feat: add 'date.trunc' function for date trunc support

prqlc/bindings/prqlc-python/src/lib.rs

@@ -232,15 +232,15 @@ mod test {
     fn debug_prql_lineage() {
         assert_snapshot!(
             debug::prql_lineage(r#"from a | select { beta, gamma }"#).unwrap(),
-            @r#"{"frames":[["1:9-31",{"columns":[{"Single":{"name":["a","beta"],"target_id":120,"target_name":null}},{"Single":{"name":["a","gamma"],"target_id":121,"target_name":null}}],"inputs":[{"id":118,"name":"a","table":["default_db","a"]}]}]],"nodes":[{"id":118,"kind":"Ident","span":"1:0-6","ident":{"Ident":["default_db","a"]},"parent":123},{"id":120,"kind":"Ident","span":"1:18-22","ident":{"Ident":["this","a","beta"]},"targets":[118],"parent":122},{"id":121,"kind":"Ident","span":"1:24-29","ident":{"Ident":["this","a","gamma"]},"targets":[118],"parent":122},{"id":122,"kind":"Tuple","span":"1:16-31","children":[120,121],"parent":123},{"id":123,"kind":"TransformCall: Select","span":"1:9-31","children":[118,122]}],"ast":{"name":"Project","stmts":[{"VarDef":{"kind":"Main","name":"main","value":{"Pipeline":{"exprs":[{"FuncCall":{"name":{"Ident":["from"],"span":"1:0-4"},"args":[{"Ident":["a"],"span":"1:5-6"}]},"span":"1:0-6"},{"FuncCall":{"name":{"Ident":["select"],"span":"1:9-15"},"args":[{"Tuple":[{"Ident":["beta"],"span":"1:18-22"},{"Ident":["gamma"],"span":"1:24-29"}],"span":"1:16-31"}]},"span":"1:9-31"}]},"span":"1:0-31"}},"span":"1:0-31"}]}}"#
+            @r#"{"frames":[["1:9-31",{"columns":[{"Single":{"name":["a","beta"],"target_id":121,"target_name":null}},{"Single":{"name":["a","gamma"],"target_id":122,"target_name":null}}],"inputs":[{"id":119,"name":"a","table":["default_db","a"]}]}]],"nodes":[{"id":119,"kind":"Ident","span":"1:0-6","ident":{"Ident":["default_db","a"]},"parent":124},{"id":121,"kind":"Ident","span":"1:18-22","ident":{"Ident":["this","a","beta"]},"targets":[119],"parent":123},{"id":122,"kind":"Ident","span":"1:24-29","ident":{"Ident":["this","a","gamma"]},"targets":[119],"parent":123},{"id":123,"kind":"Tuple","span":"1:16-31","children":[121,122],"parent":124},{"id":124,"kind":"TransformCall: Select","span":"1:9-31","children":[119,123]}],"ast":{"name":"Project","stmts":[{"VarDef":{"kind":"Main","name":"main","value":{"Pipeline":{"exprs":[{"FuncCall":{"name":{"Ident":["from"],"span":"1:0-4"},"args":[{"Ident":["a"],"span":"1:5-6"}]},"span":"1:0-6"},{"FuncCall":{"name":{"Ident":["select"],"span":"1:9-15"},"args":[{"Tuple":[{"Ident":["beta"],"span":"1:18-22"},{"Ident":["gamma"],"span":"1:24-29"}],"span":"1:16-31"}]},"span":"1:9-31"}]},"span":"1:0-31"}},"span":"1:0-31"}]}}"#
         );
     }
 
     #[test]
     fn debug_pl_to_lineage() {
         assert_snapshot!(
             prql_to_pl(r#"from a | select { beta, gamma }"#).and_then(|x| debug::pl_to_lineage(&x)).unwrap(),
-            @r#"{"frames":[["1:9-31",{"columns":[{"Single":{"name":["a","beta"],"target_id":120,"target_name":null}},{"Single":{"name":["a","gamma"],"target_id":121,"target_name":null}}],"inputs":[{"id":118,"name":"a","table":["default_db","a"]}]}]],"nodes":[{"id":118,"kind":"Ident","span":"1:0-6","ident":{"Ident":["default_db","a"]},"parent":123},{"id":120,"kind":"Ident","span":"1:18-22","ident":{"Ident":["this","a","beta"]},"targets":[118],"parent":122},{"id":121,"kind":"Ident","span":"1:24-29","ident":{"Ident":["this","a","gamma"]},"targets":[118],"parent":122},{"id":122,"kind":"Tuple","span":"1:16-31","children":[120,121],"parent":123},{"id":123,"kind":"TransformCall: Select","span":"1:9-31","children":[118,122]}],"ast":{"name":"Project","stmts":[{"VarDef":{"kind":"Main","name":"main","value":{"Pipeline":{"exprs":[{"FuncCall":{"name":{"Ident":["from"],"span":"1:0-4"},"args":[{"Ident":["a"],"span":"1:5-6"}]},"span":"1:0-6"},{"FuncCall":{"name":{"Ident":["select"],"span":"1:9-15"},"args":[{"Tuple":[{"Ident":["beta"],"span":"1:18-22"},{"Ident":["gamma"],"span":"1:24-29"}],"span":"1:16-31"}]},"span":"1:9-31"}]},"span":"1:0-31"}},"span":"1:0-31"}]}}"#
+            @r#"{"frames":[["1:9-31",{"columns":[{"Single":{"name":["a","beta"],"target_id":121,"target_name":null}},{"Single":{"name":["a","gamma"],"target_id":122,"target_name":null}}],"inputs":[{"id":119,"name":"a","table":["default_db","a"]}]}]],"nodes":[{"id":119,"kind":"Ident","span":"1:0-6","ident":{"Ident":["default_db","a"]},"parent":124},{"id":121,"kind":"Ident","span":"1:18-22","ident":{"Ident":["this","a","beta"]},"targets":[119],"parent":123},{"id":122,"kind":"Ident","span":"1:24-29","ident":{"Ident":["this","a","gamma"]},"targets":[119],"parent":123},{"id":123,"kind":"Tuple","span":"1:16-31","children":[121,122],"parent":124},{"id":124,"kind":"TransformCall: Select","span":"1:9-31","children":[119,123]}],"ast":{"name":"Project","stmts":[{"VarDef":{"kind":"Main","name":"main","value":{"Pipeline":{"exprs":[{"FuncCall":{"name":{"Ident":["from"],"span":"1:0-4"},"args":[{"Ident":["a"],"span":"1:5-6"}]},"span":"1:0-6"},{"FuncCall":{"name":{"Ident":["select"],"span":"1:9-15"},"args":[{"Tuple":[{"Ident":["beta"],"span":"1:18-22"},{"Ident":["gamma"],"span":"1:24-29"}],"span":"1:16-31"}]},"span":"1:9-31"}]},"span":"1:0-31"}},"span":"1:0-31"}]}}"#
         );
     }
 }

prqlc/prqlc/src/cli/test.rs

@@ -434,61 +434,61 @@ fn debug() {
           name:
           - tracks
           - artist
-          target_id: 120
+          target_id: 121
           target_name: null
         - !Single
           name:
           - tracks
           - album
-          target_id: 121
+          target_id: 122
           target_name: null
         inputs:
-        - id: 118
+        - id: 119
           name: tracks
           table:
           - default_db
           - tracks
     nodes:
-    - id: 118
+    - id: 119
       kind: Ident
       span: 1:0-11
       ident: !Ident
       - default_db
       - tracks
-      parent: 123
-    - id: 120
+      parent: 124
+    - id: 121
       kind: Ident
       span: 1:22-28
       ident: !Ident
       - this
       - tracks
       - artist
       targets:
-      - 118
-      parent: 122
-    - id: 121
+      - 119
+      parent: 123
+    - id: 122
       kind: Ident
       span: 1:30-35
       ident: !Ident
       - this
       - tracks
       - album
       targets:
-      - 118
-      parent: 122
-    - id: 122
+      - 119
+      parent: 123
+    - id: 123
       kind: Tuple
       span: 1:21-36
       children:
-      - 120
       - 121
-      parent: 123
-    - id: 123
+      - 122
+      parent: 124
+    - id: 124
       kind: 'TransformCall: Select'
       span: 1:14-36
       children:
-      - 118
-      - 122
+      - 119
+      - 123
     ast:
       name: Project
       stmts:

prqlc/prqlc/src/semantic/resolver/snapshots/prqlc__semantic__resolver__test__append_union_different_tables.snap

@@ -7,21 +7,21 @@ columns:
       name:
         - employees
         - name
-      target_id: 134
+      target_id: 135
       target_name: ~
   - Single:
       name:
         - employees
         - salary
-      target_id: 135
+      target_id: 136
       target_name: ~
 inputs:
-  - id: 132
+  - id: 133
     name: employees
     table:
       - default_db
       - employees
-  - id: 121
+  - id: 122
     name: managers
     table:
       - default_db

prqlc/prqlc/src/semantic/resolver/snapshots/prqlc__semantic__resolver__test__frames_and_names-2.snap

@@ -1,21 +1,22 @@
 ---
 source: prqlc/prqlc/src/semantic/resolver/mod.rs
+assertion_line: 190
 expression: "resolve_lineage(r#\"\n            from table_1\n            join customers (==customer_no)\n            \"#).unwrap()"
 ---
 columns:
   - All:
-      input_id: 119
+      input_id: 120
       except: []
   - All:
-      input_id: 116
+      input_id: 117
       except: []
 inputs:
-  - id: 119
+  - id: 120
     name: table_1
     table:
       - default_db
       - table_1
-  - id: 116
+  - id: 117
     name: customers
     table:
       - default_db

prqlc/prqlc/src/semantic/resolver/snapshots/prqlc__semantic__resolver__test__frames_and_names-3.snap

@@ -1,32 +1,33 @@
 ---
 source: prqlc/prqlc/src/semantic/resolver/mod.rs
+assertion_line: 198
 expression: "resolve_lineage(r#\"\n            from e = employees\n            join salaries (==emp_no)\n            group {e.emp_no, e.gender} (\n                aggregate {\n                    emp_salary = average salaries.salary\n                }\n            )\n            \"#).unwrap()"
 ---
 columns:
   - Single:
       name:
         - e
         - emp_no
-      target_id: 129
+      target_id: 130
       target_name: ~
   - Single:
       name:
         - e
         - gender
-      target_id: 130
+      target_id: 131
       target_name: ~
   - Single:
       name:
         - emp_salary
-      target_id: 148
+      target_id: 149
       target_name: ~
 inputs:
-  - id: 122
+  - id: 123
     name: e
     table:
       - default_db
       - employees
-  - id: 119
+  - id: 120
     name: salaries
     table:
       - default_db

prqlc/prqlc/src/semantic/resolver/snapshots/prqlc__semantic__resolver__test__frames_and_names.snap

@@ -7,26 +7,26 @@ columns:
       name:
         - orders
         - customer_no
-      target_id: 123
+      target_id: 124
       target_name: ~
   - Single:
       name:
         - orders
         - gross
-      target_id: 124
+      target_id: 125
       target_name: ~
   - Single:
       name:
         - orders
         - tax
-      target_id: 125
+      target_id: 126
       target_name: ~
   - Single:
       name: ~
-      target_id: 126
+      target_id: 127
       target_name: ~
 inputs:
-  - id: 121
+  - id: 122
     name: orders
     table:
       - default_db

prqlc/prqlc/src/semantic/resolver/snapshots/prqlc__semantic__resolver__transforms__tests__aggregate_positional_arg-2.snap

@@ -21,10 +21,10 @@ TransformCall:
     lineage:
       columns:
         - All:
-            input_id: 118
+            input_id: 119
             except: []
       inputs:
-        - id: 118
+        - id: 119
           name: c_invoice
           table:
             - default_db
@@ -94,14 +94,14 @@ lineage:
         name:
           - c_invoice
           - issued_at
-        target_id: 120
+        target_id: 121
         target_name: ~
     - Single:
         name: ~
-        target_id: 136
+        target_id: 137
         target_name: ~
   inputs:
-    - id: 118
+    - id: 119
       name: c_invoice
       table:
         - default_db

prqlc/prqlc/src/semantic/std.prql

@@ -230,6 +230,7 @@ module date {
   let now = -> <timestamp> internal std.date.now
   let to_text = format<text> column -> <text> internal std.date.to_text
   let diff = `noresolve.unit` start end -> <int> internal std.date.diff
+  let trunc = unit<text> column -> internal std.date.trunc
 }
 
 ## File-reading functions, primarily for DuckDB

prqlc/prqlc/src/sql/gen_expr.rs

@@ -108,6 +108,7 @@ pub(super) fn translate_expr(expr: rq::Expr, ctx: &mut Context) -> Result<ExprOr
                 "std.date.to_text" => {
                     return Ok(process_date_to_text(&expr, name, args, ctx)?.into())
                 }
+                "std.date.trunc" => return Ok(process_date_trunc(&expr, name, args, ctx)?.into()),
                 _ => match try_into_between(expr.clone(), ctx)? {
                     Some(between_expr) => return Ok(between_expr.into()),
                     None => {
@@ -208,6 +209,92 @@ fn process_array_in(
     }
 }
 
+/// Translates PRQL date truncation to dialect-specific SQL.
+/// BigQuery uses unquoted uppercase date_part after the column: DATE_TRUNC(col, DAY)
+/// MSSQL uses unquoted lowercase datepart before the column: DATETRUNC(day, col)
+/// All other dialects use: DATE_TRUNC('unit', col)
+fn process_date_trunc(
+    expr: &rq::Expr,
+    op_name: &str,
+    args: &[rq::Expr],
+    ctx: &mut Context,
+) -> Result<sql_ast::Expr> {
+    if let [unit_expr @ rq::Expr {
+        kind: rq::ExprKind::Literal(Literal::String(unit)),
+        ..
+    }, col_expr] = args
+    {
+        use crate::sql::dialect::{BigQueryDialect, MsSqlDialect};
+
+        if ctx.dialect.is::<BigQueryDialect>() {
+            // BigQuery: DATE_TRUNC(col, DAY) — unit is an unquoted uppercase keyword
+            let col = translate_expr(col_expr.clone(), ctx)?.into_ast();
+            let unit_upper = unit.to_uppercase();
+            return Ok(sql_ast::Expr::Function(Function {
+                name: ObjectName(vec![sqlparser::ast::ObjectNamePart::Identifier(
+                    sql_ast::Ident::new("DATE_TRUNC"),
+                )]),
+                args: sql_ast::FunctionArguments::List(FunctionArgumentList {
+                    args: vec![
+                        FunctionArg::Unnamed(FunctionArgExpr::Expr(col)),
+                        FunctionArg::Unnamed(FunctionArgExpr::Expr(sql_ast::Expr::Identifier(
+                            sql_ast::Ident::new(unit_upper),
+                        ))),
+                    ],
+                    clauses: vec![],
+                    duplicate_treatment: None,
+                }),
+                over: None,
+                filter: None,
+                null_treatment: None,
+                within_group: vec![],
+                parameters: sql_ast::FunctionArguments::None,
+                uses_odbc_syntax: false,
+            }));
+        } else if ctx.dialect.is::<MsSqlDialect>() {
+            // MSSQL: DATETRUNC(day, col) — unit is an unquoted lowercase keyword (SQL Server 2022+)
+            let col = translate_expr(col_expr.clone(), ctx)?.into_ast();
+            let unit_lower = unit.to_lowercase();
+            return Ok(sql_ast::Expr::Function(Function {
+                name: ObjectName(vec![sqlparser::ast::ObjectNamePart::Identifier(
+                    sql_ast::Ident::new("DATETRUNC"),
+                )]),
+                args: sql_ast::FunctionArguments::List(FunctionArgumentList {
+                    args: vec![
+                        FunctionArg::Unnamed(FunctionArgExpr::Expr(sql_ast::Expr::Identifier(
+                            sql_ast::Ident::new(unit_lower),
+                        ))),
+                        FunctionArg::Unnamed(FunctionArgExpr::Expr(col)),
+                    ],
+                    clauses: vec![],
+                    duplicate_treatment: None,
+                }),
+                over: None,
+                filter: None,
+                null_treatment: None,
+                within_group: vec![],
+                parameters: sql_ast::FunctionArguments::None,
+                uses_odbc_syntax: false,
+            }));
+        }
+
+        // All other dialects: DATE_TRUNC('unit', col) via the s-string in std.sql.prql
+        let expr = rq::Expr {
+            kind: rq::ExprKind::Operator {
+                name: op_name.to_string(),
+                args: vec![unit_expr.clone(), col_expr.clone()],
+            },
+            ..expr.clone()
+        };
+        Ok(super::operators::translate_operator_expr(expr, ctx)?.into_ast())
+    } else {
+        Err(
+            Error::new_simple("`std.date.trunc` only supports a string literal as unit")
+                .with_span(expr.span),
+        )
+    }
+}
+
 /// Translates PRQL date format (based on `chrono` crate) to dialect specific date format
 /// For now only date format as string literal is supported
 fn process_date_to_text(

prqlc/prqlc/src/sql/std.sql.prql

@@ -130,9 +130,18 @@ module date {
   # Snowflake: https://docs.snowflake.com/en/sql-reference/functions/datediff
   # Redshift: https://docs.aws.amazon.com/redshift/latest/dg/r_DATEDIFF_function.html
   let diff = unit start end -> s"DATEDIFF({unit:0}, {start:0}, {end:0})"
-
   # CURRENT_TIMESTAMP is ANSI SQL standard; dialects may override.
   let now = -> s"CURRENT_TIMESTAMP"
+  # Clickhouse: https://clickhouse.com/docs/en/sql-reference/functions/date-time-functions#date_trunc
+  # DuckDB: https://duckdb.org/docs/sql/functions/timestamp#date_truncdate_part-timestamp
+  # MariaDB: https://mariadb.com/kb/en/date_format/
+  # MySQL: https://dev.mysql.com/doc/refman/8.4/en/date-and-time-functions.html#function_date-trunc
+  # Postgres: https://www.postgresql.org/docs/current/functions-datetime.html#FUNCTIONS-DATETIME-TRUNC
+  # SQLite: https://www.sqlite.org/lang_datefunc.html
+  # MSSQL: https://learn.microsoft.com/en-us/sql/t-sql/functions/datetrunc-transact-sql
+  # BigQuery: https://cloud.google.com/bigquery/docs/reference/standard-sql/date_functions#date_trunc
+  # Snowflake: https://docs.snowflake.com/en/sql-reference/functions/date_trunc
+  let trunc = unit column -> s"DATE_TRUNC({unit:0}, {column:0})"
 }
 
 # Source-reading functions, primarily for DuckDB
@@ -323,6 +332,8 @@ module mysql {
     let diff = unit start end -> s"TIMESTAMPDIFF({unit:0}, {start:0}, {end:0})"
     # MySQL uses NOW() function
     let now = -> s"NOW()"
+    # https://dev.mysql.com/doc/refman/8.4/en/date-and-time-functions.html#function_date-trunc
+    let trunc = unit column -> s"DATE_TRUNC({unit:0}, {column:0})"
   }
 
   # 'c' for case-sensitive