feat(ingestion/montecarlo): add Monte Carlo connector

datahub-web-react/src/app/entityV2/shared/tabs/Dataset/Validations/CustomAssertionDescription.tsx

@@ -0,0 +1,51 @@
+import { Button, Typography } from 'antd';
+import React, { useState } from 'react';
+import { useTranslation } from 'react-i18next';
+import styled from 'styled-components';
+
+import { DatasetAssertionLogicModal } from '@app/entityV2/shared/tabs/Dataset/Validations/DatasetAssertionLogicModal';
+
+import { AssertionInfo } from '@types';
+
+const ViewLogicButton = styled(Button)`
+    padding: 0px;
+    margin: 0px;
+`;
+
+type Props = {
+    assertionInfo: AssertionInfo;
+};
+
+/**
+ * A human-readable description of a Custom Assertion, surfacing its logic
+ * (e.g. the SQL behind a Monte Carlo custom-SQL rule) behind a "View Logic" link.
+ */
+export const CustomAssertionDescription = ({ assertionInfo }: Props) => {
+    const { t } = useTranslation('entity.profile.validations');
+    const [isLogicVisible, setIsLogicVisible] = useState(false);
+
+    const { description } = assertionInfo;
+    const logic = assertionInfo.customAssertion?.logic;
+
+    return (
+        <>
+            <Typography.Text>{description}</Typography.Text>
+            {logic && (
+                <>
+                    <div>
+                        <ViewLogicButton onClick={() => setIsLogicVisible(true)} type="link">
+                            {t('datasetDescription.popover.viewLogic')}
+                        </ViewLogicButton>
+                    </div>
+                    <DatasetAssertionLogicModal
+                        logic={logic}
+                        title={description || undefined}
+                        description={assertionInfo.customAssertion?.type || undefined}
+                        visible={isLogicVisible}
+                        onClose={() => setIsLogicVisible(false)}
+                    />
+                </>
+            )}
+        </>
+    );
+};

datahub-web-react/src/app/entityV2/shared/tabs/Dataset/Validations/DatasetAssertionLogicModal.tsx

@@ -6,15 +6,17 @@ import Query from '@app/entityV2/shared/tabs/Dataset/Queries/Query';
 
 type Props = {
     logic: string;
+    title?: string;
+    description?: string;
     visible: boolean;
     onClose: () => void;
 };
 
-export const DatasetAssertionLogicModal = ({ logic, visible, onClose }: Props) => {
+export const DatasetAssertionLogicModal = ({ logic, title, description, visible, onClose }: Props) => {
     const { t: tc } = useTranslation('common.actions');
     return (
         <Modal visible={visible} onCancel={onClose} footer={<Button onClick={onClose}>{tc('close')}</Button>}>
-            <Query query={logic} />
+            <Query query={logic} title={title} description={description} />
         </Modal>
     );
 };

datahub-web-react/src/app/entityV2/shared/tabs/Dataset/Validations/assertion/profile/summary/utils.tsx

@@ -5,6 +5,7 @@ import React from 'react';
 import { useTranslation } from 'react-i18next';
 import { useTheme } from 'styled-components';
 
+import { CustomAssertionDescription } from '@app/entityV2/shared/tabs/Dataset/Validations/CustomAssertionDescription';
 import { DatasetAssertionDescription } from '@app/entityV2/shared/tabs/Dataset/Validations/DatasetAssertionDescription';
 import { FieldAssertionDescription } from '@app/entityV2/shared/tabs/Dataset/Validations/FieldAssertionDescription';
 import {
@@ -365,7 +366,9 @@ export const useBuildAssertionPrimaryLabel = (
 ): JSX.Element => {
     const { t } = useTranslation('entity.profile.validations');
     let primaryLabel = <Typography.Text>{t('datasetDescription.fallback.noDescription')}</Typography.Text>;
-    if (assertionInfo?.description && assertionInfo?.type !== AssertionType.Field) {
+    if (assertionInfo?.type === AssertionType.Custom && assertionInfo?.customAssertion?.logic) {
+        primaryLabel = <CustomAssertionDescription assertionInfo={assertionInfo} />;
+    } else if (assertionInfo?.description && assertionInfo?.type !== AssertionType.Field) {
         primaryLabel = <Typography.Text>{assertionInfo.description}</Typography.Text>;
     } else {
         switch (assertionInfo?.type) {

datahub-web-react/src/app/ingest/source/builder/sources.json

@@ -1,4 +1,13 @@
 [
+    {
+        "urn": "urn:li:dataPlatform:montecarlo",
+        "name": "montecarlo",
+        "displayName": "Monte Carlo",
+        "description": "Import monitors, custom rules, and alerts from Monte Carlo as DataHub assertions.",
+        "docsUrl": "https://docs.datahub.com/docs/generated/ingestion/sources/montecarlo",
+        "recipe": "source:\n  type: montecarlo\n  config:\n    api_id: ${MCD_ID}\n    api_token: ${MCD_TOKEN}\n    connection_to_platform_map:\n      \"<mc-warehouse-uuid>\":\n        platform: snowflake\n        platform_instance: prod\n        env: PROD\n    emit_assertions: true\n    emit_alerts: true\n    alerts_lookback_days: 30\n    stateful_ingestion:\n      enabled: true",
+        "category": "Observability"
+    },
     {
         "urn": "urn:li:dataPlatform:aerospike",
         "name": "aerospike",

datahub-web-react/src/app/ingestV2/source/builder/constants.ts

@@ -37,6 +37,7 @@ import metabaseLogo from '@images/metabaselogo.png';
 import mlflowLogo2 from '@images/mlflowlogo2.png';
 import modeLogo from '@images/modelogo.png';
 import mongodbLogo from '@images/mongodblogo.png';
+import montecarloLogo from '@images/montecarlologo.png';
 import mssqlLogo from '@images/mssqllogo.png';
 import mysqlLogo from '@images/mysqllogo-2.png';
 import neo4j from '@images/neo4j.svg';
@@ -203,6 +204,8 @@ export const VERTEX_AI = 'vertexai';
 export const VERTEXAI_URN = `urn:li:dataPlatform:${VERTEX_AI}`;
 export const INFORMATICA = 'informatica';
 export const INFORMATICA_URN = `urn:li:dataPlatform:${INFORMATICA}`;
+export const MONTECARLO = 'montecarlo';
+export const MONTECARLO_URN = `urn:li:dataPlatform:${MONTECARLO}`;
 export const SNAPLOGIC = 'snaplogic';
 export const SNAPLOGIC_URN = `urn:li:dataPlatform:${SNAPLOGIC}`;
 export const DLT = 'dlt';
@@ -282,6 +285,7 @@ export const PLATFORM_URN_TO_LOGO = {
     [NEO4J_URN]: neo4j,
     [VERTEXAI_URN]: vertexAI,
     [INFORMATICA_URN]: informaticaLogo,
+    [MONTECARLO_URN]: montecarloLogo,
     [SNAPLOGIC_URN]: snaplogicLogo,
     [FABRIC_URN]: fabricLogo,
     [FABRIC_DATA_FACTORY_URN]: fabricDataFactoryLogo,

datahub-web-react/src/app/ingestV2/source/builder/sources.json

@@ -1,4 +1,14 @@
 [
+    {
+        "urn": "urn:li:dataPlatform:montecarlo",
+        "name": "montecarlo",
+        "displayName": "Monte Carlo",
+        "description": "Import monitors, custom rules, and alerts from Monte Carlo as DataHub assertions.",
+        "docsUrl": "https://docs.datahub.com/docs/generated/ingestion/sources/montecarlo",
+        "recipe": "source:\n  type: montecarlo\n  config:\n    api_id: ${MCD_ID}\n    api_token: ${MCD_TOKEN}\n    connection_to_platform_map:\n      \"<mc-warehouse-uuid>\":\n        platform: snowflake\n        platform_instance: prod\n        env: PROD\n    emit_assertions: true\n    emit_alerts: true\n    alerts_lookback_days: 30\n    stateful_ingestion:\n      enabled: true",
+        "category": "Observability",
+        "isPopular": false
+    },
     {
         "urn": "urn:li:dataPlatform:bigquery",
         "name": "bigquery",

metadata-ingestion/constraints.txt

@@ -343,6 +343,7 @@ databricks-sql-connector==2.9.6
     # via acryl-datahub
 dataclasses-json==0.6.7
     # via
+    #   pycarlo
     #   unstructured
     #   unstructured-ingest
 dataflows-tabulator==1.54.3
@@ -554,6 +555,7 @@ graphql-core==3.2.8
     # via
     #   acryl-datahub
     #   gql
+    #   sgqlc
 greenlet==3.3.2
     # via
     #   acryl-datahub
@@ -1179,6 +1181,8 @@ pyasn1-modules==0.4.2
     #   python-ldap
 pyathena==2.25.2
     # via acryl-datahub
+pycarlo==0.12.457
+    # via acryl-datahub
 pycountry==26.2.16
     # via schwifty
 pycparser==3.0 ; implementation_name != 'PyPy'
@@ -1291,6 +1295,8 @@ pytest-rerunfailures==16.1
     # via acryl-datahub
 pytest-timeout==2.4.0
     # via acryl-datahub
+python-box==7.4.1
+    # via pycarlo
 python-dateutil==2.9.0.post0
     # via
     #   acryl-datahub
@@ -1430,6 +1436,7 @@ requests==2.32.5
     #   mlflow-skinny
     #   moto
     #   msal
+    #   pycarlo
     #   pydruid
     #   pyiceberg
     #   redash-toolbelt
@@ -1478,6 +1485,7 @@ responses==0.26.0
     # via
     #   acryl-datahub
     #   moto
+    #   pycarlo
 rfc3339-validator==0.1.4
     # via
     #   jsonschema
@@ -1551,6 +1559,8 @@ setuptools==81.0.0
     #   redshift-connector
     #   spacy
     #   thinc
+sgqlc==16.5
+    # via pycarlo
 shellingham==1.5.4
     # via typer
 simple-salesforce==1.12.9

metadata-ingestion/docs/sources/montecarlo/README.md

@@ -0,0 +1,17 @@
+## Overview
+
+[Monte Carlo](https://www.montecarlodata.com/) is a data observability platform that monitors warehouse and lake tables for freshness, volume, schema and field-quality issues and raises alerts/incidents when they breach.
+
+This connector ingests Monte Carlo **monitors**, **custom (SQL) rules** and **alerts/incidents** and models them as DataHub **Assertions**, so the native "Validation" tab on a dataset reflects Monte Carlo's observability coverage and incident history.
+
+## Concept Mapping
+
+| Monte Carlo Concept    | DataHub Concept                                                                           | Notes                                                                     |
+| ---------------------- | ----------------------------------------------------------------------------------------- | ------------------------------------------------------------------------- |
+| `"montecarlo"`         | [Data Platform](https://docs.datahub.com/docs/generated/metamodel/entities/dataplatform/) |                                                                           |
+| Monitor                | [Assertion](https://docs.datahub.com/docs/generated/metamodel/entities/assertion/)        | One `CUSTOM` assertion per monitor; native type kept in custom props.     |
+| Custom (SQL) rule      | [Assertion](https://docs.datahub.com/docs/generated/metamodel/entities/assertion/)        | One `CUSTOM` assertion per rule; SQL captured in `customAssertion.logic`. |
+| Monitored asset (MCON) | [Dataset](https://docs.datahub.com/docs/generated/metamodel/entities/dataset/)            | Resolved via `getTable` and `connection_to_platform_map`.                 |
+| Alert / Incident       | Assertion Run Event                                                                       | Emitted as an `AssertionRunEvent` failure on the corresponding assertion. |
+
+Every monitor/rule is modeled as a `CUSTOM` assertion (matching the established connector pattern, e.g. Snowflake DMFs), with the Monte Carlo native type, resource id and data-quality dimension round-tripped via `customProperties`.

metadata-ingestion/docs/sources/montecarlo/montecarlo_post.md

@@ -0,0 +1,54 @@
+### Capabilities
+
+Use the **Important Capabilities** table above as the source of truth for supported features.
+
+#### Assertion modeling
+
+Every Monte Carlo monitor and custom SQL rule is modeled as a `CUSTOM` assertion. The Monte Carlo
+native type (e.g. `freshness`, `volume`, `field_quality`, `custom_sql`) is preserved in
+`customProperties` alongside the resource identifier and data-quality dimension, so no information
+is lost. For custom SQL rules, the raw SQL expression is captured in `customAssertion.logic`.
+
+#### Alert and incident ingestion
+
+Monte Carlo alerts and incidents are ingested as `AssertionRunEvent` failures on their
+corresponding assertion. Each event carries a timestamp, the Monte Carlo alert ID, and a link back
+to the Monte Carlo UI via `externalUrl`.
+
+### Limitations
+
+- **Failures only:** Monte Carlo's API does not expose a per-run "pass" stream, so the connector
+  emits only `FAILURE` run events (from alerts/incidents). Periodic `SUCCESS` events are not
+  synthesized.
+- **MCON resolution:** Each monitored asset requires one `getTable` call to resolve its MCON to a
+  warehouse table (results are cached per MCON). Assets whose warehouse connection type is not in
+  `connection_to_platform_map` (and not auto-mappable) are skipped with a warning.
+- **Assertion typing:** All monitors and rules are modeled as `CUSTOM` assertions. The Monte Carlo
+  native type is preserved in `customProperties` rather than coerced into DataHub's typed
+  freshness/volume/SQL/field assertion schemas.
+- **Monte Carlo Cloud only:** Requires a Monte Carlo Cloud account and API key pair. Self-hosted
+  deployments are not supported.
+
+### Troubleshooting
+
+#### Monitored assets are skipped with a warning
+
+If you see warnings like `Could not resolve MCON to a DataHub dataset URN`, the warehouse
+connection type for that asset is not covered by `connection_to_platform_map`. Add a mapping entry
+for the connection name shown in the warning.
+
+#### Assertion URNs do not match your warehouse source
+
+Assertion URNs are derived from the dataset URN resolved via `connection_to_platform_map`. If the
+`platform`, `platform_instance`, or `env` values differ from those used by your warehouse source
+connector, the assertions will not appear on the correct dataset. Align the values in
+`connection_to_platform_map` with the config of your warehouse source.
+
+#### No assertions appear after ingestion
+
+Verify that:
+
+1. The API key has read access to monitors, custom rules, and alerts in the Monte Carlo UI.
+2. At least one monitor is active and has fired an alert (the connector ingests only monitors that
+   have associated assets and alerts).
+3. The `connection_to_platform_map` covers the warehouse connections used by your monitored assets.

metadata-ingestion/docs/sources/montecarlo/montecarlo_pre.md

@@ -0,0 +1,28 @@
+### Overview
+
+[Monte Carlo](https://www.montecarlodata.com/) is a data observability platform that monitors
+warehouse and lake tables for freshness, volume, schema and field-quality issues and raises
+alerts/incidents when they breach.
+
+This connector ingests Monte Carlo **monitors**, **custom (SQL) rules** and **alerts/incidents** and
+models them as DataHub **Assertions**, so the native "Validation" tab on a dataset reflects Monte
+Carlo's observability coverage and incident history.
+
+### Prerequisites
+
+In order to ingest metadata from Monte Carlo, you will need:
+
+- A Monte Carlo Cloud account (this connector does not support self-hosted/on-prem variants).
+- An API key pair (`mcd_id` + `mcd_token`) with read access to monitors, custom rules, alerts and
+  the catalog. Create one in the Monte Carlo UI under **Settings → API** (see the
+  [Monte Carlo API docs](https://docs.getmontecarlo.com/docs/using-the-api)).
+- A `connection_to_platform_map` entry for each Monte Carlo warehouse you want ingested, so
+  monitored-asset URNs align with the URNs emitted by your warehouse sources.
+
+#### Cross-platform URN mapping
+
+A Monte Carlo MCON does not encode the DataHub platform. The connector resolves each MCON to a
+concrete table via `getTable` and maps the warehouse connection type to a DataHub platform. Use
+`connection_to_platform_map` to pin the `platform`, `platform_instance` and `env` for each Monte
+Carlo warehouse so the resulting dataset URNs line up with the URNs emitted by your warehouse
+sources (Snowflake, BigQuery, etc.).

metadata-ingestion/docs/sources/montecarlo/montecarlo_recipe.yml

@@ -0,0 +1,24 @@
+source:
+  type: montecarlo
+  config:
+    api_id: ${MCD_ID} # Monte Carlo API key id (mcd_id)
+    api_token: ${MCD_TOKEN} # Monte Carlo API key token (mcd_token)
+
+    # Map each Monte Carlo warehouse/resource UUID to a DataHub platform so the
+    # monitored-asset URNs match those emitted by your warehouse sources.
+    connection_to_platform_map:
+      "<mc-warehouse-uuid>":
+        platform: snowflake
+        platform_instance: prod
+        env: PROD
+
+    emit_assertions: true # monitors + custom rules -> Assertion entities
+    alerts:
+      enabled: true # alerts/incidents -> AssertionRunEvent failures
+      start_time: "-30 days" # how far back to fetch alerts (default: 30 days)
+
+    # stateful_ingestion:
+    #   enabled: true
+
+sink:
+  # sink configs