fraiseql-core 2.16.0

//! Integration tests for aggregation queries.
//!
//! These tests verify the end-to-end aggregation pipeline:
//! - JSON query parsing
//! - Execution plan generation
//! - SQL generation
//! - Result projection
//!
//! To run unit tests:
//!   cargo test -p fraiseql-core --test `aggregation_integration`
//!
//! For database integration tests:
//!   1. Start test database: docker compose -f docker-compose.test.yml up -d
//!   2. Run tests: cargo test -p fraiseql-core --features test-postgres --test
//!      `aggregation_integration`

#![allow(
    clippy::unwrap_used,
    clippy::panic,
    clippy::print_stdout,
    clippy::print_stderr
)] // Reason: test code, panics acceptable
#![allow(clippy::match_wildcard_for_single_variants)] // Reason: test match uses wildcard arm for forward-compatibility
use fraiseql_core::{
    compiler::{
        aggregate_types::{AggregateFunction, TemporalBucket},
        aggregation::{AggregateSelection, AggregationRequest, GroupBySelection},
        fact_table::{
            CalendarBucket, CalendarDimension, CalendarGranularity, DimensionColumn, DimensionPath,
            FactTableMetadata, FilterColumn, MeasureColumn, SqlType,
        },
    },
    runtime::{AggregateQueryParser, AggregationProjector, AggregationSqlGenerator},
};
use serde_json::json;

/// Helper to create test fact table metadata for `tf_sales`
fn create_test_metadata() -> FactTableMetadata {
    FactTableMetadata {
        table_name:               "tf_sales".to_string(),
        type_name:                None,
        measures:                 vec![
            MeasureColumn {
                name:     "revenue".to_string(),
                sql_type: SqlType::Decimal,
                nullable: false,
            },
            MeasureColumn {
                name:     "quantity".to_string(),
                sql_type: SqlType::Int,
                nullable: false,
            },
        ],
        dimensions:               DimensionColumn {
            name:  "data".to_string(),
            paths: vec![
                DimensionPath {
                    name:      "category".to_string(),
                    json_path: "data->>'category'".to_string(),
                    data_type: "text".to_string(),
                },
                DimensionPath {
                    name:      "product".to_string(),
                    json_path: "data->>'product'".to_string(),
                    data_type: "text".to_string(),
                },
            ],
        },
        denormalized_filters:     vec![FilterColumn {
            name:      "occurred_at".to_string(),
            sql_type:  SqlType::Timestamp,
            indexed:   true,
            hierarchy: None,
        }],
        calendar_dimensions:      vec![CalendarDimension {
            source_column: "occurred_at".to_string(),
            granularities: vec![CalendarGranularity {
                column_name: "date_info".to_string(),
                buckets:     vec![
                    CalendarBucket {
                        json_key:    "day".to_string(),
                        bucket_type: TemporalBucket::Day,
                        data_type:   "date".to_string(),
                    },
                    CalendarBucket {
                        json_key:    "month".to_string(),
                        bucket_type: TemporalBucket::Month,
                        data_type:   "integer".to_string(),
                    },
                ],
            }],
        }],
        partial_period:           None,
        native_measures:          std::collections::HashMap::new(),
        native_dimension_mapping: std::collections::HashMap::new(),
    }
}

// ============================================================================
// Query Parsing Tests
// ============================================================================

#[test]
fn test_parse_simple_aggregate_query() {
    let metadata = create_test_metadata();
    let query = json!({
        "table": "tf_sales",
        "aggregates": [
            {"count": {}}
        ]
    });

    let request =
        AggregateQueryParser::parse(&query, &metadata, &std::collections::HashMap::new()).unwrap();

    assert_eq!(request.table_name, "tf_sales");
    assert_eq!(request.aggregates.len(), 1);
    assert_eq!(request.aggregates[0].alias(), "count");
}

#[test]
fn test_parse_group_by_with_aggregates() {
    let metadata = create_test_metadata();
    let query = json!({
        "table": "tf_sales",
        "groupBy": {
            "category": true,
            "occurred_at_day": true
        },
        "aggregates": [
            {"count": {}},
            {"revenue_sum": {}},
            {"revenue_avg": {}}
        ]
    });

    let request =
        AggregateQueryParser::parse(&query, &metadata, &std::collections::HashMap::new()).unwrap();

    assert_eq!(request.group_by.len(), 2);
    assert_eq!(request.aggregates.len(), 3);

    // Verify GROUP BY selections
    match &request.group_by[0] {
        GroupBySelection::Dimension { path, alias } => {
            assert_eq!(path, "category");
            assert_eq!(alias, "category");
        },
        _ => panic!("Expected Dimension selection"),
    }

    // Parser prefers CalendarDimension over TemporalBucket when calendar dimensions are defined
    match &request.group_by[1] {
        GroupBySelection::CalendarDimension {
            source_column,
            calendar_column,
            json_key,
            bucket,
            alias,
        } => {
            assert_eq!(source_column, "occurred_at");
            assert_eq!(calendar_column, "date_info");
            assert_eq!(json_key, "day");
            assert_eq!(*bucket, TemporalBucket::Day);
            assert_eq!(alias, "occurred_at_day");
        },
        GroupBySelection::TemporalBucket {
            column,
            bucket,
            alias,
        } => {
            // Fallback if no calendar dimension (shouldn't happen in this test)
            assert_eq!(column, "occurred_at");
            assert_eq!(*bucket, TemporalBucket::Day);
            assert_eq!(alias, "occurred_at_day");
        },
        _ => panic!("Expected CalendarDimension or TemporalBucket selection"),
    }
}

// ============================================================================
// SQL Generation Tests
// ============================================================================

#[test]
fn test_sql_generation_postgres() {
    use fraiseql_core::{compiler::aggregation::AggregationPlanner, db::types::DatabaseType};

    let metadata = create_test_metadata();
    let request = AggregationRequest {
        table_name:   "tf_sales".to_string(),
        where_clause: None,
        group_by:     vec![GroupBySelection::Dimension {
            path:  "category".to_string(),
            alias: "category".to_string(),
        }],
        aggregates:   vec![
            AggregateSelection::Count {
                alias: "count".to_string(),
            },
            AggregateSelection::MeasureAggregate {
                measure:  "revenue".to_string(),
                function: AggregateFunction::Sum,
                alias:    "revenue_sum".to_string(),
            },
        ],
        having:       vec![],
        order_by:     vec![],
        limit:        None,
        offset:       None,
    };

    // Generate execution plan
    let plan = AggregationPlanner::plan(request, metadata).unwrap();

    // Generate SQL
    let sql_generator = AggregationSqlGenerator::new(DatabaseType::PostgreSQL);
    let sql = sql_generator.generate_parameterized(&plan).unwrap();

    // Verify SQL contains expected clauses
    assert!(sql.sql.contains("data->>'category'"));
    assert!(sql.sql.contains("COUNT(*)"));
    assert!(sql.sql.contains("SUM(revenue)"));
    assert!(sql.sql.contains("GROUP BY"));
    assert!(sql.sql.contains("FROM tf_sales"));
}

#[test]
fn test_result_projection() {
    use std::collections::HashMap;

    use fraiseql_core::compiler::aggregation::{
        AggregateExpression, AggregationPlan, GroupByExpression,
    };

    let metadata = create_test_metadata();
    let request = AggregationRequest {
        table_name:   "tf_sales".to_string(),
        where_clause: None,
        group_by:     vec![GroupBySelection::Dimension {
            path:  "category".to_string(),
            alias: "category".to_string(),
        }],
        aggregates:   vec![AggregateSelection::Count {
            alias: "count".to_string(),
        }],
        having:       vec![],
        order_by:     vec![],
        limit:        None,
        offset:       None,
    };

    let plan = AggregationPlan {
        metadata,
        request,
        group_by_expressions: vec![GroupByExpression::JsonbPath {
            jsonb_column: "data".to_string(),
            path:         "category".to_string(),
            alias:        "category".to_string(),
        }],
        aggregate_expressions: vec![AggregateExpression::Count {
            alias: "count".to_string(),
        }],
        having_conditions: vec![],
    };

    // Mock SQL results
    let rows = vec![
        {
            let mut row = HashMap::new();
            row.insert("category".to_string(), json!("Electronics"));
            row.insert("count".to_string(), json!(42));
            row
        },
        {
            let mut row = HashMap::new();
            row.insert("category".to_string(), json!("Books"));
            row.insert("count".to_string(), json!(15));
            row
        },
    ];

    // Project results
    let projected = AggregationProjector::project(rows, &plan).unwrap();

    // Verify projection
    assert!(projected.is_array());
    let arr = projected.as_array().unwrap();
    assert_eq!(arr.len(), 2);
    assert_eq!(arr[0]["category"], "Electronics");
    assert_eq!(arr[0]["count"], 42);
    assert_eq!(arr[1]["category"], "Books");
    assert_eq!(arr[1]["count"], 15);
}

#[test]
fn test_wrap_in_graphql_envelope() {
    let projected = json!([
        {"category": "Electronics", "count": 42}
    ]);

    let response = AggregationProjector::wrap_in_data_envelope(projected, "sales_aggregate");

    assert!(response.get("data").is_some());
    assert!(response["data"].get("sales_aggregate").is_some());
    assert_eq!(response["data"]["sales_aggregate"][0]["category"], "Electronics");
}

// ============================================================================
// Database Integration Tests
// ============================================================================

#[cfg(feature = "test-postgres")]
#[tokio::test]
async fn test_end_to_end_aggregate_query() {
    use std::sync::Arc;

    use fraiseql_core::{db::postgres::PostgresAdapter, runtime::Executor, schema::CompiledSchema};

    let Some(pg) = fraiseql_test_support::postgres().await else {
        eprintln!("SKIP test_end_to_end_aggregate_query: no postgres (set DATABASE_URL)");
        return;
    };

    // Setup test database (pg is held for the test's lifetime so a locally spawned
    // container, if any, is not torn down mid-test).
    let adapter = Arc::new(
        PostgresAdapter::new(pg.url())
            .await
            .expect("Failed to connect to test database"),
    );

    // Create test schema
    let schema = CompiledSchema::new();
    let executor = Executor::new(schema, adapter);

    // Create test fact table metadata
    let metadata = create_test_metadata();

    // Create aggregate query
    let query_json = json!({
        "table": "tf_sales",
        "groupBy": {
            "category": true
        },
        "aggregates": [
            {"count": {}},
            {"revenue_sum": {}}
        ],
        "limit": 10
    });

    // Execute aggregate query
    let result = executor
        .execute_aggregate_query(&query_json, "sales_aggregate", &metadata)
        .await
        .expect("Failed to execute aggregate query");

    // Parse response
    let response = result;

    // Verify response structure
    assert!(response.get("data").is_some());
    assert!(response["data"].get("sales_aggregate").is_some());
    assert!(response["data"]["sales_aggregate"].is_array());

    println!("Aggregate query result: {}", serde_json::to_string_pretty(&response).unwrap());
}