feat(cubesql): Support PERCENTILE_CONT SQL push down (#8697)

Author: MazterQyou

packages/cubejs-backend-native/Cargo.lock

@@ -3831,6 +3831,8 @@ export class BaseQuery {
         // DATEADD is being rewritten to DATE_ADD
         // DATEADD: 'DATEADD({{ date_part }}, {{ interval }}, {{ args[2] }})',
         DATE: 'DATE({{ args_concat }})',
+
+        PERCENTILECONT: 'PERCENTILE_CONT({{ args_concat }})',
       },
       statements: {
         select: '{% if ctes %} WITH \n' +
@@ -3893,6 +3895,7 @@ export class BaseQuery {
         like: '{{ expr }} {% if negated %}NOT {% endif %}LIKE {{ pattern }}',
         ilike: '{{ expr }} {% if negated %}NOT {% endif %}ILIKE {{ pattern }}',
         like_escape: '{{ like_expr }} ESCAPE {{ escape_char }}',
+        within_group: '{{ fun_sql }} WITHIN GROUP (ORDER BY {{ within_group_concat }})',
         concat_strings: '{{ strings | join(\' || \' ) }}',
       },
       tesseract: {

packages/cubejs-schema-compiler/src/adapter/BaseQuery.js

@@ -259,6 +259,7 @@ export class BigqueryQuery extends BaseQuery {
     templates.functions.DATE_ADD = 'DATETIME_ADD(DATETIME({{ args[0] }}), INTERVAL {{ interval }} {{ date_part }})';
     templates.functions.CURRENTDATE = 'CURRENT_DATE';
     delete templates.functions.TO_CHAR;
+    delete templates.functions.PERCENTILECONT;
     templates.expressions.binary = '{% if op == \'%\' %}MOD({{ left }}, {{ right }}){% else %}({{ left }} {{ op }} {{ right }}){% endif %}';
     templates.expressions.interval = 'INTERVAL {{ interval }}';
     templates.expressions.extract = 'EXTRACT({% if date_part == \'DOW\' %}DAYOFWEEK{% elif date_part == \'DOY\' %}DAYOFYEAR{% else %}{{ date_part }}{% endif %} FROM {{ expr }})';

packages/cubejs-schema-compiler/src/adapter/BigqueryQuery.ts

@@ -272,6 +272,7 @@ export class ClickHouseQuery extends BaseQuery {
     // TODO: Introduce additional filter in jinja? or parseDateTimeBestEffort?
     // https://github.com/ClickHouse/ClickHouse/issues/19351
     templates.expressions.timestamp_literal = 'parseDateTimeBestEffort(\'{{ value }}\')';
+    delete templates.functions.PERCENTILECONT;
     delete templates.expressions.like_escape;
     templates.quotes.identifiers = '`';
     templates.quotes.escape = '\\`';

packages/cubejs-schema-compiler/src/adapter/ClickHouseQuery.ts

@@ -246,6 +246,8 @@ export class MssqlQuery extends BaseQuery {
     const templates = super.sqlTemplates();
     templates.functions.LEAST = 'LEAST({{ args_concat }})';
     templates.functions.GREATEST = 'GREATEST({{ args_concat }})';
+    // PERCENTILE_CONT works but requires PARTITION BY
+    delete templates.functions.PERCENTILECONT;
     delete templates.expressions.ilike;
     // NOTE: this template contains a comma; two order expressions are being generated
     templates.expressions.sort = '{{ expr }} IS NULL {% if nulls_first %}DESC{% else %}ASC{% endif %}, {{ expr }} {% if asc %}ASC{% else %}DESC{% endif %}';

packages/cubejs-schema-compiler/src/adapter/MssqlQuery.ts

@@ -165,6 +165,8 @@ export class MysqlQuery extends BaseQuery {
 
   public sqlTemplates() {
     const templates = super.sqlTemplates();
+    // PERCENTILE_CONT works but requires PARTITION BY
+    delete templates.functions.PERCENTILECONT;
     templates.quotes.identifiers = '`';
     templates.quotes.escape = '\\`';
     // NOTE: this template contains a comma; two order expressions are being generated

packages/cubejs-schema-compiler/src/adapter/MysqlQuery.ts

@@ -137,6 +137,7 @@ export class PrestodbQuery extends BaseQuery {
     const templates = super.sqlTemplates();
     templates.functions.DATETRUNC = 'DATE_TRUNC({{ args_concat }})';
     templates.functions.DATEPART = 'DATE_PART({{ args_concat }})';
+    delete templates.functions.PERCENTILECONT;
     templates.statements.select = 'SELECT {{ select_concat | map(attribute=\'aliased\') | join(\', \') }} \n' +
       'FROM (\n  {{ from }}\n) AS {{ from_alias }} \n' +
       '{% if group_by %} GROUP BY {{ group_by }}{% endif %}' +

packages/cubejs-schema-compiler/src/adapter/PrestodbQuery.ts

@@ -10,14 +10,14 @@ homepage = "https://cube.dev"
 
 [dependencies]
 arc-swap = "1"
-datafusion = { git = 'https://github.com/cube-js/arrow-datafusion.git', rev = "ecd17b3a6f3f902ac42814344bddc891c41686e6", default-features = false, features = [
+datafusion = { git = 'https://github.com/cube-js/arrow-datafusion.git', rev = "454020d6ad0eb5987aa0dbc3a3b4db8c24ccc1b5", default-features = false, features = [
     "regex_expressions",
     "unicode_expressions",
 ] }
 thiserror = "2"
 cubeclient = { path = "../cubeclient" }
 pg-srv = { path = "../pg-srv" }
-sqlparser = { git = 'https://github.com/cube-js/sqlparser-rs.git', rev = "6a54d27d3b75a04b9f9cbe309a83078aa54b32fd" }
+sqlparser = { git = 'https://github.com/cube-js/sqlparser-rs.git', rev = "34f22de680caa5fe586def5b336d56efe43c8cc4" }
 base64 = "0.13.0"
 tokio = { version = "^1.35", features = ["full", "rt", "tracing"] }
 serde = { version = "^1.0", features = ["derive"] }