Datafold diff_created API

openapi: 3.1.0
info:
  contact:
    email: support@datafold.com
    name: API Support
  description: "The Datafold API reference is a guide to our available endpoints and authentication methods.\nIf you're just getting started with Datafold, we recommend first checking out our [documentation](https://docs.datafold.com).\n\n:::info\n  To use the Datafold API, you should first create a Datafold API Key,\n  which should be stored as a local environment variable named DATAFOLD_API_KEY.\n  This can be set in your Datafold Cloud's Settings under the Account page.\n:::"
  title: Datafold Audit Logs diff_created API
  version: latest
servers:
- description: Default server
  url: https://app.datafold.com
security:
- ApiKeyAuth: []
tags:
- name: diff_created
paths:
  /api/v1/datadiffs:
    post:
      description: 'Launches a new data diff to compare two datasets (tables or queries).


        A data diff identifies differences between two datasets by comparing:

        - Row-level changes (added, removed, modified rows)

        - Schema differences

        - Column-level statistics


        The diff runs asynchronously. Use the returned diff ID to poll for status and retrieve results.'
      operationId: create_datadiff_api_v1_datadiffs_post
      requestBody:
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/ApiDataDiffData'
        required: true
      responses:
        '200':
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ApiDataDiffFull'
          description: Successful Response
        '422':
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/HTTPValidationError'
          description: Validation Error
      summary: Create a data diff
      tags:
      - diff_created
components:
  schemas:
    TableModifiers:
      enum:
      - case_insensitive_strings
      - null_equals_empty_string
      title: TableModifiers
      type: string
    ValidationError:
      properties:
        loc:
          items:
            anyOf:
            - type: string
            - type: integer
          title: Location
          type: array
        msg:
          title: Message
          type: string
        type:
          title: Error Type
          type: string
      required:
      - loc
      - msg
      - type
      title: ValidationError
      type: object
    TDataDiffDataAppMetadata:
      properties:
        data_app_id:
          title: Data App Id
          type: integer
        data_app_model1_id:
          anyOf:
          - type: string
          - type: 'null'
          title: Data App Model1 Id
        data_app_model1_name:
          anyOf:
          - type: string
          - type: 'null'
          title: Data App Model1 Name
        data_app_model2_id:
          anyOf:
          - type: string
          - type: 'null'
          title: Data App Model2 Id
        data_app_model2_name:
          anyOf:
          - type: string
          - type: 'null'
          title: Data App Model2 Name
        data_app_model_type:
          title: Data App Model Type
          type: string
        meta_data:
          additionalProperties: true
          title: Meta Data
          type: object
      required:
      - data_app_id
      - data_app_model_type
      - meta_data
      title: TDataDiffDataAppMetadata
      type: object
    HTTPValidationError:
      properties:
        detail:
          items:
            $ref: '#/components/schemas/ValidationError'
          title: Detail
          type: array
      title: HTTPValidationError
      type: object
    ApiDataDiffFull:
      properties:
        affected_columns:
          anyOf:
          - items:
              type: string
            type: array
          - type: 'null'
          title: Affected Columns
        algorithm:
          anyOf:
          - $ref: '#/components/schemas/DiffAlgorithm'
          - type: 'null'
          description: 'Diff algorithm. ''join'' for same-database diffs, ''fetch_and_join'' for cross-database or file diffs. Auto-selected if omitted: ''join'' when both data sources are the same, ''fetch_and_join'' otherwise.'
        archived:
          default: false
          title: Archived
          type: boolean
        bisection_factor:
          anyOf:
          - type: integer
          - type: 'null'
          title: Bisection Factor
        bisection_threshold:
          anyOf:
          - type: integer
          - type: 'null'
          title: Bisection Threshold
        ci_base_branch:
          anyOf:
          - type: string
          - type: 'null'
          title: Ci Base Branch
        ci_pr_branch:
          anyOf:
          - type: string
          - type: 'null'
          title: Ci Pr Branch
        ci_pr_num:
          anyOf:
          - type: integer
          - type: 'null'
          title: Ci Pr Num
        ci_pr_sha:
          anyOf:
          - type: string
          - type: 'null'
          title: Ci Pr Sha
        ci_pr_url:
          anyOf:
          - type: string
          - type: 'null'
          title: Ci Pr Url
        ci_pr_user_display_name:
          anyOf:
          - type: string
          - type: 'null'
          title: Ci Pr User Display Name
        ci_pr_user_email:
          anyOf:
          - type: string
          - type: 'null'
          title: Ci Pr User Email
        ci_pr_user_id:
          anyOf:
          - type: string
          - type: 'null'
          title: Ci Pr User Id
        ci_pr_username:
          anyOf:
          - type: string
          - type: 'null'
          title: Ci Pr Username
        ci_run_id:
          anyOf:
          - type: integer
          - type: 'null'
          title: Ci Run Id
        ci_sha_url:
          anyOf:
          - type: string
          - type: 'null'
          title: Ci Sha Url
        column_mapping:
          anyOf:
          - items:
              maxItems: 2
              minItems: 2
              prefixItems:
              - type: string
              - type: string
              type: array
            type: array
          - type: 'null'
          description: Map columns with different names between datasets. List of [column_in_A, column_in_B] pairs.
          title: Column Mapping
        column_rules:
          anyOf:
          - items:
              $ref: '#/components/schemas/ApiColumnRule'
            type: array
          - type: 'null'
          description: Extra equality clauses that extend the in-memory diff value comparator. Useful for treating specific value pairs (e.g. legacy sentinels vs NULL) as non-differences. See `ApiColumnRule` for entry shape.
          title: Column Rules
        columns_to_compare:
          anyOf:
          - items:
              type: string
            type: array
          - type: 'null'
          description: Columns to compare between datasets. If set, only these columns are diffed (primary key columns are always included). Column names must match the dataset schema.
          title: Columns To Compare
        compare_duplicates:
          anyOf:
          - type: boolean
          - type: 'null'
          description: Compare rows with duplicate primary keys. Defaults to true.
          title: Compare Duplicates
        created_at:
          anyOf:
          - format: date-time
            type: string
          - type: 'null'
          title: Created At
        data_app_metadata:
          anyOf:
          - $ref: '#/components/schemas/TDataDiffDataAppMetadata'
          - type: 'null'
        data_app_type:
          anyOf:
          - type: string
          - type: 'null'
          title: Data App Type
        data_source1_id:
          description: ID of the first data source (Dataset A).
          title: Data Source1 Id
          type: integer
        data_source1_session_parameters:
          anyOf:
          - additionalProperties: true
            type: object
          - type: 'null'
          description: 'Snowflake session parameters for Dataset A, e.g. {"QUERY_TAG": "datadiff", "WAREHOUSE": "COMPUTE_WH"}.'
          title: Data Source1 Session Parameters
        data_source2_id:
          description: ID of the second data source (Dataset B). Can be the same as data_source1_id.
          title: Data Source2 Id
          type: integer
        data_source2_session_parameters:
          anyOf:
          - additionalProperties: true
            type: object
          - type: 'null'
          description: Snowflake session parameters for Dataset B.
          title: Data Source2 Session Parameters
        datetime_tolerance:
          anyOf:
          - type: integer
          - type: 'null'
          description: Datetime precision for comparison. 0=seconds, 1=tenths, 2=hundredths, 3=milliseconds, 4=tenth-ms, 5=hundredth-ms, 6=microseconds.
          title: Datetime Tolerance
        diff_stats:
          anyOf:
          - $ref: '#/components/schemas/DiffStats'
          - type: 'null'
        diff_tolerance:
          anyOf:
          - type: number
          - type: 'null'
          description: 'Default tolerance for float comparisons. In absolute mode: values within this distance are equal. In relative mode: fraction of difference allowed.'
          title: Diff Tolerance
        diff_tolerances_per_column:
          anyOf:
          - items:
              $ref: '#/components/schemas/ColumnTolerance'
            type: array
          - type: 'null'
          description: 'Per-column tolerance overrides. Each entry: {column_name, tolerance_value (>= 0), tolerance_mode: ''absolute''|''relative''}.'
          title: Diff Tolerances Per Column
        done:
          anyOf:
          - type: boolean
          - type: 'null'
          default: false
          title: Done
        download_limit:
          anyOf:
          - type: integer
          - type: 'null'
          title: Download Limit
        exclude_columns:
          anyOf:
          - items:
              type: string
            type: array
          - type: 'null'
          description: Columns to exclude from comparison. Ignored if include_columns is set.
          title: Exclude Columns
        execute_as_user:
          anyOf:
          - type: boolean
          - type: 'null'
          title: Execute As User
        file1:
          anyOf:
          - format: uri
            minLength: 1
            type: string
          - type: 'null'
          description: File URL for Dataset A (s3://, gs://, abfss://, https://). Mutually exclusive with table1 and query1. Requires file1_options.
          title: File1
        file1_options:
          anyOf:
          - discriminator:
              mapping:
                csv: '#/components/schemas/CSVFileOptions'
                excel: '#/components/schemas/ExcelFileOptions'
                parquet: '#/components/schemas/ParquetFileOptions'
              propertyName: file_type
            oneOf:
            - $ref: '#/components/schemas/CSVFileOptions'
            - $ref: '#/components/schemas/ExcelFileOptions'
            - $ref: '#/components/schemas/ParquetFileOptions'
          - type: 'null'
          description: File format options for file1 (file_type, delimiter, sheet, skip rows).
          title: File1 Options
        file2:
          anyOf:
          - format: uri
            minLength: 1
            type: string
          - type: 'null'
          description: File URL for Dataset B (s3://, gs://, abfss://, https://). Mutually exclusive with table2 and query2. Requires file2_options.
          title: File2
        file2_options:
          anyOf:
          - discriminator:
              mapping:
                csv: '#/components/schemas/CSVFileOptions'
                excel: '#/components/schemas/ExcelFileOptions'
                parquet: '#/components/schemas/ParquetFileOptions'
              propertyName: file_type
            oneOf:
            - $ref: '#/components/schemas/CSVFileOptions'
            - $ref: '#/components/schemas/ExcelFileOptions'
            - $ref: '#/components/schemas/ParquetFileOptions'
          - type: 'null'
          description: File format options for file2 (file_type, delimiter, sheet, skip rows).
          title: File2 Options
        filter1:
          anyOf:
          - type: string
          - type: 'null'
          description: SQL WHERE clause for Dataset A (omit the WHERE keyword), e.g. 'status = 1'.
          title: Filter1
        filter2:
          anyOf:
          - type: string
          - type: 'null'
          description: SQL WHERE clause for Dataset B (omit the WHERE keyword), e.g. 'status = 1'.
          title: Filter2
        finished_at:
          anyOf:
          - format: date-time
            type: string
          - type: 'null'
          title: Finished At
        id:
          anyOf:
          - type: integer
          - type: 'null'
          title: Id
        include_columns:
          anyOf:
          - items:
              type: string
            type: array
          - type: 'null'
          description: Explicit list of columns to compare. If set, only these columns are diffed.
          title: Include Columns
        infer_pk:
          default: false
          description: 'Infer primary key columns automatically before running the diff. When true, pk_columns may be omitted; the inferred PK is stored back on the diff. Supported configurations: in-db diffs (all input types); cross-db diffs with query inputs on both sides (PK is inferred inside DuckDB after fetch). Not supported: cross-db with table inputs, or any file-based diff — use POST /api/v1/data_sources/{id}/guess_pk and pass pk_columns explicitly. For in-db query inputs, materialization is auto-forced.'
          title: Infer Pk
          type: boolean
        infer_pk_avoid_names:
          anyOf:
          - items:
              type: string
            type: array
          - type: 'null'
          description: Column names to skip during PK inference. Matched case-insensitively against actual column names (exact match, no glob/regex). Requires infer_pk=true and pk_columns empty; rejected with 422 otherwise.
          title: Infer Pk Avoid Names
        infer_pk_avoid_types:
          anyOf:
          - items:
              type: string
            type: array
          - type: 'null'
          description: Extra column types to skip during PK inference, on top of the built-in defaults (float, timestamp, array, boolean, binary, unsupported). Use canonical type names (e.g. 'integer', 'text') or 'db:<RAW_TYPE>' to target a raw warehouse type (e.g. 'db:JSONB'). Requires infer_pk=true and pk_columns empty; rejected with 422 otherwise.
          title: Infer Pk Avoid Types
        kind:
          $ref: '#/components/schemas/DiffKind'
        materialization_destination_id:
          anyOf:
          - type: integer
          - type: 'null'
          description: Data source ID where materialized diff results are stored.
          title: Materialization Destination Id
        materialize_dataset1:
          anyOf:
          - type: boolean
          - type: 'null'
          description: Materialize Dataset A before diffing. Improves speed for heavy queries, filtered non-indexed columns, or transformed primary keys.
          title: Materialize Dataset1
        materialize_dataset2:
          anyOf:
          - type: boolean
          - type: 'null'
          description: Materialize Dataset B before diffing. Same use cases as materialize_dataset1.
          title: Materialize Dataset2
        materialize_without_sampling:
          anyOf:
          - type: boolean
          - type: 'null'
          default: false
          description: Skip sampling when materializing results.
          title: Materialize Without Sampling
        monitor_error:
          anyOf:
          - $ref: '#/components/schemas/QueryError'
          - type: 'null'
        monitor_id:
          anyOf:
          - type: integer
          - type: 'null'
          title: Monitor Id
        monitor_state:
          anyOf:
          - $ref: '#/components/schemas/MonitorRunState'
          - type: 'null'
        per_column_diff_limit:
          anyOf:
          - type: integer
          - type: 'null'
          title: Per Column Diff Limit
        pk_columns:
          items:
            type: string
          title: Pk Columns
          type: array
        purged:
          default: false
          title: Purged
          type: boolean
        query1:
          anyOf:
          - type: string
          - type: 'null'
          description: SQL query for Dataset A. Mutually exclusive with table1 and file1.
          title: Query1
        query2:
          anyOf:
          - type: string
          - type: 'null'
          description: SQL query for Dataset B. Mutually exclusive with table2 and file2.
          title: Query2
        result:
          anyOf:
          - enum:
            - error
            - bad-pks
            - different
            - missing-pks
            - identical
            - empty
            type: string
          - type: 'null'
          title: Result
        result_revisions:
          additionalProperties:
            type: integer
          default: {}
          title: Result Revisions
          type: object
        result_statuses:
          anyOf:
          - additionalProperties:
              type: string
            type: object
          - type: 'null'
          title: Result Statuses
        run_profiles:
          anyOf:
          - type: boolean
          - type: 'null'
          description: Run column profiling on diff results.
          title: Run Profiles
        runtime:
          anyOf:
          - type: number
          - type: 'null'
          title: Runtime
        sampling_confidence:
          anyOf:
          - type: number
          - type: 'null'
          description: 'Sampling confidence level, between 0 and 100 exclusive. Common values: 90, 95, 99, 99.5, 99.9. Use with sampling_tolerance.'
          title: Sampling Confidence
        sampling_max_rows:
          anyOf:
          - type: integer
          - type: 'null'
          description: Maximum number of rows to sample (absolute count). Alternative to tolerance+confidence and sampling_ratio.
          title: Sampling Max Rows
        sampling_ratio:
          anyOf:
          - type: number
          - type: 'null'
          description: Sample this fraction of rows. Value between 0 and 1 exclusive (e.g. 0.1 = 10% of rows). Alternative to tolerance+confidence.
          title: Sampling Ratio
        sampling_threshold:
          anyOf:
          - type: integer
          - type: 'null'
          description: Minimum row count to activate sampling. Sampling is disabled if the largest table has fewer rows than this.
          title: Sampling Threshold
        sampling_tolerance:
          anyOf:
          - type: number
          - type: 'null'
          description: 'Sampling tolerance: max fraction of rows with PK errors before sampling is disabled. Value between 0 and 1 exclusive (e.g. 0.001 = 0.1%). Use with sampling_confidence.'
          title: Sampling Tolerance
        share_token:
          anyOf:
          - type: string
          - type: 'null'
          title: Share Token
        share_url:
          anyOf:
          - type: string
          - type: 'null'
          title: Share Url
        sort_array_columns:
          anyOf:
          - type: boolean
          - type: 'null'
          description: Sort array elements before comparison to ignore insertion order differences.
          title: Sort Array Columns
        source:
          anyOf:
          - $ref: '#/components/schemas/JobSource'
          - type: 'null'
        status:
          anyOf:
          - $ref: '#/components/schemas/JobStatus'
          - type: 'null'
        table1:
          anyOf:
          - items:
              type: string
            type: array
          - type: 'null'
          description: Table path for Dataset A as a list of path components, e.g. ['schema', 'table'] or ['database', 'schema', 'table']. Mutually exclusive with query1 and file1.
          title: Table1
        table2:
          anyOf:
          - items:
              type: string
            type: array
          - type: 'null'
          description: Table path for Dataset B as a list of path components, e.g. ['schema', 'table'] or ['database', 'schema', 'table']. Mutually exclusive with query2 and file2.
          title: Table2
        table_modifiers:
          anyOf:
          - items:
              $ref: '#/components/schemas/TableModifiers'
            type: array
          - type: 'null'
          description: 'Table-level modifiers. Allowed values: ''case_insensitive_strings'' (ignore string case), ''null_equals_empty_string'' (treat NULL and empty string as equal when one warehouse uses NULL and another uses '''' for missing text).'
          title: Table Modifiers
        tags:
          anyOf:
          - items:
              type: string
            type: array
          - type: 'null'
          description: Tags for organizing and filtering diffs.
          title: Tags
        temp_schema_override:
          anyOf:
          - items:
              type: string
            type: array
          - type: 'null'
          title: Temp Schema Override
        time_aggregate:
          anyOf:
          - $ref: '#/components/schemas/TimeAggregateEnum'
          - type: 'null'
          description: Time aggregation level when using time_column.
        time_column:
          anyOf:
          - type: string
          - type: 'null'
          description: Column name used for time-based filtering or aggregation.
          title: Time Column
        time_interval_end:
          anyOf:
          - format: date-time
            type: string
          - type: 'null'
          title: Time Interval End
        time_interval_start:
          anyOf:
          - format: date-time
            type: string
          - type: 'null'
          title: Time Interval Start
        time_travel_point1:
          anyOf:
          - type: integer
          - format: date-time
            type: string
          - type: string
          - type: 'null'
          description: 'Time travel point for Dataset A. Accepts: negative integer offset (e.g. -130), UTC timestamp (e.g. ''2024-01-15T00:00:00''), or a time point hash. Only supported by Snowflake and Databricks.'
          title: Time Travel Point1
        time_travel_point2:
          anyOf:
          - type: integer
          - format: date-time
            type: string
          - type: string
          - type: 'null'
          description: Time travel point for Dataset B. Same format as time_travel_point1.
          title: Time Travel Point2
        tolerance_mode:
          anyOf:
          - $ref: '#/components/schemas/ToleranceModeEnum'
          - type: 'null'
          description: 'How diff_tolerance is applied: ''absolute'' or ''relative''.'
        updated_at:
          anyOf:
          - format: date-time
            type: string
          - type: 'null'
          title: Updated At
        user_id:
          anyOf:
          - type: integer
          - type: 'null'
          title: User Id
      required:
      - data_source1_id
      - data_source2_id
      - pk_columns
      - kind
      title: ApiDataDiffFull
      type: object
    ParquetFileOptions:
      properties:
        file_type:
          const: parquet
          default: parquet
          title: File Type
          type: string
      title: ParquetFileOptions
      type: object
    JobSource:
      enum:
      - interactive
      - demo_signup
      - manual
      - api
      - ci
      - schedule
      - auto
      title: JobSource
      type: string
    CSVFileOptions:
      properties:
        delimiter:
          anyOf:
          - type: string
          - type: 'null'
          title: Delimiter
        file_type:
          const: csv
          default: csv
          title: File Type
          type: string
        skip_head_rows:
          anyOf:
          - type: integer
          - type: 'null'
          title: Skip Head Rows
        skip_tail_rows:
          anyOf:
          - type: integer
          - type: 'null'
          title: Skip Tail Rows
      title: CSVFileOptions
      type: object
    ApiColumnRule:
      properties:
        column:
          anyOf:
          - minLength: 1
            type: string
          - type: 'null'
          description: Column name the rule applies to. When column remapping is enabled, this refers to the dataset-A (canonical) column name — the same name the comparator operates on after remapping. Mutually exclusive with `type`. Exactly one of `column` or `type` must be set.
          title: Column
        equal_if:
          description: SQL boolean expression OR-ed into the value comparator for the matching column(s). Use {a} and {b} as placeholders for the side-A and side-B column references. Multiple rules matching the same column are all OR-ed together.
          title: Equal If
          type: string
        type:
          anyOf:
          - $ref: '#/components/schemas/DatatypeName'
          - type: 'null'
          description: Type-based selector — matches all columns whose Datatype.typename equals this value. The rule is applied to every column of that type after data lands in DuckDB post-fetch. Mutually exclusive with `column`.
      required:
      - equal_if
      title: ApiColumnRule
      type: object
    ExcelFileOptions:
      properties:
        file_type:
          const: excel
          default: excel
          title: File Type
          type: string
        sheet:
          anyOf:
          - type: string
          - type: 'null'
          title: Sheet
        skip_head_rows:
          anyOf:
          - type: integer
          - type: 'null'
          title: Skip Head Rows
        skip_tail_rows:
          anyOf:
          - type: integer
          - type: 'null'
          title: Skip Tail Rows
      title: ExcelFileOptions
      type: object
    ToleranceModeEnum:
      enum:
      - absolute
      - relative
      title: ToleranceModeEnum
      type: string
    DiffAlgorithm:
      enum:
      - join
      - hash
      - hash_v2_alpha
      - fetch_and_join
      title: DiffAlgorithm
      type: string
    TimeAggregateEnum:
      enum:
      - minute
      - hour
      - day
      - week
      - month
      - year
      title: TimeAggregateEnum
      type: string
    MonitorRunState:
      enum:
      - ok
      - alert
      - error
      - learning
      - checking
      - created
      - skipped
      - cancelled
      - queued
      title: MonitorRunState
      type: string
    QueryError:
      properties:
        error_type:
          title: Error Type
          type: string
        error_value:
          title: Error Value
          type: string
      required:
      - error_type
      - error_value
      title: QueryError
      type: object
    DatatypeName:
      enum:
      - boolean
      - integer
      - float
      - decimal
      - unbounded_decimal
      - money
      - string
      - fixed_length_string
      - varbinary
      - binary
      - date
      - super
      - time
      - time_interval
      - timestamp
      - year
      - uuid
      - array
      - struct
      - json
      - xml
      - map
      - void
      - unsupported
      title: DatatypeName
      type: string
    JobStatus:
      enum:
      - needs_confirmation
      - needs_authentication
      - waiting
      - queued
      - processing
      - done
      - failed
      - cancelled
      title: JobStatus
      type: string
    DiffStats:
      properties:
        diff_duplicate_pks:
          anyOf:
          - type: number
          - type: 'null'
          title: Diff Duplicate Pks
        diff_null_pks:
          anyOf:
          - type: number
          - type: 'null'
          title: Diff Null Pks
        diff_pks:
          anyOf:
          - type: number
          - type: 'null'
          title: Diff Pks
        diff_rows:
          anyOf:
          - type: number
          - type: 'null'
          title: Diff Rows
        diff_rows_count:
          anyOf:
          - type: integer
          - type: 'null'
          title: Diff Rows Count
        diff_rows_number:
          anyOf:
          - type: number
          - type: 'null'
          title: Diff Rows Number
        diff_schema:
          anyOf:
          - type: number
          - type: 'null'
          title: Diff Schema
        diff_values:
          anyOf:
          - type: number
          - type: 'null'
          title: Diff Values
        errors:
          anyOf:
          - type: integer
          - type: 'null'
          title: Errors
        match_ratio:
          anyOf:
          - type: number
          - type: 'null'
          title: Match Ratio
        rows_added:
          anyOf:
          - type: integer
          - type: 'null'
          title: Rows Added
        rows_removed:
          anyOf:
          - type: integer
          - type: 'null'
          title: Rows Removed
        sampled:
          anyOf:
          - type: boolean
          - type: 'null'
          title: Sampled
        table_a_row_count:
          anyOf:
          - type: integer
          - type: 'null'
          title: Table A Row Count
        table_b_row_count:
          anyOf:
          - type: integer
          - type: 'null'
          title: Table B Row Count
        version:
          title: Version
          type: string
      required:
      - version
      title: DiffStats
      type: object
    ColumnTolerance:
      properties:
        column_name:
          description: Column name to apply tolerance to.
          title: Column Name
          type: string
        tolerance_mode:
          $ref: '#/components/schemas/ToleranceModeEnum'
          description: '''absolute'' or ''relative''.'
        tolerance_value:
          description: Tolerance value. Must be >= 0.
          title: Tolerance Value
          type: number
      required:
      - column_name
      - tolerance_value
      - tolerance_mode
      title: ColumnTolerance
      type: object
    ApiDataDiffData:
      properties:
        archived:
          default: false
          title: Archived
          type: boolean
        bisection_factor:
          anyOf:
          - type: integer
          - type: 'null'
          title: Bisection Factor
        bisection_threshold:
          anyOf:
          - type: integer
          - type: 'null'
          title: Bisection Threshold
        column_mapping:
          anyOf:
          - items:
              maxItems: 2
              minItems: 2
              prefixItems:
              - type: string
              - type: string
              type: array
            type: array
          - type: 'null'
          description: Map columns with different names between datasets. List of [column_in_A, column_in_B] pairs.
          title: Column Mapping
        column_rules:
          anyOf:
          - items:
              $ref: '#/components/schemas/ApiColumnRule'
            type: array
          - type: 'null'
          description: Extra equality clauses that extend the in-memory diff value comparator. Useful for treating specific value pairs (e.g. legacy sentinels vs NULL) as non-differences. See `ApiColumnRule` for entry shape.
          title: Column Rules
        columns_to_compare:
          anyOf:
          - items:
              type: string
            type: array
          - type: 'null'
          description: Columns to compare between datasets. If set, only these columns are diffed (primary key columns are always included). Column names must match the dataset schema.
          title: Columns To Compare
        compare_duplicates:
          anyOf:
          - type: boolean
          - type: 'null'
          description: Compare rows with duplicate primary keys. Defaults to true.
          title: Compare Duplicates
        data_source1_id:
          description: ID of the first data source (Dataset A).
          title: Data Source1 Id
          type: integer
        data_source1_session_parameters:
          anyOf:
          - additionalProperties: true
            type: object
          - type: 'null'
          description: 'Snowflake session parameters for Dataset A, e.g. {"QUERY_TAG": "datadiff", "WAREHOUSE": "COMPUTE_WH"}.'
          title: Data Source1 Session Parameters
        data_source2_id:
          description: ID of the second data source (Dataset B). Can be the same as data_source1_id.
          title: Data 

# --- truncated at 32 KB (44 KB total) ---
# Full source: https://raw.githubusercontent.com/api-evangelist/datafold/refs/heads/main/openapi/datafold-diff-created-api-openapi.yml