feat: add 'date.to_start_of_interval' function for truncating values to interval boundaries

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":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"}]}}"#
+            @r#"{"frames":[["1:9-31",{"columns":[{"Single":{"name":["a","beta"],"target_id":122,"target_name":null}},{"Single":{"name":["a","gamma"],"target_id":123,"target_name":null}}],"inputs":[{"id":120,"name":"a","table":["default_db","a"]}]}]],"nodes":[{"id":120,"kind":"Ident","span":"1:0-6","ident":{"Ident":["default_db","a"]},"parent":125},{"id":122,"kind":"Ident","span":"1:18-22","ident":{"Ident":["this","a","beta"]},"targets":[120],"parent":124},{"id":123,"kind":"Ident","span":"1:24-29","ident":{"Ident":["this","a","gamma"]},"targets":[120],"parent":124},{"id":124,"kind":"Tuple","span":"1:16-31","children":[122,123],"parent":125},{"id":125,"kind":"TransformCall: Select","span":"1:9-31","children":[120,124]}],"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":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"}]}}"#
+            @r#"{"frames":[["1:9-31",{"columns":[{"Single":{"name":["a","beta"],"target_id":122,"target_name":null}},{"Single":{"name":["a","gamma"],"target_id":123,"target_name":null}}],"inputs":[{"id":120,"name":"a","table":["default_db","a"]}]}]],"nodes":[{"id":120,"kind":"Ident","span":"1:0-6","ident":{"Ident":["default_db","a"]},"parent":125},{"id":122,"kind":"Ident","span":"1:18-22","ident":{"Ident":["this","a","beta"]},"targets":[120],"parent":124},{"id":123,"kind":"Ident","span":"1:24-29","ident":{"Ident":["this","a","gamma"]},"targets":[120],"parent":124},{"id":124,"kind":"Tuple","span":"1:16-31","children":[122,123],"parent":125},{"id":125,"kind":"TransformCall: Select","span":"1:9-31","children":[120,124]}],"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: 121
+          target_id: 122
           target_name: null
         - !Single
           name:
           - tracks
           - album
-          target_id: 122
+          target_id: 123
           target_name: null
         inputs:
-        - id: 119
+        - id: 120
           name: tracks
           table:
           - default_db
           - tracks
     nodes:
-    - id: 119
+    - id: 120
       kind: Ident
       span: 1:0-11
       ident: !Ident
       - default_db
       - tracks
-      parent: 124
-    - id: 121
+      parent: 125
+    - id: 122
       kind: Ident
       span: 1:22-28
       ident: !Ident
       - this
       - tracks
       - artist
       targets:
-      - 119
-      parent: 123
-    - id: 122
+      - 120
+      parent: 124
+    - id: 123
       kind: Ident
       span: 1:30-35
       ident: !Ident
       - this
       - tracks
       - album
       targets:
-      - 119
-      parent: 123
-    - id: 123
+      - 120
+      parent: 124
+    - id: 124
       kind: Tuple
       span: 1:21-36
       children:
-      - 121
       - 122
-      parent: 124
-    - id: 124
+      - 123
+      parent: 125
+    - id: 125
       kind: 'TransformCall: Select'
       span: 1:14-36
       children:
-      - 119
-      - 123
+      - 120
+      - 124
     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: 135
+      target_id: 136
       target_name: ~
   - Single:
       name:
         - employees
         - salary
-      target_id: 136
+      target_id: 137
       target_name: ~
 inputs:
-  - id: 133
+  - id: 134
     name: employees
     table:
       - default_db
       - employees
-  - id: 122
+  - id: 123
     name: managers
     table:
       - default_db

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

@@ -1,22 +1,21 @@
 ---
 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: 120
+      input_id: 121
       except: []
   - All:
-      input_id: 117
+      input_id: 118
       except: []
 inputs:
-  - id: 120
+  - id: 121
     name: table_1
     table:
       - default_db
       - table_1
-  - id: 117
+  - id: 118
     name: customers
     table:
       - default_db

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

@@ -1,33 +1,32 @@
 ---
 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: 130
+      target_id: 131
       target_name: ~
   - Single:
       name:
         - e
         - gender
-      target_id: 131
+      target_id: 132
       target_name: ~
   - Single:
       name:
         - emp_salary
-      target_id: 149
+      target_id: 150
       target_name: ~
 inputs:
-  - id: 123
+  - id: 124
     name: e
     table:
       - default_db
       - employees
-  - id: 120
+  - id: 121
     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: 124
+      target_id: 125
       target_name: ~
   - Single:
       name:
         - orders
         - gross
-      target_id: 125
+      target_id: 126
       target_name: ~
   - Single:
       name:
         - orders
         - tax
-      target_id: 126
+      target_id: 127
       target_name: ~
   - Single:
       name: ~
-      target_id: 127
+      target_id: 128
       target_name: ~
 inputs:
-  - id: 122
+  - id: 123
     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: 119
+            input_id: 120
             except: []
       inputs:
-        - id: 119
+        - id: 120
           name: c_invoice
           table:
             - default_db
@@ -94,14 +94,14 @@ lineage:
         name:
           - c_invoice
           - issued_at
-        target_id: 121
+        target_id: 122
         target_name: ~
     - Single:
         name: ~
-        target_id: 137
+        target_id: 138
         target_name: ~
   inputs:
-    - id: 119
+    - id: 120
       name: c_invoice
       table:
         - default_db

prqlc/prqlc/src/semantic/std.prql

@@ -231,6 +231,7 @@ module date {
   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
+  let to_start_of_interval = amount<int> `noresolve.unit` column -> internal std.date.to_start_of_interval
 }
 
 ## File-reading functions, primarily for DuckDB

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

@@ -142,6 +142,7 @@ module date {
   # 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})"
+  let to_start_of_interval = amount unit column -> null
 }
 
 # Source-reading functions, primarily for DuckDB
@@ -229,6 +230,7 @@ module bigquery {
     let diff = unit start end -> s"DATE_DIFF({end:0}, {start:0}, {unit:0})"
     # BigQuery requires parentheses for CURRENT_TIMESTAMP
     let now = -> s"CURRENT_TIMESTAMP()"
+    let to_start_of_interval = amount unit column -> s"TIMESTAMP_BUCKET(CAST({column:0} AS TIMESTAMP), INTERVAL {amount:0} {unit:0})"
   }
 
   let regex_search = text pattern -> s"REGEXP_CONTAINS({text:0}, {pattern:0})"
@@ -250,6 +252,8 @@ module clickhouse {
     let diff = unit start end -> s"dateDiff('{unit:0}', {start:0}, {end:0})"
     # Clickhouse uses now() for current timestamp
     let now = -> s"now()"
+    # https://clickhouse.com/docs/en/sql-reference/functions/date-time-functions#toStartOfInterval
+    let to_start_of_interval = amount unit column -> s"toStartOfInterval({column:0}, INTERVAL {amount:0} {unit:0})"
   }
 
   let regex_search = text pattern -> s"match({text:0}, {pattern:0})"
@@ -278,6 +282,7 @@ module duckdb {
     let to_text = format column -> s"strftime({column:0}, {format:0})"
     # https://duckdb.org/docs/sql/functions/date
     let diff = unit start end -> s"datediff('{unit:0}', {start:0}, {end:0})"
+    let to_start_of_interval = amount unit column -> s"time_bucket(INTERVAL '{amount:0} {unit:0}', {column:0})"
   }
 
   let regex_search = text pattern -> s"REGEXP_MATCHES({text:0}, {pattern:0})"
@@ -309,6 +314,7 @@ module mssql {
   module date {
     # https://learn.microsoft.com/en-us/sql/t-sql/functions/format-transact-sql
     let to_text = format column -> s"FORMAT({column:0}, {format:0})"
+    let to_start_of_interval = amount unit column -> null
   }
 
   let regex_search = text pattern -> null
@@ -334,6 +340,7 @@ module mysql {
     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})"
+    let to_start_of_interval = amount unit column -> null
   }
 
   # 'c' for case-sensitive
@@ -366,6 +373,7 @@ module postgres {
     let to_text = format column -> s"TO_CHAR({column:0}, {format:0})"
     # PostgreSQL has no built-in DATEDIFF equivalent
     let diff = unit start end -> null
+    let to_start_of_interval = amount unit column -> s"date_bin('{amount:0} {unit:0}', {column:0}, TIMESTAMP '1970-01-01 00:00:00')"
   }
 
   @{binding_strength=9}