feat(schema-compiler): Custom granularities support for Amazon Athena/Presto dialect (cube-js#9472)

* Add docs

* code hygiene

* feat(schema-compiler): Custom granularities support for AWS Athena/Presto dialect

---------

Co-authored-by: Konstantin Burkalev <[email protected]>

Author: 2 people

docs/pages/reference/data-model/dimensions.mdx

@@ -618,6 +618,14 @@ examples.
 
 </ReferenceBox>
 
+<WarningBox>
+
+Custom granularities are supported for the following [data sources][ref-data-sources]:
+Amazon Athena, Amazon Redshift, DuckDB, Databricks, Google BigQuery, ClickHouse, Microsoft SQL Server, MySQL, Postgres, and Snowflake.
+Please [file an issue](https://github.com/cube-js/cube/issues) if you need support for another data source.
+
+</WarningBox>
+
 For each custom granularity, the `interval` parameter is required. It specifies
 the duration of the time interval and has the following format:
 `quantity unit [quantity unit...]`, e.g., `5 days` or `1 year 6 months`.
@@ -713,4 +721,5 @@ cube(`orders`, {
 [ref-time-dimensions]: /reference/data-model/types-and-formats#time-1
 [link-date-time-string]: https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/Date#date_time_string_format
 [ref-custom-granularity-recipe]: /guides/recipes/data-modeling/custom-granularity
-[ref-ref-hierarchies]: /reference/data-model/hierarchies
+[ref-ref-hierarchies]: /reference/data-model/hierarchies
+[ref-data-sources]: /product/configuration/data-sources

packages/cubejs-schema-compiler/src/adapter/BaseQuery.js

@@ -3144,7 +3144,7 @@ export class BaseQuery {
    */
   // eslint-disable-next-line @typescript-eslint/no-unused-vars
   dateBin(interval, source, origin) {
-    throw new Error('Date bin function is not implemented');
+    throw new Error('Date bin function, required for custom time dimension granularities, is not implemented for this data source');
     // Different syntax possible in different DBs
   }
 

packages/cubejs-schema-compiler/src/adapter/PreAggregations.js

@@ -157,7 +157,7 @@ export class PreAggregations {
     const tableName = this.preAggregationTableName(cube, preAggregationName, preAggregation);
     const invalidateKeyQueries = this.query.preAggregationInvalidateKeyQueries(cube, preAggregation, preAggregationName);
     const queryForSqlEvaluation = this.query.preAggregationQueryForSqlEvaluation(cube, preAggregation);
-    const partitionInvalidateKeyQueries = queryForSqlEvaluation.partitionInvalidateKeyQueries && queryForSqlEvaluation.partitionInvalidateKeyQueries(cube, preAggregation);
+    const partitionInvalidateKeyQueries = queryForSqlEvaluation.partitionInvalidateKeyQueries?.(cube, preAggregation);
 
     const allBackAliasMembers = this.query.allBackAliasMembers();
 

packages/cubejs-schema-compiler/src/adapter/PrestodbQuery.ts

@@ -1,3 +1,4 @@
+import { parseSqlInterval } from '@cubejs-backend/shared';
 import { BaseQuery } from './BaseQuery';
 import { BaseFilter } from './BaseFilter';
 
@@ -55,6 +56,30 @@ export class PrestodbQuery extends BaseQuery {
       field;
   }
 
+  /**
+   * Returns sql for source expression floored to timestamps aligned with
+   * intervals relative to origin timestamp point.
+   * Athena doesn't support INTERVALs directly — using date_diff/date_add
+   */
+  public dateBin(interval: string, source: string, origin: string): string {
+    const intervalParsed = parseSqlInterval(interval);
+    const intervalParts = Object.entries(intervalParsed);
+
+    if (intervalParts.length > 1) {
+      throw new Error('Athena/Presto supports only simple intervals with one date part');
+    }
+
+    const [unit, count] = intervalParts[0];
+    const originExpr = this.timeStampCast(`'${origin}'`);
+
+    return `date_add('${unit}',
+      floor(
+        date_diff('${unit}', ${originExpr}, ${source}) / ${count}
+      ) * ${count},
+      ${originExpr}
+    )`;
+  }
+
   public timeGroupedColumn(granularity, dimension) {
     return `date_trunc('${GRANULARITY_TO_INTERVAL[granularity]}', ${dimension})`;
   }

packages/cubejs-server-core/src/core/DriverResolvers.ts

@@ -49,13 +49,13 @@ export const lookupDriverClass = (dbType): Constructor<BaseDriver> & {
  */
 export const isDriver = (val: any): boolean => {
   let isDriverInstance = val instanceof BaseDriver;
-  if (!isDriverInstance && val && val.constructor) {
+  if (!isDriverInstance && val?.constructor) {
     let end = false;
     let obj = val.constructor;
     while (!isDriverInstance && !end) {
       obj = Object.getPrototypeOf(obj);
       end = !obj;
-      isDriverInstance = obj && obj.name ? obj.name === 'BaseDriver' : false;
+      isDriverInstance = obj?.name ? obj.name === 'BaseDriver' : false;
     }
   }
   return isDriverInstance;
@@ -82,11 +82,11 @@ export const getDriverMaxPool = async (
     const queryQueueOptions = await options
       .queryCacheOptions
       .queueOptions(context.dataSource);
-  
+
     const preAggregationsQueueOptions = await options
       .preAggregationsOptions
       .queueOptions(context.dataSource);
-  
+
     return 2 * (
       queryQueueOptions.concurrency +
         preAggregationsQueueOptions.concurrency

packages/cubejs-testing-drivers/fixtures/athena.json

@@ -151,15 +151,8 @@
     "---------------------------------------",
     "Custom Granularities                   ",
     "---------------------------------------",
-    "querying custom granularities ECommerce: count by half_year + no dimension",
-    "querying custom granularities ECommerce: count by half_year_by_1st_april + no dimension",
     "querying custom granularities ECommerce: count by three_months_by_march + no dimension",
-    "querying custom granularities ECommerce: count by half_year + dimension",
-    "querying custom granularities ECommerce: count by half_year_by_1st_april + dimension",
     "querying custom granularities ECommerce: count by three_months_by_march + dimension",
-    "querying custom granularities ECommerce: count by two_mo_by_feb + no dimension + rollingCountByUnbounded",
-    "querying custom granularities ECommerce: count by two_mo_by_feb + no dimension + rollingCountByTrailing",
-    "querying custom granularities ECommerce: count by two_mo_by_feb + no dimension + rollingCountByLeading",
 
     "SKIPPED SQL API (Need work)",
     "---------------------------------------",