Add SQL dialect function overriding for duckdb

Author: georgestagg

ggsql-python/src/lib.rs

@@ -127,6 +127,8 @@ struct PyReaderBridge {
     obj: Py<PyAny>,
 }
 
+static ANSI_DIALECT: ggsql::reader::AnsiDialect = ggsql::reader::AnsiDialect;
+
 impl Reader for PyReaderBridge {
     fn execute_sql(&self, sql: &str) -> ggsql::Result<DataFrame> {
         Python::attach(|py| {
@@ -161,6 +163,10 @@ impl Reader for PyReaderBridge {
             Ok(())
         })
     }
+
+    fn dialect(&self) -> &dyn ggsql::reader::SqlDialect {
+        &ANSI_DIALECT
+    }
 }
 
 // ============================================================================

src/execute/casting.rs

@@ -4,7 +4,8 @@
 //! scale requirements and updating type info accordingly.
 
 use crate::plot::scale::coerce_dtypes;
-use crate::plot::{CastTargetType, Layer, ParameterValue, Plot, SqlTypeNames};
+use crate::plot::{CastTargetType, Layer, ParameterValue, Plot};
+use crate::reader::SqlDialect;
 use crate::{naming, DataSource};
 use polars::prelude::{DataType, TimeUnit};
 use std::collections::{HashMap, HashSet};
@@ -57,7 +58,7 @@ pub fn literal_to_sql(lit: &ParameterValue) -> String {
 pub fn determine_type_requirements(
     spec: &Plot,
     layer_type_info: &[Vec<TypeInfo>],
-    type_names: &SqlTypeNames,
+    dialect: &dyn SqlDialect,
 ) -> Vec<Vec<TypeRequirement>> {
     use crate::plot::scale::TransformKind;
 
@@ -123,7 +124,7 @@ pub fn determine_type_requirements(
 
             // Check if this specific column needs casting
             if let Some(cast_target) = scale_type.required_cast_type(col_dtype, &target_dtype) {
-                if let Some(sql_type) = type_names.for_target(cast_target) {
+                if let Some(sql_type) = dialect.type_name_for(cast_target) {
                     // Don't add duplicate requirements for same column
                     if !requirements.iter().any(|r| r.column == col_name) {
                         requirements.push(TypeRequirement {
@@ -155,7 +156,7 @@ pub fn determine_type_requirements(
                     };
 
                     if needs_int_cast {
-                        if let Some(sql_type) = type_names.for_target(CastTargetType::Integer) {
+                        if let Some(sql_type) = dialect.type_name_for(CastTargetType::Integer) {
                             // Don't add duplicate requirements for same column
                             if !requirements.iter().any(|r| r.column == col_name) {
                                 requirements.push(TypeRequirement {

src/execute/layer.rs

@@ -4,9 +4,9 @@
 //! transformations, stat transforms, and post-query operations.
 
 use crate::plot::{
-    AestheticValue, DefaultAestheticValue, Layer, ParameterValue, Scale, Schema, SqlTypeNames,
-    StatResult,
+    AestheticValue, DefaultAestheticValue, Layer, ParameterValue, Scale, Schema, StatResult,
 };
+use crate::reader::SqlDialect;
 use crate::{naming, DataFrame, GgsqlError, Result};
 use polars::prelude::DataType;
 use std::collections::{HashMap, HashSet};
@@ -194,14 +194,14 @@ pub fn literal_to_series(name: &str, lit: &ParameterValue, len: usize) -> polars
 /// * `layer` - The layer configuration
 /// * `schema` - The layer's schema (used for column dtype lookup)
 /// * `scales` - All resolved scales
-/// * `type_names` - SQL type names for the database backend
+/// * `dialect` - SQL dialect for the database backend
 pub fn apply_pre_stat_transform(
     query: &str,
     layer: &Layer,
     full_schema: &Schema,
     aesthetic_schema: &Schema,
     scales: &[Scale],
-    type_names: &SqlTypeNames,
+    dialect: &dyn SqlDialect,
 ) -> String {
     let mut transform_exprs: Vec<(String, String)> = vec![];
     let mut transformed_columns: HashSet<String> = HashSet::new();
@@ -237,7 +237,7 @@ pub fn apply_pre_stat_transform(
                 // Get pre-stat SQL transformation from scale type (if applicable)
                 // Each scale type's pre_stat_transform_sql() returns None if not applicable
                 if let Some(sql) =
-                    scale_type.pre_stat_transform_sql(&aes_col_name, &col_dtype, scale, type_names)
+                    scale_type.pre_stat_transform_sql(&aes_col_name, &col_dtype, scale, dialect)
                 {
                     transformed_columns.insert(aes_col_name.clone());
                     transform_exprs.push((aes_col_name, sql));
@@ -347,7 +347,7 @@ pub fn build_layer_base_query(
 /// * `base_query` - The base query from build_layer_base_query
 /// * `schema` - The layer's schema (with min/max from base_query)
 /// * `scales` - All resolved scales
-/// * `type_names` - SQL type names for the database backend
+/// * `dialect` - SQL dialect for the database backend
 /// * `execute_query` - Function to execute queries (needed for some stat transforms)
 ///
 /// # Returns
@@ -358,7 +358,7 @@ pub fn apply_layer_transforms<F>(
     base_query: &str,
     schema: &Schema,
     scales: &[Scale],
-    type_names: &SqlTypeNames,
+    dialect: &dyn SqlDialect,
     execute_query: &F,
 ) -> Result<String>
 where
@@ -398,7 +398,7 @@ where
         schema,
         &aesthetic_schema,
         scales,
-        type_names,
+        dialect,
     );
 
     // Build group_by columns from partition_by
@@ -427,6 +427,7 @@ where
         &group_by,
         &layer.parameters,
         execute_query,
+        dialect,
     )?;
 
     // Apply literal default remappings from geom defaults (e.g., y2 => 0.0 for bar baseline).

src/execute/mod.rs

@@ -876,7 +876,7 @@ pub struct PreparedData {
 /// * `reader` - A Reader implementation for executing SQL
 pub fn prepare_data_with_reader<R: Reader>(query: &str, reader: &R) -> Result<PreparedData> {
     let execute_query = |sql: &str| reader.execute_sql(sql);
-    let type_names = reader.sql_type_names();
+    let dialect = reader.dialect();
 
     // Parse once and create SourceTree
     let source_tree = parser::SourceTree::new(query)?;
@@ -1014,7 +1014,7 @@ pub fn prepare_data_with_reader<R: Reader>(query: &str, reader: &R) -> Result<Pr
 
     // Determine which columns need type casting
     let type_requirements =
-        casting::determine_type_requirements(&specs[0], &layer_type_info, &type_names);
+        casting::determine_type_requirements(&specs[0], &layer_type_info, dialect);
 
     // Update type info with post-cast dtypes
     // This ensures subsequent schema extraction and scale resolution see the correct types
@@ -1083,7 +1083,7 @@ pub fn prepare_data_with_reader<R: Reader>(query: &str, reader: &R) -> Result<Pr
             &layer_base_queries[idx],
             &layer_schemas[idx],
             &scales,
-            &type_names,
+            dialect,
             &execute_query,
         )?;
         layer_queries.push(layer_query);

src/lib.rs

@@ -48,7 +48,6 @@ pub mod writer;
 
 pub mod execute;
 
-pub mod utils;
 pub mod validate;
 
 // Re-export key types for convenience

src/plot/layer/geom/bar.rs

@@ -7,6 +7,7 @@ use super::types::get_column_name;
 use super::{DefaultAesthetics, DefaultParam, DefaultParamValue, GeomTrait, GeomType, StatResult};
 use crate::naming;
 use crate::plot::types::{DefaultAestheticValue, ParameterValue};
+use crate::reader::SqlDialect;
 use crate::{DataFrame, GgsqlError, Mappings, Result};
 
 use super::types::Schema;
@@ -81,6 +82,7 @@ impl GeomTrait for Bar {
         group_by: &[String],
         _parameters: &HashMap<String, ParameterValue>,
         _execute_query: &dyn Fn(&str) -> Result<DataFrame>,
+        _dialect: &dyn SqlDialect,
     ) -> Result<StatResult> {
         stat_bar_count(query, schema, aesthetics, group_by)
     }

src/plot/layer/geom/boxplot.rs

@@ -9,7 +9,7 @@ use crate::{
         geom::types::get_column_name, DefaultAestheticValue, DefaultParam, DefaultParamValue,
         ParameterValue, StatResult,
     },
-    utils::{sql_greatest, sql_least, sql_percentile},
+    reader::SqlDialect,
     DataFrame, GgsqlError, Mappings, Result,
 };
 
@@ -86,8 +86,9 @@ impl GeomTrait for Boxplot {
         group_by: &[String],
         parameters: &HashMap<String, ParameterValue>,
         _execute_query: &dyn Fn(&str) -> Result<DataFrame>,
+        dialect: &dyn SqlDialect,
     ) -> Result<StatResult> {
-        stat_boxplot(query, aesthetics, group_by, parameters)
+        stat_boxplot(query, aesthetics, group_by, parameters, dialect)
     }
 }
 
@@ -102,6 +103,7 @@ fn stat_boxplot(
     aesthetics: &Mappings,
     group_by: &[String],
     parameters: &HashMap<String, ParameterValue>,
+    dialect: &dyn SqlDialect,
 ) -> Result<StatResult> {
     let y = get_column_name(aesthetics, "pos2").ok_or_else(|| {
         GgsqlError::ValidationError("Boxplot requires 'y' aesthetic mapping".to_string())
@@ -147,7 +149,7 @@ fn stat_boxplot(
     }
 
     // Query for boxplot summary statistics
-    let summary = boxplot_sql_compute_summary(query, &groups, &value_col, coef);
+    let summary = boxplot_sql_compute_summary(query, &groups, &value_col, coef, dialect);
     let stats_query = boxplot_sql_append_outliers(&summary, &groups, &value_col, query, outliers);
 
     Ok(StatResult::Transformed {
@@ -162,13 +164,13 @@ fn stat_boxplot(
     })
 }
 
-fn boxplot_sql_compute_summary(from: &str, groups: &[String], value: &str, coef: &f64) -> String {
+fn boxplot_sql_compute_summary(from: &str, groups: &[String], value: &str, coef: &f64, dialect: &dyn SqlDialect) -> String {
     let groups_str = groups.join(", ");
-    let lower_expr = sql_greatest(&[&format!("q1 - {coef} * (q3 - q1)"), "min"]);
-    let upper_expr = sql_least(&[&format!("q3 + {coef} * (q3 - q1)"), "max"]);
-    let q1 = sql_percentile(value, 0.25, from, groups);
-    let median = sql_percentile(value, 0.50, from, groups);
-    let q3 = sql_percentile(value, 0.75, from, groups);
+    let lower_expr = dialect.sql_greatest(&[&format!("q1 - {coef} * (q3 - q1)"), "min"]);
+    let upper_expr = dialect.sql_least(&[&format!("q3 + {coef} * (q3 - q1)"), "max"]);
+    let q1 = dialect.sql_percentile(value, 0.25, from, groups);
+    let median = dialect.sql_percentile(value, 0.50, from, groups);
+    let q3 = dialect.sql_percentile(value, 0.75, from, groups);
     format!(
         "SELECT
           *,
@@ -293,6 +295,7 @@ fn boxplot_sql_append_outliers(
 mod tests {
     use super::*;
     use crate::plot::AestheticValue;
+    use crate::reader::AnsiDialect;
 
     // ==================== Helper Functions ====================
 
@@ -314,7 +317,7 @@ mod tests {
     #[test]
     fn test_sql_compute_summary_basic() {
         let groups = vec!["category".to_string()];
-        let result = boxplot_sql_compute_summary("data", &groups, "value", &1.5);
+        let result = boxplot_sql_compute_summary("data", &groups, "value", &1.5, &AnsiDialect);
         assert!(result.contains("NTILE(4) OVER (ORDER BY value)"));
         assert!(result.contains("AS q1"));
         assert!(result.contains("AS median"));
@@ -330,15 +333,15 @@ mod tests {
     #[test]
     fn test_sql_compute_summary_multiple_groups() {
         let groups = vec!["cat".to_string(), "region".to_string()];
-        let result = boxplot_sql_compute_summary("tbl", &groups, "val", &1.5);
+        let result = boxplot_sql_compute_summary("tbl", &groups, "val", &1.5, &AnsiDialect);
         assert!(result.contains("GROUP BY cat, region"));
         assert!(result.contains("NTILE(4) OVER (ORDER BY val)"));
     }
 
     #[test]
     fn test_sql_compute_summary_custom_coef() {
         let groups = vec!["pos1".to_string()];
-        let result = boxplot_sql_compute_summary("q", &groups, "pos2", &2.5);
+        let result = boxplot_sql_compute_summary("q", &groups, "pos2", &2.5, &AnsiDialect);
         assert!(result.contains("2.5"));
         assert!(
             result.contains("(CASE WHEN (q1 - 2.5 * (q3 - q1)) >= (min) THEN (q1 - 2.5 * (q3 - q1)) ELSE (min) END)")
@@ -364,11 +367,11 @@ mod tests {
     #[test]
     fn test_boxplot_sql_compute_summary_single_group() {
         let groups = vec!["category".to_string()];
-        let result = boxplot_sql_compute_summary("SELECT * FROM sales", &groups, "price", &1.5);
+        let result = boxplot_sql_compute_summary("SELECT * FROM sales", &groups, "price", &1.5, &AnsiDialect);
 
-        let q1 = sql_percentile("price", 0.25, "SELECT * FROM sales", &groups);
-        let median = sql_percentile("price", 0.50, "SELECT * FROM sales", &groups);
-        let q3 = sql_percentile("price", 0.75, "SELECT * FROM sales", &groups);
+        let q1 = AnsiDialect.sql_percentile("price", 0.25, "SELECT * FROM sales", &groups);
+        let median = AnsiDialect.sql_percentile("price", 0.50, "SELECT * FROM sales", &groups);
+        let q3 = AnsiDialect.sql_percentile("price", 0.75, "SELECT * FROM sales", &groups);
         let expected = format!(
             r#"SELECT
           *,
@@ -394,11 +397,11 @@ mod tests {
     #[test]
     fn test_boxplot_sql_compute_summary_multiple_groups() {
         let groups = vec!["region".to_string(), "product".to_string()];
-        let result = boxplot_sql_compute_summary("SELECT * FROM data", &groups, "revenue", &1.5);
+        let result = boxplot_sql_compute_summary("SELECT * FROM data", &groups, "revenue", &1.5, &AnsiDialect);
 
-        let q1 = sql_percentile("revenue", 0.25, "SELECT * FROM data", &groups);
-        let median = sql_percentile("revenue", 0.50, "SELECT * FROM data", &groups);
-        let q3 = sql_percentile("revenue", 0.75, "SELECT * FROM data", &groups);
+        let q1 = AnsiDialect.sql_percentile("revenue", 0.25, "SELECT * FROM data", &groups);
+        let median = AnsiDialect.sql_percentile("revenue", 0.50, "SELECT * FROM data", &groups);
+        let q3 = AnsiDialect.sql_percentile("revenue", 0.75, "SELECT * FROM data", &groups);
         let expected = format!(
             r#"SELECT
           *,
@@ -507,7 +510,7 @@ mod tests {
         );
         parameters.insert("outliers".to_string(), ParameterValue::Boolean(true));
 
-        let result = stat_boxplot("SELECT * FROM data", &aesthetics, &groups, &parameters);
+        let result = stat_boxplot("SELECT * FROM data", &aesthetics, &groups, &parameters, &AnsiDialect);
 
         assert!(result.is_err());
         assert!(result.unwrap_err().to_string().contains("coef"));
@@ -522,7 +525,7 @@ mod tests {
         parameters.insert("outliers".to_string(), ParameterValue::Boolean(true));
         // Missing coef
 
-        let result = stat_boxplot("SELECT * FROM data", &aesthetics, &groups, &parameters);
+        let result = stat_boxplot("SELECT * FROM data", &aesthetics, &groups, &parameters, &AnsiDialect);
 
         assert!(result.is_err());
         assert!(result.unwrap_err().to_string().contains("coef"));
@@ -540,7 +543,7 @@ mod tests {
             ParameterValue::String("yes".to_string()),
         );
 
-        let result = stat_boxplot("SELECT * FROM data", &aesthetics, &groups, &parameters);
+        let result = stat_boxplot("SELECT * FROM data", &aesthetics, &groups, &parameters, &AnsiDialect);
 
         assert!(result.is_err());
         assert!(result.unwrap_err().to_string().contains("outliers"));
@@ -555,7 +558,7 @@ mod tests {
         parameters.insert("coef".to_string(), ParameterValue::Number(1.5));
         // Missing outliers
 
-        let result = stat_boxplot("SELECT * FROM data", &aesthetics, &groups, &parameters);
+        let result = stat_boxplot("SELECT * FROM data", &aesthetics, &groups, &parameters, &AnsiDialect);
 
         assert!(result.is_err());
         assert!(result.unwrap_err().to_string().contains("outliers"));