feat(cubesql): Improve DataGrip compatibility (cube-js#9825)

List of changes:

- Parse explicit `DISTINCT` option with set operators (e.g. `UNION DISTINCT`)
- `array_agg` function can now be prefixed with `pg_catalog.`
- New introspection tables are supported:
  - `pg_auth_members`
  - `pg_available_extension_versions`
  - `pg_cast`
  - `pg_event_trigger`
  - `pg_foreign_data_wrapper`
  - `pg_foreign_server`
  - `pg_foreign_table`
  - `pg_language`
  - `pg_locks`
  - `pg_operator`
  - `pg_rewrite`
  - `pg_tablespace`
  - `pg_timezone_abbrevs`
  - `pg_timezone_names`
  - `pg_user_mapping`
- New introspection functions are supported:
  - `pg_is_in_recovery`
  - `pg_postmaster_start_time`
  - `pg_tablespace_location`
  - `txid_current`
- System field `xmin` added to the following introspection tables:
  - `pg_am`
  - `pg_attribute`
  - `pg_class`
  - `pg_constraint`
  - `pg_description`
  - `pg_extension`
  - `pg_index`
  - `pg_namespace`
  - `pg_proc`
  - `pg_type`
- `pg_namespace`'s `nspacl` field now has correct type (`List(Utf8)`)
- `AGE` function now accepts values of type `xid`
- `"char"` casts are being replaced with `text` casts
- `xid` casts are being replaced with `unsigned int` casts
- New `PgType`s added:
  - `[ARRAY]PGAM`
  - `[ARRAY]PGCAST`
  - `[ARRAY]PGEVENTTRIGGER`
  - `[ARRAY]PGEXTENSION`
  - `[ARRAY]PGFOREIGNDATAWRAPPER`
  - `[ARRAY]PGFOREIGNSERVER`
  - `[ARRAY]PGLANGUAGE`
- `CTID` system field is replaced with `NULL` in data queries

Signed-off-by: Alex Qyoun-ae <[email protected]>

Author: MazterQyou

packages/cubejs-backend-native/Cargo.lock

@@ -10,14 +10,14 @@ homepage = "https://cube.dev"
 
 [dependencies]
 arc-swap = "1"
-datafusion = { git = 'https://github.com/cube-js/arrow-datafusion.git', rev = "7803f7e1cb8fa3d7b25451de35b8765e1ce4a3f6", default-features = false, features = [
+datafusion = { git = 'https://github.com/cube-js/arrow-datafusion.git', rev = "016c22f74b82f241fc01abd205020cb52b6c911e", 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 = "34f22de680caa5fe586def5b336d56efe43c8cc4" }
+sqlparser = { git = 'https://github.com/cube-js/sqlparser-rs.git', rev = "e14d5bf45367edd8679cbc15ccee56693da8e4fb" }
 base64 = "0.13.0"
 tokio = { version = "^1.35", features = ["full", "rt", "tracing"] }
 serde = { version = "^1.0", features = ["derive"] }

rust/cubenativeutils/Cargo.lock

@@ -1,6 +1,5 @@
 ---
 source: cubesql/e2e/tests/postgres.rs
-assertion_line: 316
 expression: "self.print_query_result(res, true, false).await"
 ---
 oid type: 23 (int4)
@@ -36,4 +35,5 @@ relminmxid type: 23 (int4)
 relacl type: 25 (text)
 reloptions type: 25 (text)
 relpartbound type: 25 (text)
+xmin type: 23 (int4)
 relhasoids type: 16 (bool)

rust/cubesql/Cargo.lock

@@ -1,6 +1,5 @@
 ---
 source: cubesql/e2e/tests/postgres.rs
-assertion_line: 316
 expression: "self.print_query_result(res, true, false).await"
 ---
 oid type: 23 (int4)
@@ -33,3 +32,4 @@ probin type: 25 (text)
 prosqlbody type: 25 (text)
 proconfig type: 25 (text)
 proacl type: 25 (text)
+xmin type: 23 (int4)

rust/cubesql/cubesql/Cargo.toml

@@ -1,6 +1,5 @@
 ---
 source: cubesql/e2e/tests/postgres.rs
-assertion_line: 316
 expression: "self.print_query_result(res, true, false).await"
 ---
 oid type: 23 (int4)
@@ -35,3 +34,4 @@ typcollation type: 25 (text)
 typdefaultbin type: 25 (text)
 typdefault type: 25 (text)
 typacl type: 25 (text)
+xmin type: 23 (int4)

rust/cubesql/cubesql/e2e/tests/snapshots/e2e__tests__postgres__system_pg_catalog.pg_class.snap

@@ -17,16 +17,21 @@ use super::information_schema::postgres::{
     InfoSchemaSqlImplementationInfoProvider as PostgresInfoSchemaSqlImplementationInfoProvider,
     InfoSchemaSqlSizingProvider as PostgresInfoSchemaSqlSizingProvider,
     InfoSchemaTestingBlockingProvider, InfoSchemaTestingDatasetProvider, PgCatalogAmProvider,
-    PgCatalogAttrdefProvider, PgCatalogAttributeProvider, PgCatalogClassProvider,
+    PgCatalogAttrdefProvider, PgCatalogAttributeProvider, PgCatalogAuthMembersProvider,
+    PgCatalogAvailableExtensionVersionsProvider, PgCatalogCastProvider, PgCatalogClassProvider,
     PgCatalogConstraintProvider, PgCatalogDatabaseProvider, PgCatalogDependProvider,
-    PgCatalogDescriptionProvider, PgCatalogEnumProvider, PgCatalogExtensionProvider,
-    PgCatalogIndexProvider, PgCatalogInheritsProvider, PgCatalogMatviewsProvider,
-    PgCatalogNamespaceProvider, PgCatalogPartitionedTableProvider, PgCatalogProcProvider,
-    PgCatalogRangeProvider, PgCatalogRolesProvider, PgCatalogSequenceProvider,
+    PgCatalogDescriptionProvider, PgCatalogEnumProvider, PgCatalogEventTriggerProvider,
+    PgCatalogExtensionProvider, PgCatalogForeignDataWrapperProvider,
+    PgCatalogForeignServerProvider, PgCatalogForeignTableProvider, PgCatalogIndexProvider,
+    PgCatalogInheritsProvider, PgCatalogLanguageProvider, PgCatalogLocksProvider,
+    PgCatalogMatviewsProvider, PgCatalogNamespaceProvider, PgCatalogOperatorProvider,
+    PgCatalogPartitionedTableProvider, PgCatalogProcProvider, PgCatalogRangeProvider,
+    PgCatalogRewriteProvider, PgCatalogRolesProvider, PgCatalogSequenceProvider,
     PgCatalogSettingsProvider, PgCatalogShdescriptionProvider, PgCatalogStatActivityProvider,
     PgCatalogStatUserTablesProvider, PgCatalogStatioUserTablesProvider, PgCatalogStatsProvider,
-    PgCatalogTableProvider, PgCatalogTypeProvider, PgCatalogUserProvider, PgCatalogViewsProvider,
-    PgPreparedStatementsProvider,
+    PgCatalogTableProvider, PgCatalogTablespaceProvider, PgCatalogTimezoneAbbrevsProvider,
+    PgCatalogTimezoneNamesProvider, PgCatalogTypeProvider, PgCatalogUserMappingProvider,
+    PgCatalogUserProvider, PgCatalogViewsProvider, PgPreparedStatementsProvider,
 };
 use crate::{
     compile::{
@@ -138,6 +143,36 @@ impl DatabaseProtocol {
             "pg_catalog.pg_stat_user_tables".to_string()
         } else if let Some(_) = any.downcast_ref::<PgCatalogShdescriptionProvider>() {
             "pg_catalog.pg_shdescription".to_string()
+        } else if let Some(_) = any.downcast_ref::<PgCatalogLocksProvider>() {
+            "pg_catalog.pg_locks".to_string()
+        } else if let Some(_) = any.downcast_ref::<PgCatalogTimezoneNamesProvider>() {
+            "pg_catalog.pg_timezone_names".to_string()
+        } else if let Some(_) = any.downcast_ref::<PgCatalogTimezoneAbbrevsProvider>() {
+            "pg_catalog.pg_timezone_abbrevs".to_string()
+        } else if let Some(_) = any.downcast_ref::<PgCatalogAuthMembersProvider>() {
+            "pg_catalog.pg_auth_members".to_string()
+        } else if let Some(_) = any.downcast_ref::<PgCatalogTablespaceProvider>() {
+            "pg_catalog.pg_tablespace".to_string()
+        } else if let Some(_) = any.downcast_ref::<PgCatalogEventTriggerProvider>() {
+            "pg_catalog.pg_event_trigger".to_string()
+        } else if let Some(_) = any.downcast_ref::<PgCatalogForeignDataWrapperProvider>() {
+            "pg_catalog.pg_foreign_data_wrapper".to_string()
+        } else if let Some(_) = any.downcast_ref::<PgCatalogForeignServerProvider>() {
+            "pg_catalog.pg_foreign_server".to_string()
+        } else if let Some(_) = any.downcast_ref::<PgCatalogUserMappingProvider>() {
+            "pg_catalog.pg_user_mapping".to_string()
+        } else if let Some(_) = any.downcast_ref::<PgCatalogAvailableExtensionVersionsProvider>() {
+            "pg_catalog.pg_available_extension_versions".to_string()
+        } else if let Some(_) = any.downcast_ref::<PgCatalogLanguageProvider>() {
+            "pg_catalog.pg_language".to_string()
+        } else if let Some(_) = any.downcast_ref::<PgCatalogCastProvider>() {
+            "pg_catalog.pg_cast".to_string()
+        } else if let Some(_) = any.downcast_ref::<PgCatalogForeignTableProvider>() {
+            "pg_catalog.pg_foreign_table".to_string()
+        } else if let Some(_) = any.downcast_ref::<PgCatalogOperatorProvider>() {
+            "pg_catalog.pg_operator".to_string()
+        } else if let Some(_) = any.downcast_ref::<PgCatalogRewriteProvider>() {
+            "pg_catalog.pg_rewrite".to_string()
         } else if let Some(_) = any.downcast_ref::<RedshiftPgExternalSchemaProvider>() {
             "pg_catalog.pg_external_schema".to_string()
         } else if let Some(_) = any.downcast_ref::<RedshiftSvvTablesTableProvider>() {
@@ -404,6 +439,31 @@ impl DatabaseProtocol {
                     )))
                 }
                 "pg_shdescription" => return Some(Arc::new(PgCatalogShdescriptionProvider::new())),
+                "pg_locks" => return Some(Arc::new(PgCatalogLocksProvider::new())),
+                "pg_timezone_names" => {
+                    return Some(Arc::new(PgCatalogTimezoneNamesProvider::new()))
+                }
+                "pg_timezone_abbrevs" => {
+                    return Some(Arc::new(PgCatalogTimezoneAbbrevsProvider::new()))
+                }
+                "pg_auth_members" => return Some(Arc::new(PgCatalogAuthMembersProvider::new())),
+                "pg_tablespace" => return Some(Arc::new(PgCatalogTablespaceProvider::new())),
+                "pg_event_trigger" => return Some(Arc::new(PgCatalogEventTriggerProvider::new())),
+                "pg_foreign_data_wrapper" => {
+                    return Some(Arc::new(PgCatalogForeignDataWrapperProvider::new()))
+                }
+                "pg_foreign_server" => {
+                    return Some(Arc::new(PgCatalogForeignServerProvider::new()))
+                }
+                "pg_user_mapping" => return Some(Arc::new(PgCatalogUserMappingProvider::new())),
+                "pg_available_extension_versions" => {
+                    return Some(Arc::new(PgCatalogAvailableExtensionVersionsProvider::new()))
+                }
+                "pg_language" => return Some(Arc::new(PgCatalogLanguageProvider::new())),
+                "pg_cast" => return Some(Arc::new(PgCatalogCastProvider::new())),
+                "pg_foreign_table" => return Some(Arc::new(PgCatalogForeignTableProvider::new())),
+                "pg_operator" => return Some(Arc::new(PgCatalogOperatorProvider::new())),
+                "pg_rewrite" => return Some(Arc::new(PgCatalogRewriteProvider::new())),
                 "pg_external_schema" => {
                     return Some(Arc::new(RedshiftPgExternalSchemaProvider::new()))
                 }