Merge remote-tracking branch 'upstream/master' into parallelize-file-cache-metadata-download

src/Analyzer/Passes/QueryAnalysisPass.cpp

@@ -1918,6 +1918,9 @@ void QueryAnalyzer::evaluateScalarSubqueryIfNeeded(QueryTreeNodePtr & node, Iden
         subquery_settings.max_result_rows = 1;
         subquery_settings.extremes = false;
         subquery_context->setSettings(subquery_settings);
+        /// When execute `INSERT INTO t WITH ... SELECT ...`, it may lead to `Unknown columns`
+        /// exception with this settings enabled(https://github.com/ClickHouse/ClickHouse/issues/52494).
+        subquery_context->setSetting("use_structure_from_insertion_table_in_table_functions", false);
 
         auto options = SelectQueryOptions(QueryProcessingStage::Complete, scope.subquery_depth, true /*is_subquery*/);
         auto interpreter = std::make_unique<InterpreterSelectQueryAnalyzer>(node->toAST(), subquery_context, options);

src/Functions/JSONPath/Parsers/ParserJSONPathMemberAccess.cpp

@@ -58,6 +58,10 @@ bool ParserJSONPathMemberAccess::parseImpl(Pos & pos, ASTPtr & node, Expected &
             member_name = std::make_shared<ASTIdentifier>(String(last_begin, pos->end));
             ++pos;
         }
+        else if (!pos.isValid() && pos->type == TokenType::EndOfStream)
+        {
+            member_name = std::make_shared<ASTIdentifier>(String(last_begin, last_end));
+        }
         else
         {
             return false;

src/Interpreters/ExecuteScalarSubqueriesVisitor.cpp

@@ -77,6 +77,10 @@ static auto getQueryInterpreter(const ASTSubquery & subquery, ExecuteScalarSubqu
     subquery_settings.max_result_rows = 1;
     subquery_settings.extremes = false;
     subquery_context->setSettings(subquery_settings);
+
+    /// When execute `INSERT INTO t WITH ... SELECT ...`, it may lead to `Unknown columns`
+    /// exception with this settings enabled(https://github.com/ClickHouse/ClickHouse/issues/52494).
+    subquery_context->getQueryContext()->setSetting("use_structure_from_insertion_table_in_table_functions", false);
     if (!data.only_analyze && subquery_context->hasQueryContext())
     {
         /// Save current cached scalars in the context before analyzing the query

tests/queries/0_stateless/01889_sql_json_functions.reference

@@ -77,6 +77,10 @@ SELECT JSON_QUERY('{"array":[[0, 1, 2, 3, 4, 5], [0, -1, -2, -3, -4, -5]]}', '$.
 [0, 1, 4, 0, -1, -4]
 SELECT JSON_QUERY('{"1key":1}', '$.1key');
 [1]
+SELECT JSON_QUERY('{"123":1}', '$.123');
+[1]
+SELECT JSON_QUERY('{"123":1}', '$[123]');
+
 SELECT JSON_QUERY('{"hello":1}', '$[hello]');
 [1]
 SELECT JSON_QUERY('{"hello":1}', '$["hello"]');

tests/queries/0_stateless/01889_sql_json_functions.sql

@@ -43,6 +43,8 @@ SELECT JSON_QUERY( '{hello:{"world":"!"}}}', '$.hello'); -- invalid json => defa
 SELECT JSON_QUERY('', '$.hello');
 SELECT JSON_QUERY('{"array":[[0, 1, 2, 3, 4, 5], [0, -1, -2, -3, -4, -5]]}', '$.array[*][0 to 2, 4]');
 SELECT JSON_QUERY('{"1key":1}', '$.1key');
+SELECT JSON_QUERY('{"123":1}', '$.123');
+SELECT JSON_QUERY('{"123":1}', '$[123]');
 SELECT JSON_QUERY('{"hello":1}', '$[hello]');
 SELECT JSON_QUERY('{"hello":1}', '$["hello"]');
 SELECT JSON_QUERY('{"hello":1}', '$[\'hello\']');

tests/queries/0_stateless/02843_insertion_table_schema_infer.reference

@@ -0,0 +1,3 @@
+user127	1
+user405	1
+user902	1

tests/queries/0_stateless/02843_insertion_table_schema_infer.sh

@@ -0,0 +1,16 @@
+#!/usr/bin/env bash
+# Tags: no-parallel, no-fasttest
+
+set -e
+
+CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)
+# shellcheck source=../shell_config.sh
+. "$CUR_DIR"/../shell_config.sh
+
+DATA_DIR=$CUR_DIR/data_tsv
+
+$CLICKHOUSE_LOCAL --multiquery \
+"CREATE VIEW users AS SELECT * FROM file('$DATA_DIR/mock_data.tsv', TSVWithNamesAndTypes); 
+ CREATE TABLE users_output (name String, tag UInt64)ENGINE = Memory;
+ INSERT INTO users_output WITH (SELECT groupUniqArrayArray(mapKeys(Tags)) FROM users) AS unique_tags SELECT UserName AS name, length(unique_tags) AS tag FROM users;
+ SELECT * FROM users_output;"

tests/queries/0_stateless/02862_sorted_distinct_sparse_fix.reference

@@ -10,3 +10,4 @@ set max_threads=1;
 select trimLeft(explain) from (explain pipeline SELECT DISTINCT id, v FROM t_sparse_distinct) where explain ilike '%DistinctSortedChunkTransform%';
 DistinctSortedChunkTransform
 SELECT DISTINCT id, v FROM t_sparse_distinct format Null;
+DROP TABLE t_sparse_distinct;

tests/queries/0_stateless/data_tsv/mock_data.tsv

@@ -0,0 +1,5 @@
+UserName	Age	Tags
+String	Int8	Map(String, UInt64)
+user127	20	{'test': 123}
+user405	43	{'test': 123}
+user902	43	{'test': 123}