Struct FactTableDetector 

pub struct FactTableDetector;

Detects and introspects fact tables

Implementations

impl FactTableDetector

pub fn is_fact_table(table_name: &str) -> bool

Detect if a table name follows the fact table pattern.

Fact tables must follow the naming convention: tf_<table_name> where the table name contains only lowercase letters, numbers, and underscores.

Arguments

• table_name - Table name to check

Returns

true if the table name starts with tf_ and follows naming conventions, false otherwise.

Notes

• The check is strict: tf_ is required as a prefix
• Table names like TF_sales (uppercase prefix) are NOT recognized as fact tables
• Empty strings and tables named just tf_ without additional suffix are not valid

Examples

use fraiseql_core::compiler::fact_table::FactTableDetector;

assert!(FactTableDetector::is_fact_table("tf_sales"));
assert!(FactTableDetector::is_fact_table("tf_events"));
assert!(FactTableDetector::is_fact_table("tf_page_views_daily"));
assert!(!FactTableDetector::is_fact_table("ta_sales_by_day"));
assert!(!FactTableDetector::is_fact_table("v_user"));
assert!(!FactTableDetector::is_fact_table("TF_sales")); // uppercase prefix not recognized
assert!(!FactTableDetector::is_fact_table("tf_")); // incomplete name

pub async fn introspect( introspector: &impl DatabaseIntrospector, table_name: &str, ) -> Result<FactTableMetadata>

Introspect a fact table from the database

Queries the database schema to extract:

• Measures (numeric columns)
• Dimensions (JSONB/JSON columns)
• Denormalized filters (indexed columns)

Arguments

• introspector - Database introspection implementation
• table_name - Fact table name (must start with “tf_”)

Errors

Returns error if:

• Table is not a fact table (doesn’t start with “tf_”)
• Database query fails
• Table structure is invalid

Example

use fraiseql_core::compiler::fact_table::{FactTableDetector, DatabaseIntrospector};

let metadata = FactTableDetector::introspect(&db, "tf_sales").await?;
println!("Found {} measures", metadata.measures.len());

pub fn validate(metadata: &FactTableMetadata) -> Result<()>

Validate fact table structure

Errors

Returns error if:

• No measures found
• No dimension column found
• Measures are not numeric types

pub fn extract_dimension_paths( sample: &Value, column_name: &str, db_type: DatabaseType, ) -> Vec<DimensionPath>

Extract dimension paths from a sample JSON value

Walks through the JSON structure and extracts top-level keys as dimension paths. Nested objects are represented with dot notation (e.g., “customer.region”).

Arguments

• sample - Sample JSON value from the dimension column
• column_name - Name of the JSONB column (e.g., “dimensions”)
• db_type - Database type for generating correct JSON path syntax

Returns

Vec of DimensionPath extracted from the sample data

pub fn from_columns( table_name: String, columns: Vec<(&str, SqlType, bool)>, ) -> Result<FactTableMetadata>

Create metadata from column definitions (for testing)

Errors

Returns FraiseQLError::Validation if no time dimension column is found.