{ "$defs": { "AnyNonCollectionType": { "anyOf": [ { "type": "string" }, { "type": "number" }, { "type": "integer" }, { "type": "boolean" }, { "type": "null" } ] }, "AnyType": { "anyOf": [ { "type": "string" }, { "type": "number" }, { "type": "integer" }, { "type": "boolean" }, { "type": "null" }, { "type": "array" }, { "type": "object" } ] }, "AuthoritativeDefinitions": { "description": "List of links to sources that provide more details on the dataset; examples would be a link to an external definition, a training video, a GitHub repo, Collibra, or another tool. Authoritative definitions follow the same structure in the standard.", "items": { "properties": { "type": { "description": "Type of definition for authority: v2.3 adds standard values: `businessDefinition`, `transformationImplementation`, `videoTutorial`, `tutorial`, and `implementation`.", "examples": [ "businessDefinition", "transformationImplementation", "videoTutorial", "tutorial", "implementation" ], "type": "string" }, "url": { "description": "URL to the authority.", "type": "string" } }, "required": [ "url", "type" ], "type": "object" }, "type": "array" }, "CustomProperties": { "description": "A list of key/value pairs for custom properties.", "items": { "$ref": "#/$defs/CustomProperty" }, "type": "array" }, "CustomProperty": { "properties": { "property": { "description": "The name of the key. Names should be in camel case–the same as if they were permanent properties in the contract.", "type": "string" }, "value": { "$ref": "#/$defs/AnyType", "description": "The value of the key." } }, "type": "object" }, "DataQuality": { "allOf": [ { "if": { "properties": { "type": { "const": "library" } } }, "then": { "$ref": "#/$defs/DataQualityLibrary" } }, { "if": { "properties": { "type": { "const": "sql" } }, "required": [ "type" ] }, "then": { "$ref": "#/$defs/DataQualitySql" } }, { "if": { "properties": { "type": { "const": "custom" } }, "required": [ "type" ] }, "then": { "$ref": "#/$defs/DataQualityCustom" } } ], "properties": { "authoritativeDefinitions": { "$ref": "#/$defs/AuthoritativeDefinitions" }, "businessImpact": { "description": "Consequences of the rule failure.", "examples": [ "operational", "regulatory" ], "type": "string" }, "customProperties": { "description": "Additional properties required for rule execution.", "items": { "$ref": "#/$defs/CustomProperty" }, "type": "array" }, "description": { "description": "Describe the quality check to be completed.", "type": "string" }, "dimension": { "description": "The key performance indicator (KPI) or dimension for data quality.", "enum": [ "accuracy", "completeness", "conformity", "consistency", "coverage", "timeliness", "uniqueness" ], "type": "string" }, "method": { "examples": [ "reconciliation" ], "type": "string" }, "name": { "description": "Name of the data quality check.", "type": "string" }, "schedule": { "description": "Rule execution schedule details.", "examples": [ "0 20 * * *" ], "type": "string" }, "scheduler": { "description": "The name or type of scheduler used to start the data quality check.", "examples": [ "cron" ], "type": "string" }, "severity": { "description": "The severance of the quality rule.", "examples": [ "info", "warning", "error" ], "type": "string" }, "tags": { "$ref": "#/$defs/Tags" }, "type": { "default": "library", "description": "The type of quality check. 'text' is human-readable text that describes the quality of the data. 'library' is a set of maintained predefined quality attributes such as row count or unique. 'sql' is an individual SQL query that returns a value that can be compared. 'custom' is quality attributes that are vendor-specific, such as Soda or Great Expectations.", "enum": [ "text", "library", "sql", "custom" ], "type": "string" }, "unit": { "description": "Unit the rule is using, popular values are `rows` or `percent`, but any value is allowed.", "examples": [ "rows", "percent" ], "type": "string" } }, "type": "object" }, "DataQualityChecks": { "description": "Data quality rules with all the relevant information for rule setup and execution.", "items": { "$ref": "#/$defs/DataQuality" }, "type": "array" }, "DataQualityCustom": { "properties": { "engine": { "description": "Name of the engine which executes the data quality checks.", "examples": [ "soda", "great-expectations", "monte-carlo", "dbt" ], "type": "string" }, "implementation": { "oneOf": [ { "type": "string" }, { "type": "object" } ] } }, "required": [ "engine", "implementation" ], "type": "object" }, "DataQualityLibrary": { "properties": { "mustBe": { "description": "Must be equal to the value to be valid. When using numbers, it is equivalent to '='." }, "mustBeBetween": { "description": "Must be between the two numbers to be valid. Smallest number first in the array.", "items": { "type": "number" }, "maxItems": 2, "minItems": 2, "type": "array", "uniqueItems": true }, "mustBeGreaterOrEqualTo": { "description": "Must be greater than or equal to the value to be valid. It is equivalent to '\u003e='.", "type": "number" }, "mustBeGreaterThan": { "description": "Must be greater than the value to be valid. It is equivalent to '\u003e'.", "type": "number" }, "mustBeLessOrEqualTo": { "description": "Must be less than or equal to the value to be valid. It is equivalent to '\u003c='.", "type": "number" }, "mustBeLessThan": { "description": "Must be less than the value to be valid. It is equivalent to '\u003c'.", "type": "number" }, "mustNotBe": { "description": "Must not be equal to the value to be valid. When using numbers, it is equivalent to '!='." }, "mustNotBeBetween": { "description": "Must not be between the two numbers to be valid. Smallest number first in the array.", "items": { "type": "number" }, "maxItems": 2, "minItems": 2, "type": "array", "uniqueItems": true }, "rule": { "description": "Define a data quality check based on the predefined rules as per ODCS.", "examples": [ "duplicateCount", "validValues", "rowCount" ], "type": "string" } }, "required": [ "rule" ], "type": "object" }, "DataQualitySql": { "properties": { "query": { "description": "Query string that adheres to the dialect of the provided server.", "examples": [ "SELECT COUNT(*) FROM ${table} WHERE ${column} IS NOT NULL" ], "type": "string" } }, "required": [ "query" ], "type": "object" }, "Pricing": { "properties": { "priceAmount": { "description": "Subscription price per unit of measure in `priceUnit`.", "type": "number" }, "priceCurrency": { "description": "Currency of the subscription price in `price.priceAmount`.", "type": "string" }, "priceUnit": { "description": "The unit of measure for calculating cost. Examples megabyte, gigabyte.", "type": "string" } }, "type": "object" }, "Role": { "properties": { "access": { "description": "The type of access provided by the IAM role.", "type": "string" }, "description": { "description": "Description of the IAM role and its permissions.", "type": "string" }, "firstLevelApprovers": { "description": "The name(s) of the first-level approver(s) of the role.", "type": "string" }, "role": { "description": "Name of the IAM role that provides access to the dataset.", "type": "string" }, "secondLevelApprovers": { "description": "The name(s) of the second-level approver(s) of the role.", "type": "string" } }, "required": [ "role" ], "type": "object" }, "SchemaBaseProperty": { "allOf": [ { "$ref": "#/$defs/SchemaElement" }, { "if": { "properties": { "logicalType": { "const": "string" } } }, "then": { "properties": { "logicalTypeOptions": { "additionalProperties": false, "properties": { "format": { "description": "Provides extra context about what format the string follows.", "examples": [ "password", "byte", "binary", "email", "uuid", "uri", "hostname", "ipv4", "ipv6" ], "type": "string" }, "maxLength": { "description": "Maximum length of the string.", "minimum": 0, "type": "integer" }, "minLength": { "description": "Minimum length of the string.", "minimum": 0, "type": "integer" }, "pattern": { "description": "Regular expression pattern to define valid value. Follows regular expression syntax from ECMA-262 (https://262.ecma-international.org/5.1/#sec-15.10.1).", "type": "string" } }, "type": "object" } } } }, { "if": { "properties": { "logicalType": { "const": "date" } } }, "then": { "properties": { "logicalTypeOptions": { "additionalProperties": false, "properties": { "exclusiveMaximum": { "default": false, "description": "If set to true, all values are strictly less than the maximum value (values \u003c maximum). Otherwise, less than or equal to the maximum value (values \u003c= maximum).", "type": "boolean" }, "exclusiveMinimum": { "default": false, "description": "If set to true, all values are strictly greater than the minimum value (values \u003e minimum). Otherwise, greater than or equal to the minimum value (values \u003e= minimum).", "type": "boolean" }, "format": { "description": "Format of the date. Follows the format as prescribed by [JDK DateTimeFormatter](https://docs.oracle.com/javase/8/docs/api/java/time/format/DateTimeFormatter.html). For example, format 'yyyy-MM-dd'.", "examples": [ "yyyy-MM-dd", "yyyy-MM-dd HH:mm:ss", "HH:mm:ss" ], "type": "string" }, "maximum": { "description": "All date values are less than or equal to this value (values \u003c= maximum).", "type": "string" }, "minimum": { "description": "All date values are greater than or equal to this value (values \u003e= minimum).", "type": "string" } }, "type": "object" } } } }, { "if": { "anyOf": [ { "properties": { "logicalType": { "const": "integer" } } } ] }, "then": { "properties": { "logicalTypeOptions": { "additionalProperties": false, "properties": { "exclusiveMaximum": { "default": false, "description": "If set to true, all values are strictly less than the maximum value (values \u003c maximum). Otherwise, less than or equal to the maximum value (values \u003c= maximum).", "type": "boolean" }, "exclusiveMinimum": { "default": false, "description": "If set to true, all values are strictly greater than the minimum value (values \u003e minimum). Otherwise, greater than or equal to the minimum value (values \u003e= minimum).", "type": "boolean" }, "format": { "default": "i32", "description": "Format of the value in terms of how many bits of space it can use and whether it is signed or unsigned (follows the Rust integer types).", "enum": [ "i8", "i16", "i32", "i64", "i128", "u8", "u16", "u32", "u64", "u128" ], "type": "string" }, "maximum": { "description": "All values are less than or equal to this value (values \u003c= maximum).", "type": "number" }, "minimum": { "description": "All values are greater than or equal to this value (values \u003e= minimum).", "type": "number" }, "multipleOf": { "description": "Values must be multiples of this number. For example, multiple of 5 has valid values 0, 5, 10, -5.", "exclusiveMinimum": 0, "type": "number" } }, "type": "object" } } } }, { "if": { "anyOf": [ { "properties": { "logicalType": { "const": "number" } } } ] }, "then": { "properties": { "logicalTypeOptions": { "additionalProperties": false, "properties": { "exclusiveMaximum": { "default": false, "description": "If set to true, all values are strictly less than the maximum value (values \u003c maximum). Otherwise, less than or equal to the maximum value (values \u003c= maximum).", "type": "boolean" }, "exclusiveMinimum": { "default": false, "description": "If set to true, all values are strictly greater than the minimum value (values \u003e minimum). Otherwise, greater than or equal to the minimum value (values \u003e= minimum).", "type": "boolean" }, "format": { "default": "i32", "description": "Format of the value in terms of how many bits of space it can use (follows the Rust float types).", "enum": [ "f32", "f64" ], "type": "string" }, "maximum": { "description": "All values are less than or equal to this value (values \u003c= maximum).", "type": "number" }, "minimum": { "description": "All values are greater than or equal to this value (values \u003e= minimum).", "type": "number" }, "multipleOf": { "description": "Values must be multiples of this number. For example, multiple of 5 has valid values 0, 5, 10, -5.", "exclusiveMinimum": 0, "type": "number" } }, "type": "object" } } } }, { "if": { "properties": { "logicalType": { "const": "object" } } }, "then": { "properties": { "logicalTypeOptions": { "additionalProperties": false, "properties": { "maxProperties": { "description": "Maximum number of properties.", "minimum": 0, "type": "integer" }, "minProperties": { "default": 0, "description": "Minimum number of properties.", "minimum": 0, "type": "integer" }, "required": { "description": "Property names that are required to exist in the object.", "items": { "type": "string" }, "minItems": 1, "type": "array", "uniqueItems": true } }, "type": "object" }, "properties": { "description": "A list of properties for the object.", "items": { "$ref": "#/$defs/SchemaProperty" }, "type": "array" } } } }, { "if": { "properties": { "logicalType": { "const": "array" } } }, "then": { "properties": { "items": { "$ref": "#/$defs/SchemaItemProperty", "description": "List of items in an array (only applicable when `logicalType: array`)." }, "logicalTypeOptions": { "additionalProperties": false, "properties": { "maxItems": { "description": "Maximum number of items.", "minimum": 0, "type": "integer" }, "minItems": { "default": 0, "description": "Minimum number of items", "minimum": 0, "type": "integer" }, "uniqueItems": { "default": false, "description": "If set to true, all items in the array are unique.", "type": "boolean" } }, "type": "object" } } } } ], "properties": { "classification": { "description": "Can be anything, like confidential, restricted, and public to more advanced categorization. Some companies like PayPal, use data classification indicating the class of data in the element; expected values are 1, 2, 3, 4, or 5.", "examples": [ "confidential", "restricted", "public" ], "type": "string" }, "criticalDataElement": { "default": false, "description": "True or false indicator; If element is considered a critical data element (CDE) then true else false.", "type": "boolean" }, "encryptedName": { "description": "The element name within the dataset that contains the encrypted element value. For example, unencrypted element `email_address` might have an encryptedName of `email_address_encrypt`.", "type": "string" }, "examples": { "description": "List of sample element values.", "items": { "$ref": "#/$defs/AnyType" }, "type": "array" }, "logicalType": { "description": "The logical element data type.", "enum": [ "string", "date", "number", "integer", "object", "array", "boolean" ], "type": "string" }, "logicalTypeOptions": { "description": "Additional optional metadata to describe the logical type.", "type": "object" }, "partitionKeyPosition": { "default": -1, "description": "If element is used for partitioning, the position of the partition element. Starts from 1. Example of `country, year` being partition columns, `country` has partitionKeyPosition 1 and `year` partitionKeyPosition 2. Default to -1.", "type": "integer" }, "partitioned": { "default": false, "description": "Indicates if the element is partitioned; possible values are true and false.", "type": "boolean" }, "physicalType": { "description": "The physical element data type in the data source. For example, VARCHAR(2), DOUBLE, INT.", "type": "string" }, "primaryKey": { "description": "Boolean value specifying whether the element is primary or not. Default is false.", "type": "boolean" }, "primaryKeyPosition": { "default": -1, "description": "If element is a primary key, the position of the primary key element. Starts from 1. Example of `account_id, name` being primary key columns, `account_id` has primaryKeyPosition 1 and `name` primaryKeyPosition 2. Default to -1.", "type": "integer" }, "quality": { "$ref": "#/$defs/DataQualityChecks" }, "required": { "default": false, "description": "Indicates if the element may contain Null values; possible values are true and false. Default is false.", "type": "boolean" }, "transformDescription": { "description": "Describes the transform logic in very simple terms.", "type": "string" }, "transformLogic": { "description": "Logic used in the element transformation.", "type": "string" }, "transformSourceObjects": { "description": "List of objects in the data source used in the transformation.", "items": { "type": "string" }, "type": "array" }, "unique": { "default": false, "description": "Indicates if the element contains unique values; possible values are true and false. Default is false.", "type": "boolean" } }, "type": "object" }, "SchemaElement": { "properties": { "authoritativeDefinitions": { "$ref": "#/$defs/AuthoritativeDefinitions" }, "businessName": { "description": "The business name of the element.", "type": "string" }, "customProperties": { "$ref": "#/$defs/CustomProperties" }, "description": { "description": "Description of the element.", "type": "string" }, "name": { "description": "Name of the element.", "type": "string" }, "physicalType": { "description": "The physical element data type in the data source.", "examples": [ "table", "view", "topic", "file" ], "type": "string" }, "tags": { "$ref": "#/$defs/Tags" } }, "type": "object" }, "SchemaItemProperty": { "$ref": "#/$defs/SchemaBaseProperty", "properties": { "properties": { "description": "A list of properties for the object.", "items": { "$ref": "#/$defs/SchemaProperty" }, "type": "array" } }, "type": "object", "unevaluatedProperties": false }, "SchemaObject": { "allOf": [ { "$ref": "#/$defs/SchemaElement" } ], "properties": { "dataGranularityDescription": { "description": "Granular level of the data in the object.", "examples": [ "Aggregation by country" ], "type": "string" }, "logicalType": { "description": "The logical element data type.", "enum": [ "object" ], "type": "string" }, "physicalName": { "description": "Physical name.", "examples": [ "table_1_2_0" ], "type": "string" }, "properties": { "description": "A list of properties for the object.", "items": { "$ref": "#/$defs/SchemaProperty" }, "type": "array" }, "quality": { "$ref": "#/$defs/DataQualityChecks" } }, "required": [ "name" ], "type": "object", "unevaluatedProperties": false }, "SchemaProperty": { "$ref": "#/$defs/SchemaBaseProperty", "required": [ "name" ], "type": "object", "unevaluatedProperties": false }, "Server": { "allOf": [ { "if": { "properties": { "type": { "const": "api" } }, "required": [ "type" ] }, "then": { "$ref": "#/$defs/ServerSource/ApiServer" } }, { "if": { "properties": { "type": { "const": "athena" } }, "required": [ "type" ] }, "then": { "$ref": "#/$defs/ServerSource/AthenaServer" } }, { "if": { "properties": { "type": { "const": "azure" } }, "required": [ "type" ] }, "then": { "$ref": "#/$defs/ServerSource/AzureServer" } }, { "if": { "properties": { "type": { "const": "bigquery" } }, "required": [ "type" ] }, "then": { "$ref": "#/$defs/ServerSource/BigQueryServer" } }, { "if": { "properties": { "type": { "const": "clickhouse" } }, "required": [ "type" ] }, "then": { "$ref": "#/$defs/ServerSource/ClickHouseServer" } }, { "if": { "properties": { "type": { "const": "databricks" } }, "required": [ "type" ] }, "then": { "$ref": "#/$defs/ServerSource/DatabricksServer" } }, { "if": { "properties": { "type": { "const": "denodo" } }, "required": [ "type" ] }, "then": { "$ref": "#/$defs/ServerSource/DenodoServer" } }, { "if": { "properties": { "type": { "const": "dremio" } }, "required": [ "type" ] }, "then": { "$ref": "#/$defs/ServerSource/DremioServer" } }, { "if": { "properties": { "type": { "const": "duckdb" } }, "required": [ "type" ] }, "then": { "$ref": "#/$defs/ServerSource/DuckdbServer" } }, { "if": { "properties": { "type": { "const": "glue" } }, "required": [ "type" ] }, "then": { "$ref": "#/$defs/ServerSource/GlueServer" } }, { "if": { "properties": { "type": { "const": "cloudsql" } }, "required": [ "type" ] }, "then": { "$ref": "#/$defs/ServerSource/GoogleCloudSqlServer" } }, { "if": { "properties": { "type": { "const": "db2" } }, "required": [ "type" ] }, "then": { "$ref": "#/$defs/ServerSource/IBMDB2Server" } }, { "if": { "properties": { "type": { "const": "informix" } }, "required": [ "type" ] }, "then": { "$ref": "#/$defs/ServerSource/InformixServer" } }, { "if": { "properties": { "type": { "const": "custom" } }, "required": [ "type" ] }, "then": { "$ref": "#/$defs/ServerSource/CustomServer" } }, { "if": { "properties": { "type": { "const": "kafka" } }, "required": [ "type" ] }, "then": { "$ref": "#/$defs/ServerSource/KafkaServer" } }, { "if": { "properties": { "type": { "const": "kinesis" } }, "required": [ "type" ] }, "then": { "$ref": "#/$defs/ServerSource/KinesisServer" } }, { "if": { "properties": { "type": { "const": "local" } }, "required": [ "type" ] }, "then": { "$ref": "#/$defs/ServerSource/LocalServer" } }, { "if": { "properties": { "type": { "const": "mysql" } }, "required": [ "type" ] }, "then": { "$ref": "#/$defs/ServerSource/MySqlServer" } }, { "if": { "properties": { "type": { "const": "oracle" } }, "required": [ "type" ] }, "then": { "$ref": "#/$defs/ServerSource/OracleServer" } }, { "if": { "properties": { "type": { "const": "postgresql" } }, "required": [ "type" ] }, "then": { "$ref": "#/$defs/ServerSource/PostgresServer" } }, { "if": { "properties": { "type": { "const": "postgres" } }, "required": [ "type" ] }, "then": { "$ref": "#/$defs/ServerSource/PostgresServer" } }, { "if": { "properties": { "type": { "const": "presto" } }, "required": [ "type" ] }, "then": { "$ref": "#/$defs/ServerSource/PrestoServer" } }, { "if": { "properties": { "type": { "const": "pubsub" } }, "required": [ "type" ] }, "then": { "$ref": "#/$defs/ServerSource/PubSubServer" } }, { "if": { "properties": { "type": { "const": "redshift" } }, "required": [ "type" ] }, "then": { "$ref": "#/$defs/ServerSource/RedshiftServer" } }, { "if": { "properties": { "type": { "const": "s3" } }, "required": [ "type" ] }, "then": { "$ref": "#/$defs/ServerSource/S3Server" } }, { "if": { "properties": { "type": { "const": "sftp" } }, "required": [ "type" ] }, "then": { "$ref": "#/$defs/ServerSource/SftpServer" } }, { "if": { "properties": { "type": { "const": "snowflake" } }, "required": [ "type" ] }, "then": { "$ref": "#/$defs/ServerSource/SnowflakeServer" } }, { "if": { "properties": { "type": { "const": "sqlserver" } }, "required": [ "type" ] }, "then": { "$ref": "#/$defs/ServerSource/SqlserverServer" } }, { "if": { "properties": { "type": { "const": "synapse" } }, "required": [ "type" ] }, "then": { "$ref": "#/$defs/ServerSource/SynapseServer" } }, { "if": { "properties": { "type": { "const": "trino" } }, "required": [ "type" ] }, "then": { "$ref": "#/$defs/ServerSource/TrinoServer" } }, { "if": { "properties": { "type": { "const": "vertica" } }, "required": [ "type" ] }, "then": { "$ref": "#/$defs/ServerSource/VerticaServer" } } ], "description": "Data source details of where data is physically stored.", "properties": { "customProperties": { "$ref": "#/$defs/CustomProperties" }, "description": { "description": "Description of the server.", "type": "string" }, "environment": { "description": "Environment of the server.", "examples": [ "prod", "preprod", "dev", "uat" ], "type": "string" }, "roles": { "description": "List of roles that have access to the server.", "items": { "$ref": "#/$defs/Role" }, "type": "array" }, "server": { "description": "Identifier of the server.", "type": "string" }, "type": { "description": "Type of the server.", "enum": [ "api", "athena", "azure", "bigquery", "clickhouse", "databricks", "denodo", "dremio", "duckdb", "glue", "cloudsql", "db2", "informix", "kafka", "kinesis", "local", "mysql", "oracle", "postgresql", "postgres", "presto", "pubsub", "redshift", "s3", "sftp", "snowflake", "sqlserver", "synapse", "trino", "vertica", "custom" ], "type": "string" } }, "required": [ "server", "type" ], "type": "object" }, "ServerSource": { "ApiServer": { "properties": { "location": { "description": "The url to the API.", "examples": [ "https://api.example.com/v1" ], "format": "uri", "type": "string" } }, "required": [ "location" ], "title": "AthenaServer", "type": "object" }, "AthenaServer": { "properties": { "catalog": { "default": "awsdatacatalog", "description": "Identify the name of the Data Source, also referred to as a Catalog.", "type": "string" }, "regionName": { "description": "The region your AWS account uses.", "examples": [ "eu-west-1" ], "type": "string" }, "schema": { "description": "Identify the schema in the data source in which your tables exist.", "type": "string" }, "stagingDir": { "description": "Amazon Athena automatically stores query results and metadata information for each query that runs in a query result location that you can specify in Amazon S3.", "examples": [ "s3://my_storage_account_name/my_container/path" ], "format": "uri", "type": "string" } }, "required": [ "staging_dir", "schema" ], "title": "AthenaServer", "type": "object" }, "AzureServer": { "properties": { "delimiter": { "description": "Only for format = json. How multiple json documents are delimited within one file", "enum": [ "new_line", "array" ], "type": "string" }, "format": { "description": "File format.", "enum": [ "parquet", "delta", "json", "csv" ], "type": "string" }, "location": { "description": "Fully qualified path to Azure Blob Storage or Azure Data Lake Storage (ADLS), supports globs.", "examples": [ "az://my_storage_account_name.blob.core.windows.net/my_container/path/*.parquet", "abfss://my_storage_account_name.dfs.core.windows.net/my_container_name/path/*.parquet" ], "format": "uri", "type": "string" } }, "required": [ "location", "format" ], "title": "AzureServer", "type": "object" }, "BigQueryServer": { "properties": { "dataset": { "description": "The GCP dataset name.", "type": "string" }, "project": { "description": "The GCP project name.", "type": "string" } }, "required": [ "project", "dataset" ], "title": "BigQueryServer", "type": "object" }, "ClickHouseServer": { "properties": { "database": { "description": "The name of the database.", "type": "string" }, "host": { "description": "The host of the ClickHouse server.", "type": "string" }, "port": { "description": "The port to the ClickHouse server.", "type": "integer" } }, "required": [ "host", "port", "database" ], "title": "ClickHouseServer", "type": "object" }, "CustomServer": { "properties": { "account": { "description": "Account used by the server.", "type": "string" }, "catalog": { "description": "Name of the catalog.", "type": "string" }, "database": { "description": "Name of the database.", "type": "string" }, "dataset": { "description": "Name of the dataset.", "type": "string" }, "delimiter": { "description": "Delimiter.", "type": "string" }, "endpointUrl": { "description": "Server endpoint.", "format": "uri", "type": "string" }, "format": { "description": "File format.", "type": "string" }, "host": { "description": "Host name or IP address.", "type": "string" }, "location": { "description": "A URL to a location.", "format": "uri", "type": "string" }, "path": { "description": "Relative or absolute path to the data file(s).", "type": "string" }, "port": { "description": "Port to the server. No default value is assumed for custom servers.", "type": "integer" }, "project": { "description": "Project name.", "type": "string" }, "region": { "description": "Cloud region.", "type": "string" }, "regionName": { "description": "Region name.", "type": "string" }, "schema": { "description": "Name of the schema.", "type": "string" }, "serviceName": { "description": "Name of the service.", "type": "string" }, "stagingDir": { "description": "Staging directory.", "type": "string" }, "warehouse": { "description": "Name of the cluster or warehouse.", "type": "string" } }, "title": "CustomServer", "type": "object" }, "DatabricksServer": { "properties": { "catalog": { "description": "The name of the Hive or Unity catalog", "type": "string" }, "host": { "description": "The Databricks host", "examples": [ "dbc-abcdefgh-1234.cloud.databricks.com" ], "type": "string" }, "schema": { "description": "The schema name in the catalog", "type": "string" } }, "required": [ "catalog", "schema" ], "title": "DatabricksServer", "type": "object" }, "DenodoServer": { "properties": { "database": { "description": "The name of the database.", "type": "string" }, "host": { "description": "The host of the Denodo server.", "type": "string" }, "port": { "description": "The port of the Denodo server.", "type": "integer" } }, "required": [ "host", "port" ], "title": "DenodoServer", "type": "object" }, "DremioServer": { "properties": { "host": { "description": "The host of the Dremio server.", "type": "string" }, "port": { "description": "The port of the Dremio server.", "type": "integer" }, "schema": { "description": "The name of the schema.", "type": "string" } }, "required": [ "host", "port" ], "title": "DremioServer", "type": "object" }, "DuckdbServer": { "properties": { "database": { "description": "Path to duckdb database file.", "type": "string" }, "schema": { "description": "The name of the schema.", "type": "integer" } }, "required": [ "database" ], "title": "DuckdbServer", "type": "object" }, "GlueServer": { "properties": { "account": { "description": "The AWS Glue account", "examples": [ "1234-5678-9012" ], "type": "string" }, "database": { "description": "The AWS Glue database name", "examples": [ "my_database" ], "type": "string" }, "format": { "description": "The format of the files", "examples": [ "parquet", "csv", "json", "delta" ], "type": "string" }, "location": { "description": "The AWS S3 path. Must be in the form of a URL.", "examples": [ "s3://datacontract-example-orders-latest/data/{model}" ], "format": "uri", "type": "string" } }, "required": [ "account", "database" ], "title": "GlueServer", "type": "object" }, "GoogleCloudSqlServer": { "properties": { "database": { "description": "The name of the database.", "type": "string" }, "host": { "description": "The host of the Google Cloud Sql server.", "type": "string" }, "port": { "description": "The port of the Google Cloud Sql server.", "type": "integer" }, "schema": { "description": "The name of the schema.", "type": "string" } }, "required": [ "host", "port", "database", "schema" ], "title": "GoogleCloudSqlServer", "type": "object" }, "IBMDB2Server": { "properties": { "database": { "description": "The name of the database.", "type": "string" }, "host": { "description": "The host of the IBM DB2 server.", "type": "string" }, "port": { "description": "The port of the IBM DB2 server.", "type": "integer" }, "schema": { "description": "The name of the schema.", "type": "string" } }, "required": [ "host", "port", "database" ], "title": "IBMDB2Server", "type": "object" }, "InformixServer": { "properties": { "database": { "description": "The name of the database.", "type": "string" }, "host": { "description": "The host to the Informix server. ", "type": "string" }, "port": { "description": "The port to the Informix server. Defaults to 9088.", "type": "integer" } }, "required": [ "host", "database" ], "title": "InformixServer", "type": "object" }, "KafkaServer": { "description": "Kafka Server", "properties": { "format": { "default": "json", "description": "The format of the messages.", "examples": [ "json", "avro", "protobuf", "xml" ], "type": "string" }, "host": { "description": "The bootstrap server of the kafka cluster.", "type": "string" } }, "required": [ "host" ], "title": "KafkaServer", "type": "object" }, "KinesisServer": { "description": "Kinesis Data Streams Server", "properties": { "format": { "description": "The format of the record", "examples": [ "json", "avro", "protobuf" ], "type": "string" }, "region": { "description": "AWS region.", "examples": [ "eu-west-1" ], "type": "string" } }, "title": "KinesisDataStreamsServer", "type": "object" }, "LocalServer": { "properties": { "format": { "description": "The format of the file(s)", "examples": [ "json", "parquet", "delta", "csv" ], "type": "string" }, "path": { "description": "The relative or absolute path to the data file(s).", "examples": [ "./folder/data.parquet", "./folder/*.parquet" ], "type": "string" } }, "required": [ "path", "format" ], "title": "LocalServer", "type": "object" }, "MySqlServer": { "properties": { "database": { "description": "The name of the database.", "type": "string" }, "host": { "description": "The host of the MySql server.", "type": "string" }, "port": { "description": "The port of the MySql server.", "type": "integer" } }, "required": [ "host", "port", "database" ], "title": "MySqlServer", "type": "object" }, "OracleServer": { "properties": { "host": { "description": "The host to the oracle server", "examples": [ "localhost" ], "type": "string" }, "port": { "description": "The port to the oracle server.", "examples": [ 1523 ], "type": "integer" }, "serviceName": { "description": "The name of the service.", "examples": [ "service" ], "type": "string" } }, "required": [ "host", "port", "serviceName" ], "title": "OracleServer", "type": "object" }, "PostgresServer": { "properties": { "database": { "description": "The name of the database.", "type": "string" }, "host": { "description": "The host to the Postgres server", "type": "string" }, "port": { "description": "The port to the Postgres server.", "type": "integer" }, "schema": { "description": "The name of the schema in the database.", "type": "string" } }, "required": [ "host", "port", "database", "schema" ], "title": "PostgresServer", "type": "object" }, "PrestoServer": { "properties": { "catalog": { "description": "The name of the catalog.", "examples": [ "postgres" ], "type": "string" }, "host": { "description": "The host to the Presto server", "examples": [ "localhost:8080" ], "type": "string" }, "schema": { "description": "The name of the schema.", "examples": [ "public" ], "type": "string" } }, "required": [ "host" ], "title": "PrestoServer", "type": "object" }, "PubSubServer": { "properties": { "project": { "description": "The GCP project name.", "type": "string" } }, "required": [ "project" ], "title": "PubSubServer", "type": "object" }, "RedshiftServer": { "properties": { "account": { "description": "The account used by the server.", "type": "string" }, "database": { "description": "The name of the database.", "type": "string" }, "host": { "description": "An optional string describing the server.", "type": "string" }, "region": { "description": "AWS region of Redshift server.", "examples": [ "us-east-1" ], "type": "string" }, "schema": { "description": "The name of the schema.", "type": "string" } }, "required": [ "database", "schema" ], "title": "RedshiftServer", "type": "object" }, "S3Server": { "properties": { "delimiter": { "description": "Only for format = json. How multiple json documents are delimited within one file", "enum": [ "new_line", "array" ], "type": "string" }, "endpointUrl": { "description": "The server endpoint for S3-compatible servers.", "examples": [ "https://minio.example.com" ], "format": "uri", "type": "string" }, "format": { "description": "File format.", "enum": [ "parquet", "delta", "json", "csv" ], "type": "string" }, "location": { "description": "S3 URL, starting with `s3://`", "examples": [ "s3://datacontract-example-orders-latest/data/{model}/*.json" ], "format": "uri", "type": "string" } }, "required": [ "location" ], "title": "S3Server", "type": "object" }, "SftpServer": { "properties": { "delimiter": { "description": "Only for format = json. How multiple json documents are delimited within one file", "enum": [ "new_line", "array" ], "type": "string" }, "format": { "description": "File format.", "enum": [ "parquet", "delta", "json", "csv" ], "type": "string" }, "location": { "description": "SFTP URL, starting with `sftp://`", "examples": [ "sftp://123.123.12.123/{model}/*.json" ], "format": "uri", "pattern": "^sftp://.*", "type": "string" } }, "required": [ "location" ], "title": "SftpServer", "type": "object" }, "SnowflakeServer": { "properties": { "account": { "description": "The Snowflake account used by the server.", "type": "string" }, "database": { "description": "The name of the database.", "type": "string" }, "host": { "description": "The host to the Snowflake server", "type": "string" }, "port": { "description": "The port to the Snowflake server.", "type": "integer" }, "schema": { "description": "The name of the schema.", "type": "string" }, "warehouse": { "description": "The name of the cluster of resources that is a Snowflake virtual warehouse.", "type": "string" } }, "required": [ "account", "database", "schema" ], "title": "SnowflakeServer", "type": "object" }, "SqlserverServer": { "properties": { "database": { "description": "The name of the database.", "examples": [ "database" ], "type": "string" }, "host": { "description": "The host to the database server", "examples": [ "localhost" ], "type": "string" }, "port": { "default": 1433, "description": "The port to the database server.", "examples": [ 1433 ], "type": "integer" }, "schema": { "description": "The name of the schema in the database.", "examples": [ "dbo" ], "type": "string" } }, "required": [ "host", "database", "schema" ], "title": "SqlserverServer", "type": "object" }, "SynapseServer": { "properties": { "database": { "description": "The name of the database.", "type": "string" }, "host": { "description": "The host of the Synapse server.", "type": "string" }, "port": { "description": "The port of the Synapse server.", "type": "integer" } }, "required": [ "host", "port", "database" ], "title": "SynapseServer", "type": "object" }, "TrinoServer": { "properties": { "catalog": { "description": "The name of the catalog.", "examples": [ "hive" ], "type": "string" }, "host": { "description": "The Trino host URL.", "examples": [ "localhost" ], "type": "string" }, "port": { "description": "The Trino port.", "type": "integer" }, "schema": { "description": "The name of the schema in the database.", "examples": [ "my_schema" ], "type": "string" } }, "required": [ "host", "port", "catalog", "schema" ], "title": "TrinoServer", "type": "object" }, "VerticaServer": { "properties": { "database": { "description": "The name of the database.", "type": "string" }, "host": { "description": "The host of the Vertica server.", "type": "string" }, "port": { "description": "The port of the Vertica server.", "type": "integer" }, "schema": { "description": "The name of the schema.", "type": "string" } }, "required": [ "host", "port", "database", "schema" ], "title": "VerticaServer", "type": "object" } }, "ServiceLevelAgreementProperty": { "properties": { "driver": { "description": "Describes the importance of the SLA from the list of: `regulatory`, `analytics`, or `operational`.", "examples": [ "regulatory", "analytics", "operational" ], "type": "string" }, "element": { "description": "Element(s) to check on. Multiple elements should be extremely rare and, if so, separated by commas.", "type": "string" }, "property": { "description": "Specific property in SLA, check the periodic table. May requires units (more details to come).", "type": "string" }, "unit": { "description": "**d**, day, days for days; **y**, yr, years for years, etc. Units use the ISO standard.", "type": "string" }, "value": { "anyOf": [ { "type": "string" }, { "type": "number" }, { "type": "integer" }, { "type": "boolean" }, { "type": "null" } ], "description": "Agreement value. The label will change based on the property itself." }, "valueExt": { "$ref": "#/$defs/AnyNonCollectionType", "description": "Extended agreement value. The label will change based on the property itself." } }, "required": [ "property", "value" ], "type": "object" }, "Support": { "description": "Top level for support channels.", "items": { "$ref": "#/$defs/SupportItem" }, "type": "array" }, "SupportItem": { "properties": { "channel": { "description": "Channel name or identifier.", "type": "string" }, "description": { "description": "Description of the channel, free text.", "type": "string" }, "invitationUrl": { "description": "Some tools uses invitation URL for requesting or subscribing. Follows the [URL scheme](https://en.wikipedia.org/wiki/URL#Syntax).", "type": "string" }, "scope": { "description": "Scope can be: `interactive`, `announcements`, `issues`.", "examples": [ "interactive", "announcements", "issues" ], "type": "string" }, "tool": { "description": "Name of the tool, value can be `email`, `slack`, `teams`, `discord`, `ticket`, or `other`.", "examples": [ "email", "slack", "teams", "discord", "ticket", "other" ], "type": "string" }, "url": { "description": "Access URL using normal [URL scheme](https://en.wikipedia.org/wiki/URL#Syntax) (https, mailto, etc.).", "type": "string" } }, "required": [ "channel", "url" ], "type": "object" }, "Tags": { "description": "A list of tags that may be assigned to the elements (object or property); the tags keyword may appear at any level.", "items": { "type": "string" }, "type": "array" }, "Team": { "properties": { "dateIn": { "description": "The date when the user joined the team.", "format": "date", "type": "string" }, "dateOut": { "description": "The date when the user ceased to be part of the team.", "format": "date", "type": "string" }, "replacedByUsername": { "description": "The username of the user who replaced the previous user.", "type": "string" }, "role": { "description": "The user's job role; Examples might be owner, data steward. There is no limit on the role.", "type": "string" }, "username": { "description": "The user's username or email.", "type": "string" } }, "type": "object" } }, "$schema": "https://json-schema.org/draft/2019-09/schema", "additionalProperties": false, "description": "An open data contract specification to establish agreement between data producers and consumers.", "properties": { "apiVersion": { "default": "v3.0.0", "description": "Version of the standard used to build data contract. Default value is v3.0.0.", "enum": [ "v3.0.0", "v2.2.2", "v2.2.1", "v2.2.0" ], "type": "string" }, "contractCreatedTs": { "description": "Timestamp in UTC of when the data contract was created.", "format": "date-time", "type": "string" }, "customProperties": { "$ref": "#/$defs/CustomProperties" }, "dataProduct": { "description": "The name of the data product.", "type": "string" }, "description": { "description": "High level description of the dataset.", "properties": { "limitations": { "description": "Limitations of the dataset.", "type": "string" }, "purpose": { "description": "Purpose of the dataset.", "type": "string" }, "usage": { "description": "Intended usage of the dataset.", "type": "string" } }, "type": "object" }, "domain": { "description": "Name of the logical data domain.", "examples": [ "imdb_ds_aggregate", "receiver_profile_out", "transaction_profile_out" ], "type": "string" }, "id": { "description": "A unique identifier used to reduce the risk of dataset name collisions, such as a UUID.", "type": "string" }, "kind": { "default": "DataContract", "description": "The kind of file this is. Valid value is `DataContract`.", "enum": [ "DataContract" ], "type": "string" }, "name": { "description": "Name of the data contract.", "type": "string" }, "price": { "$ref": "#/$defs/Pricing" }, "provenance": { "additionalProperties": false, "description": "Origin metadata recording where this contract was introspected from.", "properties": { "introspectedAt": { "format": "date-time", "type": "string" }, "introspectedFrom": { "type": "string" }, "sourceDriver": { "type": "string" }, "sourceEnvironment": { "type": "string" } }, "type": "object" }, "roles": { "description": "A list of roles that will provide user access to the dataset.", "items": { "$ref": "#/$defs/Role" }, "type": "array" }, "schema": { "description": "A list of elements within the schema to be cataloged.", "items": { "$ref": "#/$defs/SchemaObject" }, "type": "array" }, "servers": { "description": "List of servers where the datasets reside.", "items": { "$ref": "#/$defs/Server" }, "type": "array" }, "slaDefaultElement": { "description": "Element (using the element path notation) to do the checks on.", "type": "string" }, "slaProperties": { "description": "A list of key/value pairs for SLA specific properties. There is no limit on the type of properties (more details to come).", "items": { "$ref": "#/$defs/ServiceLevelAgreementProperty" }, "type": "array" }, "status": { "description": "Current status of the dataset. Valid values are `production`, `test`, or `development`.", "examples": [ "production", "test", "development" ], "type": "string" }, "support": { "$ref": "#/$defs/Support" }, "tags": { "$ref": "#/$defs/Tags" }, "team": { "items": { "$ref": "#/$defs/Team" }, "type": "array" }, "tenant": { "description": "Indicates the property the data is primarily associated with. Value is case insensitive.", "type": "string" }, "version": { "description": "Current version of the data contract.", "type": "string" } }, "required": [ "version", "apiVersion", "kind", "id", "status" ], "title": "Open Data Contract Standard (ODCS)", "type": "object" }