },
\{ "name": "age", "operator": ">=", "value": 25 }
],
[
\{ "name": "lastName", "operator": "=", "value": "Smith" },
\{ "name": "department", "operator": "=", "value": "IT" }
],
[
\{ "name": "lastName", "operator": "=", "value": "Smith" },
\{ "name": "age", "operator": "<", "value": 30 }
]
]
}
{code}
*Result:* Returns users with id: 1 (Bob Smith, 25, IT) and 5 (Alice Smith, 28,
IT)
----
h3. Filter Capability Detection
The adapter analyzes the OpenAPI *requestBody* schema to detect filter
capabilities:
# 1. Field names
Presence of *or*, *and*, *where*, or *filters* fields.
# 2. Array structure
Single array vs array of arrays.
# 3. Descriptions
Keywords *"DNF"* or *"CNF"* in field descriptions.
# 4. Supported operators
Extracted from *FilterCriterion* schema enum.
If the OpenAPI specification doesn't clearly define the filter structure, or if
you need custom request formatting, you can provide a *FreeMarker template*
(.ftl file) to explicitly define the request body format.
was (Author: JIRAUSER305374):
For now I am still evaluating this functionality and possible design options,
so it is not implemented yet.
h2. REST API Filter Specification Variants
Different REST APIs use different structures for expressing filters. The
adapter automatically detects the supported format from the OpenAPI
specification. Below are the supported variants:
h3. Variant 1: DNF with "where" (AND) + "or" (OR groups)
*Description:* REST API accepts:
* {{{}where{}}}: array of filter criteria joined by AND (common conditions
from first DNF group)
* {{{}or{}}}: array of arrays, where each inner array contains criteria joined
by AND, and outer array represents OR
*OpenAPI specification:*
!http://localhost:63342/markdownPreview/1513829958/?_ijt=md615dfi7p40c910hbkffsrpg0!
{{requestBody:
content:
application/json:
schema:
properties:
where:
type: array
items:
$ref: '#/components/schemas/FilterCriterion'
description: List of AND filter criteria
or:
type: array
items:
type: array
items:
$ref: '#/components/schemas/FilterCriterion'
description: DNF - List of OR groups}}
*Example scenario:*
REST API contains the following users:
!http://localhost:63342/markdownPreview/1513829958/?_ijt=md615dfi7p40c910hbkffsrpg0!
{{[
\{ "id": 1, "name": "Bob", "lastName": "Smith", "age": 25, "department": "IT"
},
\{ "id": 2, "name": "Alice", "lastName": "Johnson", "age": 22, "department":
"HR" },
\{ "id": 3, "name": "Bob", "lastName": "Brown", "age": 19, "department": "IT"
},
\{ "id": 4, "name": "Martin", "lastName": "Smith", "age": 35, "department":
"Sales" },
\{ "id": 5, "name": "Alice", "lastName": "Smith", "age": 28, "department":
"IT" },
\{ "id": 6, "name": "John", "lastName": "Doe", "age": 30, "department": "HR"
},
\{ "id": 7, "name": "Sarah", "lastName": "Smith", "age": 26, "department":
"IT" }
]}}
*SQL Query with nested conditions:*
!http://localhost:63342/markdownPreview/1513829958/?_ijt=md615dfi7p40c910hbkffsrpg0!
{{SELECT * FROM users
WHERE (
(name = 'Bob' AND age >= 21)
OR
(name = 'Alice' AND (department = 'IT' OR age >= 25))
)
AND
(lastName = 'Smith' OR (department = 'IT' AND age < 30))}}
*DNF Expansion:*
!http://localhost:63342/markdownPreview/1513829958/?_ijt=md615dfi7p40c910hbkffsrpg0!
{{(name = 'Bob' AND age >= 21 AND lastName = 'Smith')
OR (name = 'Bob' AND age >= 21 AND department = 'IT' AND age < 30)
OR (name = 'Alice' AND department = 'IT' AND lastName = 'Smith')
OR (name = 'Alice' AND department = 'IT' AND department = 'IT' AND age < 30)
OR (name = 'Alice' AND age >= 25 AND lastName = 'Smith')
OR (name = 'Alice' AND age >= 25 AND department = 'IT' AND age < 30)}}
*Generated REST request body:*
!http://localhost:63342/markdownPreview/1513829958/?_ijt=md615dfi7p40c910hbkffsrpg0!
{{{
"page": 0,
"limit": 100,
"where": [
\{ "name": "name", "operator": "=", "value": "Bob" },
\{ "name": "age", "operator": ">=", "value": 21 },
\{ "name": "lastName", "operator": "=", "value": "Smith" }
],
"or": [
[
\{ "name": "name", "operator": "=", "value": "Bob" },
\{ "name": "age", "operator": ">=", "value": 21 },
\{ "name": "department", "operator": "=", "value": "IT" },
\{ "name": "age", "operator": "<", "value": 30 }
],
[
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "department", "operator": "=", "value": "IT" },
\{ "name": "lastName", "operator": "=", "value": "Smith" }
],
[
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "department", "operator": "=", "value": "IT" },
\{ "name": "age", "operator": "<", "value": 30 }
],
[
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "age", "operator": ">=", "value": 25 },
\{ "name": "lastName", "operator": "=", "value": "Smith" }
],
[
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "age", "operator": ">=", "value": 25 },
\{ "name": "department", "operator": "=", "value": "IT" },
\{ "name": "age", "operator": "<", "value": 30 }
]
]
}}}
*Result:* Returns users with id: 1 (Bob Smith, 25, IT) and 5 (Alice Smith, 28,
IT)
----
h3. Variant 2: DNF with "or" only (no "where")
*Description:* REST API accepts:
* {{{}or{}}}: array of arrays (DNF structure), no separate "where" field
*OpenAPI specification:*
!http://localhost:63342/markdownPreview/1513829958/?_ijt=md615dfi7p40c910hbkffsrpg0!
{{requestBody:
content:
application/json:
schema:
properties:
or:
type: array
items:
type: array
items:
$ref: '#/components/schemas/FilterCriterion'
description: DNF - List of OR groups (each group is ANDed)}}
*Example scenario:*
REST API contains the following users:
!http://localhost:63342/markdownPreview/1513829958/?_ijt=md615dfi7p40c910hbkffsrpg0!
{{[
\{ "id": 1, "name": "Bob", "lastName": "Smith", "age": 25, "department": "IT"
},
\{ "id": 2, "name": "Alice", "lastName": "Johnson", "age": 22, "department":
"HR" },
\{ "id": 3, "name": "Bob", "lastName": "Brown", "age": 19, "department": "IT"
},
\{ "id": 4, "name": "Martin", "lastName": "Smith", "age": 35, "department":
"Sales" },
\{ "id": 5, "name": "Alice", "lastName": "Smith", "age": 28, "department":
"IT" },
\{ "id": 6, "name": "John", "lastName": "Doe", "age": 30, "department": "HR"
},
\{ "id": 7, "name": "Sarah", "lastName": "Smith", "age": 26, "department":
"IT" }
]}}
*SQL Query with nested conditions:*
!http://localhost:63342/markdownPreview/1513829958/?_ijt=md615dfi7p40c910hbkffsrpg0!
{{SELECT * FROM users
WHERE (
(name = 'Bob' AND age >= 21)
OR
(name = 'Alice' AND (department = 'IT' OR age >= 25))
)
AND
(lastName = 'Smith' OR (department = 'IT' AND age < 30))}}
*DNF Expansion:*
!http://localhost:63342/markdownPreview/1513829958/?_ijt=md615dfi7p40c910hbkffsrpg0!
{{(name = 'Bob' AND age >= 21 AND lastName = 'Smith')
OR (name = 'Bob' AND age >= 21 AND department = 'IT' AND age < 30)
OR (name = 'Alice' AND department = 'IT' AND lastName = 'Smith')
OR (name = 'Alice' AND department = 'IT' AND department = 'IT' AND age < 30)
OR (name = 'Alice' AND age >= 25 AND lastName = 'Smith')
OR (name = 'Alice' AND age >= 25 AND department = 'IT' AND age < 30)}}
*Generated REST request body:*
!http://localhost:63342/markdownPreview/1513829958/?_ijt=md615dfi7p40c910hbkffsrpg0!
{{{
"page": 0,
"limit": 100,
"or": [
[
\{ "name": "name", "operator": "=", "value": "Bob" },
\{ "name": "age", "operator": ">=", "value": 21 },
\{ "name": "lastName", "operator": "=", "value": "Smith" }
],
[
\{ "name": "name", "operator": "=", "value": "Bob" },
\{ "name": "age", "operator": ">=", "value": 21 },
\{ "name": "department", "operator": "=", "value": "IT" },
\{ "name": "age", "operator": "<", "value": 30 }
],
[
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "department", "operator": "=", "value": "IT" },
\{ "name": "lastName", "operator": "=", "value": "Smith" }
],
[
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "department", "operator": "=", "value": "IT" },
\{ "name": "age", "operator": "<", "value": 30 }
],
[
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "age", "operator": ">=", "value": 25 },
\{ "name": "lastName", "operator": "=", "value": "Smith" }
],
[
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "age", "operator": ">=", "value": 25 },
\{ "name": "department", "operator": "=", "value": "IT" },
\{ "name": "age", "operator": "<", "value": 30 }
]
]
}}}
*Result:* Returns users with id: 1 (Bob Smith, 25, IT) and 5 (Alice Smith, 28,
IT)
----
h3. Variant 3: CNF with "where" (OR) + "and" (AND groups)
*Description:* REST API accepts:
* {{{}where{}}}: array of filter criteria joined by OR (conditions from first
CNF group)
* {{{}and{}}}: array of arrays, where each inner array contains criteria
joined by OR, and outer array represents AND
*OpenAPI specification:*
!http://localhost:63342/markdownPreview/1513829958/?_ijt=md615dfi7p40c910hbkffsrpg0!
{{requestBody:
content:
application/json:
schema:
properties:
where:
type: array
items:
$ref: '#/components/schemas/FilterCriterion'
description: List of OR filter criteria
and:
type: array
items:
type: array
items:
$ref: '#/components/schemas/FilterCriterion'
description: CNF - List of AND groups (each group contains OR
criteria)}}
*Example scenario:*
REST API contains the following users:
!http://localhost:63342/markdownPreview/1513829958/?_ijt=md615dfi7p40c910hbkffsrpg0!
{{[
\{ "id": 1, "name": "Bob", "lastName": "Smith", "age": 25, "department": "IT"
},
\{ "id": 2, "name": "Alice", "lastName": "Johnson", "age": 22, "department":
"HR" },
\{ "id": 3, "name": "Bob", "lastName": "Brown", "age": 19, "department": "IT"
},
\{ "id": 4, "name": "Martin", "lastName": "Smith", "age": 35, "department":
"Sales" },
\{ "id": 5, "name": "Alice", "lastName": "Smith", "age": 28, "department":
"IT" },
\{ "id": 6, "name": "John", "lastName": "Doe", "age": 30, "department": "HR"
},
\{ "id": 7, "name": "Sarah", "lastName": "Smith", "age": 26, "department":
"IT" }
]}}
*SQL Query with nested conditions:*
!http://localhost:63342/markdownPreview/1513829958/?_ijt=md615dfi7p40c910hbkffsrpg0!
{{SELECT * FROM users
WHERE (
(name = 'Bob' AND age >= 21)
OR
(name = 'Alice' AND (department = 'IT' OR age >= 25))
)
AND
(lastName = 'Smith' OR (department = 'IT' AND age < 30))}}
*CNF Expansion:*
!http://localhost:63342/markdownPreview/1513829958/?_ijt=md615dfi7p40c910hbkffsrpg0!
{{(name = 'Bob' OR name = 'Alice')
AND (name = 'Bob' OR name = 'Alice' OR age >= 25)
AND (name = 'Bob' OR name = 'Alice' OR department = 'IT')
AND (name = 'Bob' OR department = 'IT' OR age >= 25)
AND (name = 'Alice' OR age >= 21)
AND (name = 'Alice' OR age >= 21 OR department = 'IT')
AND (age >= 21 OR department = 'IT' OR age >= 25)
AND (lastName = 'Smith' OR department = 'IT')
AND (lastName = 'Smith' OR age < 30)}}
*Generated REST request body:*
!http://localhost:63342/markdownPreview/1513829958/?_ijt=md615dfi7p40c910hbkffsrpg0!
{{{
"page": 0,
"limit": 100,
"where": [
\{ "name": "name", "operator": "=", "value": "Bob" },
\{ "name": "name", "operator": "=", "value": "Alice" }
],
"and": [
[
\{ "name": "name", "operator": "=", "value": "Bob" },
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "age", "operator": ">=", "value": 25 }
],
[
\{ "name": "name", "operator": "=", "value": "Bob" },
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "department", "operator": "=", "value": "IT" }
],
[
\{ "name": "name", "operator": "=", "value": "Bob" },
\{ "name": "department", "operator": "=", "value": "IT" },
\{ "name": "age", "operator": ">=", "value": 25 }
],
[
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "age", "operator": ">=", "value": 21 }
],
[
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "age", "operator": ">=", "value": 21 },
\{ "name": "department", "operator": "=", "value": "IT" }
],
[
\{ "name": "age", "operator": ">=", "value": 21 },
\{ "name": "department", "operator": "=", "value": "IT" },
\{ "name": "age", "operator": ">=", "value": 25 }
],
[
\{ "name": "lastName", "operator": "=", "value": "Smith" },
\{ "name": "department", "operator": "=", "value": "IT" }
],
[
\{ "name": "lastName", "operator": "=", "value": "Smith" },
\{ "name": "age", "operator": "<", "value": 30 }
]
]
}}}
*Result:* Returns users with id: 1 (Bob Smith, 25, IT) and 5 (Alice Smith, 28,
IT)
----
h3. Variant 4: CNF with "and" (AND groups with OR inside)
*Description:* REST API accepts:
* {{{}and{}}}: array of arrays, where each inner array contains criteria
joined by OR, and outer array represents AND
*OpenAPI specification:*
!http://localhost:63342/markdownPreview/1513829958/?_ijt=md615dfi7p40c910hbkffsrpg0!
{{requestBody:
content:
application/json:
schema:
properties:
and:
type: array
items:
type: array
items:
$ref: '#/components/schemas/FilterCriterion'
description: CNF - List of AND groups (each group is a list of OR
criteria)}}
*Example scenario:*
REST API contains the following users:
!http://localhost:63342/markdownPreview/1513829958/?_ijt=md615dfi7p40c910hbkffsrpg0!
{{[
\{ "id": 1, "name": "Bob", "lastName": "Smith", "age": 25, "department": "IT"
},
\{ "id": 2, "name": "Alice", "lastName": "Johnson", "age": 22, "department":
"HR" },
\{ "id": 3, "name": "Bob", "lastName": "Brown", "age": 19, "department": "IT"
},
\{ "id": 4, "name": "Martin", "lastName": "Smith", "age": 35, "department":
"Sales" },
\{ "id": 5, "name": "Alice", "lastName": "Smith", "age": 28, "department":
"IT" },
\{ "id": 6, "name": "John", "lastName": "Doe", "age": 30, "department": "HR"
},
\{ "id": 7, "name": "Sarah", "lastName": "Smith", "age": 26, "department":
"IT" }
]}}
*SQL Query with nested conditions:*
!http://localhost:63342/markdownPreview/1513829958/?_ijt=md615dfi7p40c910hbkffsrpg0!
{{SELECT * FROM users
WHERE (
(name = 'Bob' AND age >= 21)
OR
(name = 'Alice' AND (department = 'IT' OR age >= 25))
)
AND
(lastName = 'Smith' OR (department = 'IT' AND age < 30))}}
*CNF Expansion:*
!http://localhost:63342/markdownPreview/1513829958/?_ijt=md615dfi7p40c910hbkffsrpg0!
{{(name = 'Bob' OR name = 'Alice')
AND (name = 'Bob' OR name = 'Alice' OR age >= 25)
AND (name = 'Bob' OR name = 'Alice' OR department = 'IT')
AND (name = 'Bob' OR department = 'IT' OR age >= 25)
AND (name = 'Alice' OR age >= 21)
AND (name = 'Alice' OR age >= 21 OR department = 'IT')
AND (age >= 21 OR department = 'IT' OR age >= 25)
AND (lastName = 'Smith' OR department = 'IT')
AND (lastName = 'Smith' OR age < 30)}}
*Generated REST request body:*
!http://localhost:63342/markdownPreview/1513829958/?_ijt=md615dfi7p40c910hbkffsrpg0!
{{{
"page": 0,
"limit": 100,
"and": [
[
\{ "name": "name", "operator": "=", "value": "Bob" },
\{ "name": "name", "operator": "=", "value": "Alice" }
],
[
\{ "name": "name", "operator": "=", "value": "Bob" },
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "age", "operator": ">=", "value": 25 }
],
[
\{ "name": "name", "operator": "=", "value": "Bob" },
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "department", "operator": "=", "value": "IT" }
],
[
\{ "name": "name", "operator": "=", "value": "Bob" },
\{ "name": "department", "operator": "=", "value": "IT" },
\{ "name": "age", "operator": ">=", "value": 25 }
],
[
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "age", "operator": ">=", "value": 21 }
],
[
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "age", "operator": ">=", "value": 21 },
\{ "name": "department", "operator": "=", "value": "IT" }
],
[
\{ "name": "age", "operator": ">=", "value": 21 },
\{ "name": "department", "operator": "=", "value": "IT" },
\{ "name": "age", "operator": ">=", "value": 25 }
],
[
\{ "name": "lastName", "operator": "=", "value": "Smith" },
\{ "name": "department", "operator": "=", "value": "IT" }
],
[
\{ "name": "lastName", "operator": "=", "value": "Smith" },
\{ "name": "age", "operator": "<", "value": 30 }
]
]
}}}
*Result:* Returns users with id: 1 (Bob Smith, 25, IT) and 5 (Alice Smith, 28,
IT)
----
h3. Variant 5: Ambiguous "filters" field (requires description hint)
*Description:* REST API uses generic field name like {{filters}} which doesn't
clearly indicate DNF or CNF structure. The OpenAPI description MUST explicitly
state "DNF" or "CNF".
h4. DNF Example
*OpenAPI specification:*
!http://localhost:63342/markdownPreview/1513829958/?_ijt=md615dfi7p40c910hbkffsrpg0!
{{requestBody:
content:
application/json:
schema:
properties:
filters:
type: array
items:
type: array
items:
$ref: '#/components/schemas/FilterCriterion'
description: "DNF: Array of OR groups, each group contains ANDed
criteria"}}
*Example scenario:*
REST API contains the following users:
!http://localhost:63342/markdownPreview/1513829958/?_ijt=md615dfi7p40c910hbkffsrpg0!
{{[
\{ "id": 1, "name": "Bob", "lastName": "Smith", "age": 25, "department": "IT"
},
\{ "id": 2, "name": "Alice", "lastName": "Johnson", "age": 22, "department":
"HR" },
\{ "id": 3, "name": "Bob", "lastName": "Brown", "age": 19, "department": "IT"
},
\{ "id": 4, "name": "Martin", "lastName": "Smith", "age": 35, "department":
"Sales" },
\{ "id": 5, "name": "Alice", "lastName": "Smith", "age": 28, "department":
"IT" },
\{ "id": 6, "name": "John", "lastName": "Doe", "age": 30, "department": "HR"
},
\{ "id": 7, "name": "Sarah", "lastName": "Smith", "age": 26, "department":
"IT" }
]}}
*SQL Query with nested conditions:*
!http://localhost:63342/markdownPreview/1513829958/?_ijt=md615dfi7p40c910hbkffsrpg0!
{{SELECT * FROM users
WHERE (
(name = 'Bob' AND age >= 21)
OR
(name = 'Alice' AND (department = 'IT' OR age >= 25))
)
AND
(lastName = 'Smith' OR (department = 'IT' AND age < 30))}}
*DNF Expansion:*
!http://localhost:63342/markdownPreview/1513829958/?_ijt=md615dfi7p40c910hbkffsrpg0!
{{(name = 'Bob' AND age >= 21 AND lastName = 'Smith')
OR (name = 'Bob' AND age >= 21 AND department = 'IT' AND age < 30)
OR (name = 'Alice' AND department = 'IT' AND lastName = 'Smith')
OR (name = 'Alice' AND department = 'IT' AND age < 30)
OR (name = 'Alice' AND age >= 25 AND lastName = 'Smith')
OR (name = 'Alice' AND age >= 25 AND department = 'IT' AND age < 30)}}
*Generated REST request body:*
!http://localhost:63342/markdownPreview/1513829958/?_ijt=md615dfi7p40c910hbkffsrpg0!
{{{
"page": 0,
"limit": 100,
"filters": [
[
\{ "name": "name", "operator": "=", "value": "Bob" },
\{ "name": "age", "operator": ">=", "value": 21 },
\{ "name": "lastName", "operator": "=", "value": "Smith" }
],
[
\{ "name": "name", "operator": "=", "value": "Bob" },
\{ "name": "age", "operator": ">=", "value": 21 },
\{ "name": "department", "operator": "=", "value": "IT" },
\{ "name": "age", "operator": "<", "value": 30 }
],
[
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "department", "operator": "=", "value": "IT" },
\{ "name": "lastName", "operator": "=", "value": "Smith" }
],
[
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "department", "operator": "=", "value": "IT" },
\{ "name": "age", "operator": "<", "value": 30 }
],
[
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "age", "operator": ">=", "value": 25 },
\{ "name": "lastName", "operator": "=", "value": "Smith" }
],
[
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "age", "operator": ">=", "value": 25 },
\{ "name": "department", "operator": "=", "value": "IT" },
\{ "name": "age", "operator": "<", "value": 30 }
]
]
}}}
*Result:* Returns users with id: 1 (Bob Smith, 25, IT) and 5 (Alice Smith, 28,
IT)
h4. CNF Example
*OpenAPI specification:*
!http://localhost:63342/markdownPreview/1513829958/?_ijt=md615dfi7p40c910hbkffsrpg0!
{{requestBody:
content:
application/json:
schema:
properties:
filters:
type: array
items:
type: array
items:
$ref: '#/components/schemas/FilterCriterion'
description: "CNF: Array of AND groups, each group contains ORed
criteria"}}
*Example scenario:*
REST API contains the following users:
!http://localhost:63342/markdownPreview/1513829958/?_ijt=md615dfi7p40c910hbkffsrpg0!
{{[
\{ "id": 1, "name": "Bob", "lastName": "Smith", "age": 25, "department": "IT"
},
\{ "id": 2, "name": "Alice", "lastName": "Johnson", "age": 22, "department":
"HR" },
\{ "id": 3, "name": "Bob", "lastName": "Brown", "age": 19, "department": "IT"
},
\{ "id": 4, "name": "Martin", "lastName": "Smith", "age": 35, "department":
"Sales" },
\{ "id": 5, "name": "Alice", "lastName": "Smith", "age": 28, "department":
"IT" },
\{ "id": 6, "name": "John", "lastName": "Doe", "age": 30, "department": "HR"
},
\{ "id": 7, "name": "Sarah", "lastName": "Smith", "age": 26, "department":
"IT" }
]}}
*SQL Query with nested conditions:*
!http://localhost:63342/markdownPreview/1513829958/?_ijt=md615dfi7p40c910hbkffsrpg0!
{{SELECT * FROM users
WHERE (
(name = 'Bob' AND age >= 21)
OR
(name = 'Alice' AND (department = 'IT' OR age >= 25))
)
AND
(lastName = 'Smith' OR (department = 'IT' AND age < 30))}}
*CNF Expansion:*
!http://localhost:63342/markdownPreview/1513829958/?_ijt=md615dfi7p40c910hbkffsrpg0!
{{(name = 'Bob' OR name = 'Alice')
AND (name = 'Bob' OR name = 'Alice' OR age >= 25)
AND (name = 'Bob' OR name = 'Alice' OR department = 'IT')
AND (name = 'Bob' OR department = 'IT' OR age >= 25)
AND (name = 'Alice' OR age >= 21)
AND (name = 'Alice' OR age >= 21 OR department = 'IT')
AND (age >= 21 OR department = 'IT' OR age >= 25)
AND (lastName = 'Smith' OR department = 'IT')
AND (lastName = 'Smith' OR age < 30)}}
*Generated REST request body:*
!http://localhost:63342/markdownPreview/1513829958/?_ijt=md615dfi7p40c910hbkffsrpg0!
{{{
"page": 0,
"limit": 100,
"filters": [
[
\{ "name": "name", "operator": "=", "value": "Bob" },
\{ "name": "name", "operator": "=", "value": "Alice" }
],
[
\{ "name": "name", "operator": "=", "value": "Bob" },
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "age", "operator": ">=", "value": 25 }
],
[
\{ "name": "name", "operator": "=", "value": "Bob" },
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "department", "operator": "=", "value": "IT" }
],
[
\{ "name": "name", "operator": "=", "value": "Bob" },
\{ "name": "department", "operator": "=", "value": "IT" },
\{ "name": "age", "operator": ">=", "value": 25 }
],
[
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "age", "operator": ">=", "value": 21 }
],
[
\{ "name": "name", "operator": "=", "value": "Alice" },
\{ "name": "age", "operator": ">=", "value": 21 },
\{ "name": "department", "operator": "=", "value": "IT" }
],
[
\{ "name": "age", "operator": ">=", "value": 21 },
\{ "name": "department", "operator": "=", "value": "IT" },
\{ "name": "age", "operator": ">=", "value": 25 }
],
[
\{ "name": "lastName", "operator": "=", "value": "Smith" },
\{ "name": "department", "operator": "=", "value": "IT" }
],
[
\{ "name": "lastName", "operator": "=", "value": "Smith" },
\{ "name": "age", "operator": "<", "value": 30 }
]
]
}}}
*Result:* Returns users with id: 1 (Bob Smith, 25, IT) and 5 (Alice Smith, 28,
IT)
*Note:* If description doesn't contain "DNF" or "CNF", the adapter will throw
an error asking user to provide a FreeMarker template.
----
h3. Filter Capability Detection
The adapter analyzes the OpenAPI {{requestBody}} schema to detect filter
capabilities:
# *Field names:* Presence of {{{}or{}}}, {{{}and{}}}, {{{}where{}}}, or
{{filters}} fields
# *Array structure:* Single array vs array of arrays
# *Descriptions:* Keywords "DNF" or "CNF" in field descriptions
# *Supported operators:* Extracted from {{FilterCriterion}} schema enum
If the OpenAPI specification doesn't clearly define the filter structure, or if
you need custom request formatting, you can provide a *FreeMarker template*
(.ftl file) to explicitly define the request body format.
> Proposal to contribute CalciteRestAPIAdapter
> --------------------------------------------
>
> Key: CALCITE-7300
> URL: https://issues.apache.org/jira/browse/CALCITE-7300
> Project: Calcite
> Issue Type: New Feature
> Reporter: Oleg Alekseev
> Priority: Minor
>
> First version
> The adapter is publicly available on GitHub:
> [https://github.com/oalekseev/CalciteRestAPIAdapter]
> What should be added:
> * Dealing with the format of the data that comes back (csv, xml, json)
> * Javadoc, tests, and user doc
> * Consider support for OpenAPI specifications in the future, as it could
> help unify and automate integration with a wide range of REST services
> Proposal to contribute here
> https://lists.apache.org/thread/jvbpz7rp7w76gqmshtz3y6bhcftk41c5
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
