Merge branch 'master' into fix_flaky_integration_tests

docs/en/faq/general/how-do-i-contribute-code-to-clickhouse.md

@@ -0,0 +1,15 @@
+---
+title: How do I contribute code to ClickHouse?
+toc_hidden: true
+toc_priority: 120
+---
+
+# How do I contribute code to ClickHouse? {#how-do-i-contribute-code-to-clickhouse}
+
+ClickHouse is an open-source project [developed on GitHub](https://github.com/ClickHouse/ClickHouse).
+
+As customary, contribution instructions are published in [CONTRIBUTING.md](https://github.com/ClickHouse/ClickHouse/blob/master/CONTRIBUTING.md) file in the root of the source code repository.
+
+If you want to suggest a substantial change to ClickHouse, consider [opening a GitHub issue](https://github.com/ClickHouse/ClickHouse/issues/new/choose) explaining what you want to do, to discuss it with maintainers and community first. [Examples of such RFC issues](https://github.com/ClickHouse/ClickHouse/issues?q=is%3Aissue+is%3Aopen+rfc).
+
+If your contributions are security related, please check out [our security policy](https://github.com/ClickHouse/ClickHouse/security/policy/) too.

docs/en/faq/general/index.md

@@ -17,6 +17,7 @@ Questions:
 -   [What is OLAP?](../../faq/general/olap.md)
 -   [What is a columnar database?](../../faq/general/columnar-database.md)
 -   [Why not use something like MapReduce?](../../faq/general/mapreduce.md)
+-   [How do I contribute code to ClickHouse?](../../faq/general/how-do-i-contribute-code-to-clickhouse.md)
 
 !!! info "Don’t see what you were looking for?"
     Check out [other F.A.Q. categories](../../faq/index.md) or browse around main documentation articles found in the left sidebar.

docs/en/interfaces/third-party/client-libraries.md

@@ -6,7 +6,7 @@ toc_title: Client Libraries
 # Client Libraries from Third-party Developers {#client-libraries-from-third-party-developers}
 
 !!! warning "Disclaimer"
-    Yandex does **not** maintain the libraries listed below and hasn’t done any extensive testing to ensure their quality.
+    ClickHouse Inc does **not** maintain the libraries listed below and hasn’t done any extensive testing to ensure their quality.
 
 -   Python
     -   [infi.clickhouse_orm](https://github.com/Infinidat/infi.clickhouse_orm)

docs/en/operations/settings/settings.md

@@ -817,9 +817,19 @@ If the number of rows to be read from a file of a [MergeTree](../../engines/tabl
 
 Possible values:
 
--   Any positive integer.
+-   Positive integer.
 
-Default value: 163840.
+Default value: `163840`.
+
+## merge_tree_min_rows_for_concurrent_read_for_remote_filesystem {#merge-tree-min-rows-for-concurrent-read-for-remote-filesystem}
+
+The minimum number of lines to read from one file before [MergeTree](../../engines/table-engines/mergetree-family/mergetree.md) engine can parallelize reading, when reading from remote filesystem.
+
+Possible values:
+
+-   Positive integer.
+
+Default value: `163840`.
 
 ## merge_tree_min_bytes_for_concurrent_read {#setting-merge-tree-min-bytes-for-concurrent-read}
 
@@ -827,9 +837,19 @@ If the number of bytes to read from one file of a [MergeTree](../../engines/tabl
 
 Possible value:
 
--   Any positive integer.
+-   Positive integer.
 
-Default value: 251658240.
+Default value: `251658240`.
+
+## merge_tree_min_bytes_for_concurrent_read_for_remote_filesystem {#merge-tree-min-bytes-for-concurrent-read-for-remote-filesystem}
+
+The minimum number of bytes to read from one file before [MergeTree](../../engines/table-engines/mergetree-family/mergetree.md) engine can parallelize reading, when reading from remote filesystem.
+
+Possible values:
+
+-   Positive integer.
+
+Default value: `251658240`.
 
 ## merge_tree_min_rows_for_seek {#setting-merge-tree-min-rows-for-seek}
 

docs/ja/faq/general/how-do-i-contribute-code-to-clickhouse.md

@@ -0,0 +1 @@
+../../../en/faq/general/how-do-i-contribute-code-to-clickhouse.md

docs/ru/faq/general/how-do-i-contribute-code-to-clickhouse.md

@@ -0,0 +1 @@
+../../../en/faq/general/how-do-i-contribute-code-to-clickhouse.md

docs/ru/operations/settings/settings.md

@@ -761,9 +761,20 @@ ClickHouse может парсить только базовый формат `Y
 
 Возможные значения:
 
--   Любое положительное целое число.
+-   Положительное целое число.
 
-Значение по умолчанию: 163840.
+Значение по умолчанию: `163840`.
+
+
+## merge_tree_min_rows_for_concurrent_read_for_remote_filesystem {#merge-tree-min-rows-for-concurrent-read-for-remote-filesystem}
+
+Минимальное количество строк для чтения из одного файла, прежде чем движок [MergeTree](../../engines/table-engines/mergetree-family/mergetree.md) может выполнять параллельное чтение из удаленной файловой системы.
+
+Возможные значения:
+
+-   Положительное целое число.
+
+Значение по умолчанию: `163840`.
 
 ## merge_tree_min_bytes_for_concurrent_read {#setting-merge-tree-min-bytes-for-concurrent-read}
 
@@ -773,7 +784,17 @@ ClickHouse может парсить только базовый формат `Y
 
 -   Положительное целое число.
 
-Значение по умолчанию: 251658240.
+Значение по умолчанию: `251658240`.
+
+## merge_tree_min_bytes_for_concurrent_read_for_remote_filesystem {#merge-tree-min-bytes-for-concurrent-read-for-remote-filesystem}
+
+Минимальное количество байтов для чтения из одного файла, прежде чем движок [MergeTree](../../engines/table-engines/mergetree-family/mergetree.md) может выполнять параллельное чтение из удаленной файловой системы.
+
+Возможное значение:
+
+-   Положительное целое число.
+
+Значение по умолчанию: `251658240`.
 
 ## merge_tree_min_rows_for_seek {#setting-merge-tree-min-rows-for-seek}
 

docs/zh/faq/general/how-do-i-contribute-code-to-clickhouse.md

@@ -0,0 +1 @@
+../../../en/faq/general/how-do-i-contribute-code-to-clickhouse.md

src/Storages/PostgreSQL/PostgreSQLReplicationHandler.cpp

@@ -278,9 +278,8 @@ ASTPtr PostgreSQLReplicationHandler::getCreateNestedTableQuery(StorageMaterializ
 {
     postgres::Connection connection(connection_info);
     pqxx::nontransaction tx(connection.getRef());
-    auto table_structure = std::make_unique<PostgreSQLTableStructure>(fetchPostgreSQLTableStructure(tx, table_name, postgres_schema, true, true, true));
-    if (!table_structure)
-        throw Exception(ErrorCodes::LOGICAL_ERROR, "Failed to get PostgreSQL table structure");
+    auto [postgres_table_schema, postgres_table_name] = getSchemaAndTableName(table_name);
+    auto table_structure = std::make_unique<PostgreSQLTableStructure>(fetchPostgreSQLTableStructure(tx, postgres_table_name, postgres_table_schema, true, true, true));
 
     auto table_override = tryGetTableOverride(current_database_name, table_name);
     return storage->getCreateNestedTableQuery(std::move(table_structure), table_override ? table_override->as<ASTTableOverride>() : nullptr);
@@ -516,17 +515,25 @@ void PostgreSQLReplicationHandler::dropPublication(pqxx::nontransaction & tx)
 
 void PostgreSQLReplicationHandler::addTableToPublication(pqxx::nontransaction & ntx, const String & table_name)
 {
-    std::string query_str = fmt::format("ALTER PUBLICATION {} ADD TABLE ONLY {}", publication_name, doubleQuoteString(table_name));
+    std::string query_str = fmt::format("ALTER PUBLICATION {} ADD TABLE ONLY {}", publication_name, doubleQuoteWithSchema(table_name));
     ntx.exec(query_str);
-    LOG_TRACE(log, "Added table `{}` to publication `{}`", table_name, publication_name);
+    LOG_TRACE(log, "Added table {} to publication `{}`", doubleQuoteWithSchema(table_name), publication_name);
 }
 
 
 void PostgreSQLReplicationHandler::removeTableFromPublication(pqxx::nontransaction & ntx, const String & table_name)
 {
-    std::string query_str = fmt::format("ALTER PUBLICATION {} DROP TABLE ONLY {}", publication_name, doubleQuoteString(table_name));
-    ntx.exec(query_str);
-    LOG_TRACE(log, "Removed table `{}` from publication `{}`", table_name, publication_name);
+    try
+    {
+        std::string query_str = fmt::format("ALTER PUBLICATION {} DROP TABLE ONLY {}", publication_name, doubleQuoteWithSchema(table_name));
+        ntx.exec(query_str);
+        LOG_TRACE(log, "Removed table `{}` from publication `{}`", doubleQuoteWithSchema(table_name), publication_name);
+    }
+    catch (const pqxx::undefined_table &)
+    {
+        /// Removing table from replication must succeed even if table does not exist in PostgreSQL.
+        LOG_WARNING(log, "Did not remove table {} from publication, because table does not exist in PostgreSQL", doubleQuoteWithSchema(table_name), publication_name);
+    }
 }
 
 

src/Storages/StorageBuffer.cpp

@@ -455,10 +455,8 @@ static void appendBlock(const Block & from, Block & to)
     size_t rows = from.rows();
     size_t bytes = from.bytes();
 
-    CurrentMetrics::add(CurrentMetrics::StorageBufferRows, rows);
-    CurrentMetrics::add(CurrentMetrics::StorageBufferBytes, bytes);
-
     size_t old_rows = to.rows();
+    size_t old_bytes = to.bytes();
 
     MutableColumnPtr last_col;
     try
@@ -468,6 +466,8 @@ static void appendBlock(const Block & from, Block & to)
         if (to.rows() == 0)
         {
             to = from;
+            CurrentMetrics::add(CurrentMetrics::StorageBufferRows, rows);
+            CurrentMetrics::add(CurrentMetrics::StorageBufferBytes, bytes);
         }
         else
         {
@@ -480,6 +480,8 @@ static void appendBlock(const Block & from, Block & to)
 
                 to.getByPosition(column_no).column = std::move(last_col);
             }
+            CurrentMetrics::add(CurrentMetrics::StorageBufferRows, rows);
+            CurrentMetrics::add(CurrentMetrics::StorageBufferBytes, to.bytes() - old_bytes);
         }
     }
     catch (...)

tests/ci/cancel_and_rerun_workflow_lambda/app.py

@@ -11,7 +11,6 @@ import boto3
 NEED_RERUN_OR_CANCELL_WORKFLOWS = {
     13241696, # PR
     15834118, # Docs
-    15522500, # MasterCI
     15516108, # ReleaseCI
     15797242, # BackportPR
 }
@@ -86,10 +85,23 @@ WorkflowDescription = namedtuple('WorkflowDescription',
 def get_workflows_description_for_pull_request(pull_request_event):
     head_branch = pull_request_event['head']['ref']
     print("PR", pull_request_event['number'], "has head ref", head_branch)
-    workflows = _exec_get_with_retry(API_URL + f"/actions/runs?branch={head_branch}")
+    workflows_data = []
+    workflows = _exec_get_with_retry(API_URL + f"/actions/runs?branch={head_branch}&event=pull_request&page=1")
+    workflows_data += workflows['workflow_runs']
+    i = 2
+    while len(workflows['workflow_runs']) > 0:
+        workflows = _exec_get_with_retry(API_URL + f"/actions/runs?branch={head_branch}&event=pull_request&page={i}")
+        workflows_data += workflows['workflow_runs']
+        i += 1
+        if i > 30:
+            print("Too many workflows found")
+            break
+
     workflow_descriptions = []
-    for workflow in workflows['workflow_runs']:
-        if workflow['workflow_id'] in NEED_RERUN_OR_CANCELL_WORKFLOWS:
+    for workflow in workflows_data:
+        # unfortunately we cannot filter workflows from forks in request to API so doing it manually
+        if (workflow['head_repository']['full_name'] == pull_request_event['head']['repo']['full_name']
+            and workflow['workflow_id'] in NEED_RERUN_OR_CANCELL_WORKFLOWS):
             workflow_descriptions.append(WorkflowDescription(
                 run_id=workflow['id'],
                 status=workflow['status'],

tests/ci/keeper_jepsen_check.py

@@ -120,6 +120,8 @@ if __name__ == "__main__":
 
     pr_info = PRInfo()
 
+    logging.info("Start at PR number %s, commit sha %s labels %s", pr_info.number, pr_info.sha, pr_info.labels)
+
     if pr_info.number != 0 and 'jepsen-test' not in pr_info.labels():
         logging.info("Not jepsen test label in labels list, skipping")
         sys.exit(0)

tests/integration/test_input_format_parallel_parsing_memory_tracking/configs/conf.xml

@@ -1,4 +1,23 @@
 <clickhouse>
-    <!-- make it fail earlier -->
-    <max_server_memory_usage>3000000000</max_server_memory_usage> <!-- 3GB -->
+    <!-- To make it fail earlier, we will limit max_server_memory_usage explicitly.
+
+         Also note, that usually it is enough 3Gi,
+         but TSan uses 2.8+- GiB of RAM w/o just at start,
+         so this limit had been increased to 4GB
+    -->
+    <max_server_memory_usage>4000000000</max_server_memory_usage>
+
+    <query_thread_log remove="remove"/>
+    <query_log remove="remove" />
+    <query_views_log remove="remove" />
+    <metric_log remove="remove"/>
+    <text_log remove="remove"/>
+    <trace_log remove="remove"/>
+    <asynchronous_metric_log remove="remove" />
+    <session_log remove="remove" />
+    <part_log remove="remove" />
+    <crash_log remove="remove" />
+    <opentelemetry_span_log remove="remove" />
+    <!-- just in case it will be enabled by default -->
+    <zookeeper_log remove="remove" />
 </clickhouse>

tests/integration/test_input_format_parallel_parsing_memory_tracking/test.py

@@ -24,19 +24,13 @@ def start_cluster():
 
 
 # max_memory_usage_for_user cannot be used, since the memory for user accounted
-# correctly, only total is not
+# correctly, only total is not (it is set via conf.xml)
 def test_memory_tracking_total():
-    if instance.is_built_with_memory_sanitizer() or instance.is_built_with_thread_sanitizer() or instance.is_built_with_address_sanitizer():
-        print("Server built with sanitizer and memory consumption can be unpredictable, skipping test")
-    else:
-        instance.query('''
-            CREATE TABLE null (row String) ENGINE=Null;
-        ''')
-        instance.exec_in_container(['bash', '-c',
-                                    'clickhouse local -q "SELECT arrayStringConcat(arrayMap(x->toString(cityHash64(x)), range(1000)), \' \') from numbers(10000)" > data.json'])
-        for it in range(0, 20):
-            # the problem can be triggered only via HTTP,
-            # since clickhouse-client parses the data by itself.
-            assert instance.exec_in_container(['curl', '--silent', '--show-error', '--data-binary', '@data.json',
-                                               'http://127.1:8123/?query=INSERT%20INTO%20null%20FORMAT%20TSV']) == '', 'Failed on {} iteration'.format(
-                it)
+    instance.query('CREATE TABLE null (row String) ENGINE=Null')
+    instance.exec_in_container(['bash', '-c',
+                                'clickhouse local -q "SELECT arrayStringConcat(arrayMap(x->toString(cityHash64(x)), range(1000)), \' \') from numbers(10000)" > data.json'])
+    for it in range(0, 20):
+        # the problem can be triggered only via HTTP,
+        # since clickhouse-client parses the data by itself.
+        assert instance.exec_in_container(['curl', '--silent', '--show-error', '--data-binary', '@data.json',
+                                           'http://127.1:8123/?query=INSERT%20INTO%20null%20FORMAT%20TSV']) == '', f'Failed on {it} iteration'

tests/integration/test_postgresql_replica_database_engine_2/test.py

@@ -178,7 +178,7 @@ def assert_number_of_columns(expected, table_name, database_name='test_database'
 def check_tables_are_synchronized(table_name, order_by='key', postgres_database='postgres_database', materialized_database='test_database', schema_name=''):
     assert_nested_table_is_created(table_name, materialized_database, schema_name)
 
-    print("Checking table is synchronized:", table_name)
+    print(f"Checking table is synchronized. Table name: {table_name}, table schema: {schema_name}")
     expected = instance.query('select * from {}.{} order by {};'.format(postgres_database, table_name, order_by))
     if len(schema_name) == 0:
         result = instance.query('select * from {}.{} order by {};'.format(materialized_database, table_name, order_by))
@@ -356,6 +356,11 @@ def test_remove_table_from_replication(started_cluster):
     for i in range(NUM_TABLES):
         cursor.execute('drop table if exists postgresql_replica_{};'.format(i))
 
+    # Removing from replication table which does not exist in PostgreSQL must be ok.
+    instance.query('DETACH TABLE test_database.postgresql_replica_0');
+    assert instance.contains_in_log("from publication, because table does not exist in PostgreSQL")
+    drop_materialized_db()
+
 
 def test_predefined_connection_configuration(started_cluster):
     drop_materialized_db()
@@ -379,6 +384,7 @@ def test_database_with_single_non_default_schema(started_cluster):
 
     NUM_TABLES=5
     schema_name = 'test_schema'
+    materialized_db = 'test_database'
     clickhouse_postgres_db = 'postgres_database_with_schema'
     global insert_counter
     insert_counter = 0
@@ -430,6 +436,14 @@ def test_database_with_single_non_default_schema(started_cluster):
     instance.query(f"INSERT INTO {clickhouse_postgres_db}.postgresql_replica_{altered_table} SELECT number, number, number from numbers(5000, 1000)")
     assert_number_of_columns(3, f'postgresql_replica_{altered_table}')
     check_tables_are_synchronized(f"postgresql_replica_{altered_table}", postgres_database=clickhouse_postgres_db);
+
+    print('DETACH-ATTACH')
+    detached_table_name = "postgresql_replica_1"
+    instance.query(f"DETACH TABLE {materialized_db}.{detached_table_name}")
+    assert not instance.contains_in_log("from publication, because table does not exist in PostgreSQL")
+    instance.query(f"ATTACH TABLE {materialized_db}.{detached_table_name}")
+    check_tables_are_synchronized(detached_table_name, postgres_database=clickhouse_postgres_db);
+
     drop_materialized_db()
 
 
@@ -440,6 +454,7 @@ def test_database_with_multiple_non_default_schemas_1(started_cluster):
     NUM_TABLES = 5
     schema_name = 'test_schema'
     clickhouse_postgres_db = 'postgres_database_with_schema'
+    materialized_db = 'test_database'
     publication_tables = ''
     global insert_counter
     insert_counter = 0
@@ -494,6 +509,15 @@ def test_database_with_multiple_non_default_schemas_1(started_cluster):
     instance.query(f"INSERT INTO {clickhouse_postgres_db}.postgresql_replica_{altered_table} SELECT number, number, number from numbers(5000, 1000)")
     assert_number_of_columns(3, f'{schema_name}.postgresql_replica_{altered_table}')
     check_tables_are_synchronized(f"postgresql_replica_{altered_table}", schema_name=schema_name, postgres_database=clickhouse_postgres_db);
+
+    print('DETACH-ATTACH')
+    detached_table_name = "postgresql_replica_1"
+    instance.query(f"DETACH TABLE {materialized_db}.`{schema_name}.{detached_table_name}`")
+    assert not instance.contains_in_log("from publication, because table does not exist in PostgreSQL")
+    instance.query(f"ATTACH TABLE {materialized_db}.`{schema_name}.{detached_table_name}`")
+    assert_show_tables("test_schema.postgresql_replica_0\ntest_schema.postgresql_replica_1\ntest_schema.postgresql_replica_2\ntest_schema.postgresql_replica_3\ntest_schema.postgresql_replica_4\n")
+    check_tables_are_synchronized(detached_table_name, schema_name=schema_name, postgres_database=clickhouse_postgres_db);
+
     drop_materialized_db()
 
 
@@ -504,6 +528,7 @@ def test_database_with_multiple_non_default_schemas_2(started_cluster):
     NUM_TABLES = 2
     schemas_num = 2
     schema_list = 'schema0, schema1'
+    materialized_db = 'test_database'
     global insert_counter
     insert_counter = 0
 
@@ -557,11 +582,23 @@ def test_database_with_multiple_non_default_schemas_2(started_cluster):
     print('ALTER')
     altered_schema = random.randint(0, schemas_num-1)
     altered_table = random.randint(0, NUM_TABLES-1)
+    clickhouse_postgres_db = f'clickhouse_postgres_db{altered_schema}'
     cursor.execute(f"ALTER TABLE schema{altered_schema}.postgresql_replica_{altered_table} ADD COLUMN value2 integer")
 
     instance.query(f"INSERT INTO clickhouse_postgres_db{altered_schema}.postgresql_replica_{altered_table} SELECT number, number, number from numbers(1000 * {insert_counter}, 1000)")
     assert_number_of_columns(3, f'schema{altered_schema}.postgresql_replica_{altered_table}')
-    check_tables_are_synchronized(f"postgresql_replica_{altered_table}", schema_name=schema_name, postgres_database=clickhouse_postgres_db);
+    check_tables_are_synchronized(f"postgresql_replica_{altered_table}", schema_name=f"schema{altered_schema}", postgres_database=clickhouse_postgres_db);
+
+    print('DETACH-ATTACH')
+    detached_table_name = "postgresql_replica_1"
+    detached_table_schema = "schema0"
+    clickhouse_postgres_db = f'clickhouse_postgres_db0'
+    instance.query(f"DETACH TABLE {materialized_db}.`{detached_table_schema}.{detached_table_name}`")
+    assert not instance.contains_in_log("from publication, because table does not exist in PostgreSQL")
+    instance.query(f"ATTACH TABLE {materialized_db}.`{detached_table_schema}.{detached_table_name}`")
+    assert_show_tables("schema0.postgresql_replica_0\nschema0.postgresql_replica_1\nschema1.postgresql_replica_0\nschema1.postgresql_replica_1\n")
+    check_tables_are_synchronized(f"postgresql_replica_{altered_table}", schema_name=detached_table_schema, postgres_database=clickhouse_postgres_db);
+
     drop_materialized_db()
 
 

tests/integration/test_system_metrics/test.py

@@ -59,3 +59,32 @@ def test_readonly_metrics(start_cluster):
         node1.query("ATTACH TABLE test.test_table")
         assert_eq_with_retry(node1, "SELECT value FROM system.metrics WHERE metric = 'ReadonlyReplica'", "0\n", retry_count=300, sleep_time=1)
 
+#For LowCardinality-columns, the bytes for N rows is not N*size of 1 row.
+def test_metrics_storage_buffer_size(start_cluster):
+    node1.query('''
+        CREATE TABLE test.test_mem_table
+        (
+            `str` LowCardinality(String)
+        )
+        ENGINE = Memory;
+
+        CREATE TABLE test.buffer_table
+        (
+            `str` LowCardinality(String)
+        )
+        ENGINE = Buffer('test', 'test_mem_table', 1, 600, 600, 1000, 100000, 100000, 10000000);
+    ''')
+
+    #before flush
+    node1.query("INSERT INTO test.buffer_table VALUES('hello');")
+    assert node1.query("SELECT value FROM system.metrics WHERE metric = 'StorageBufferRows'") == "1\n"
+    assert node1.query("SELECT value FROM system.metrics WHERE metric = 'StorageBufferBytes'") == "24\n"
+
+    node1.query("INSERT INTO test.buffer_table VALUES('hello');")
+    assert node1.query("SELECT value FROM system.metrics WHERE metric = 'StorageBufferRows'") == "2\n"
+    assert node1.query("SELECT value FROM system.metrics WHERE metric = 'StorageBufferBytes'") == "25\n"
+
+    #flush
+    node1.query("OPTIMIZE TABLE test.buffer_table")
+    assert node1.query("SELECT value FROM system.metrics WHERE metric = 'StorageBufferRows'") == "0\n"
+    assert node1.query("SELECT value FROM system.metrics WHERE metric = 'StorageBufferBytes'") == "0\n"

website/benchmark/hardware/results/xeon_gold_6266.json

@@ -1,7 +1,7 @@
 [
     {
-        "system":       "Xeon Gold 6266C, 3GHz, 4vCPU",
-        "system_full":  "Xeon Gold 6266C, 3GHz, 4vCPU, 16GiB RAM, vda1 40GB",
+        "system":       "Huawei Cloud c6.xlarge.4, 4vCPUs, 16 GiB"
+        "system_full":  "Huawei Cloud c6.xlarge.4, Xeon Gold 6266C, 3GHz, 4vCPU, 16GiB RAM, vda1 40GB",
         "cpu_vendor":   "Intel",
         "cpu_model":    "Xeon Gold 6266C",
         "time":         "2021-12-23 00:00:00",