0.8.0 Release (#400)

* initial new JSON spiking

* Basic Variant type reads

* Checkpoint for JSON data type reads

* Checkpoint for JSON data type reads

* Checkpoint for JSON data type reads

* Some lint and test cleanup

* Exclude Python 3.13 testing

* Clean up TLS test configuration

* Fix tls test lint

* Cloud test fixes

* Fix lint

* Add HTTP streaming buffer

* Improve variant type handling

* Add new mechanism for DateTime64 binding

* Add tls_mode client parameter, update changelog

.docker/clickhouse/single_node/config.xml

@@ -37,24 +37,5 @@
         <table>session_log</table>
     </session_log>
 
-    <http_options_response>
-        <header>
-            <name>Access-Control-Allow-Origin</name>
-            <value>*</value>
-        </header>
-        <header>
-            <name>Access-Control-Allow-Headers</name>
-            <value>accept, origin, x-requested-with, content-type, authorization</value>
-        </header>
-        <header>
-            <name>Access-Control-Allow-Methods</name>
-            <value>POST, GET, OPTIONS</value>
-        </header>
-        <header>
-            <name>Access-Control-Max-Age</name>
-            <value>86400</value>
-        </header>
-    </http_options_response>
-
     <custom_settings_prefixes>SQL_</custom_settings_prefixes>
 </clickhouse>

.docker/clickhouse/single_node_tls/Dockerfile

@@ -1,4 +1,4 @@
-FROM clickhouse/clickhouse-server:24.3-alpine
+FROM clickhouse/clickhouse-server:24.8-alpine
 COPY .docker/clickhouse/single_node_tls/certificates /etc/clickhouse-server/certs
 RUN chown clickhouse:clickhouse -R /etc/clickhouse-server/certs \
     && chmod 600 /etc/clickhouse-server/certs/* \

.docker/clickhouse/single_node_tls/config.xml

@@ -43,4 +43,6 @@
         <partition_by>toYYYYMM(event_date)</partition_by>
         <flush_interval_milliseconds>1000</flush_interval_milliseconds>
     </query_log>
+
+    <custom_settings_prefixes>SQL_</custom_settings_prefixes>
 </clickhouse>

.github/workflows/clickhouse_ci.yml

@@ -28,7 +28,7 @@ jobs:
       - name: "Add distribution info"  #  This lets SQLAlchemy find entry points
         run: python setup.py develop
 
-      - name: run ClickHouse Cloud SMT tests
+      - name: run ClickHouse Cloud tests
         env:
           CLICKHOUSE_CONNECT_TEST_PORT: 8443
           CLICKHOUSE_CONNECT_TEST_CLOUD: 'True'
@@ -42,18 +42,3 @@ jobs:
         run: pytest tests/integration_tests
       - name: remove latest container
         run: docker compose down -v
-
-      - name: run ClickHouse Cloud tests
-        env:
-          CLICKHOUSE_CONNECT_TEST_PORT: 8443
-          CLICKHOUSE_CONNECT_TEST_INSERT_QUORUM: 3
-          CLICKHOUSE_CONNECT_TEST_HOST: ${{ secrets.INTEGRATIONS_TEAM_TESTS_CLOUD_HOST }}
-          CLICKHOUSE_CONNECT_TEST_PASSWORD: ${{ secrets.INTEGRATIONS_TEAM_TESTS_CLOUD_PASSWORD }}
-        run: pytest tests/integration_tests
-
-      - name: Run ClickHouse Container (HEAD)
-        run: CLICKHOUSE_VERSION=head docker compose up -d clickhouse
-      - name: Run HEAD tests
-        run: pytest tests/integration_tests
-      - name: remove head container
-        run: docker compose down -v

.github/workflows/on_push.yml

@@ -58,17 +58,15 @@ jobs:
     strategy:
       matrix:
         python-version:
-          - '3.8'
           - '3.9'
           - '3.10'
           - '3.11'
           - '3.12'
         clickhouse-version:
-          - '23.8'
-          - '23.12'
-          - '24.1'
-          - '24.2'
           - '24.3'
+          - '24.6'
+          - '24.7'
+          - '24.8'
           - latest
 
     name: Local Tests Py=${{ matrix.python-version }} CH=${{ matrix.clickhouse-version }}
@@ -99,14 +97,11 @@ jobs:
           sudo echo "127.0.0.1 server1.clickhouse.test" | sudo tee -a /etc/hosts
       - name: Run tests
         env:
+          CLICKHOUSE_CONNECT_TEST_TLS: 1
           CLICKHOUSE_CONNECT_TEST_DOCKER: 'False'
           CLICKHOUSE_CONNECT_TEST_FUZZ: 50
           SQLALCHEMY_SILENCE_UBER_WARNING: 1
         run: pytest tests
-      - name: Run TLS tests
-        env:
-          CLICKHOUSE_CONNECT_TEST_TLS: 1
-        run: pytest tests/tls
 
   check-secret:
     runs-on: ubuntu-latest

CHANGELOG.md

@@ -3,10 +3,75 @@
 ### WARNING -- Impending Breaking Change - Server Settings in DSN
 When creating a DBAPI Connection method using the Connection constructor or a SQLAlchemy DSN, the library currently
 converts any unrecognized keyword argument/query parameter to a ClickHouse server setting. Starting in the next minor
-release (0.8.0), unrecognized arguments/keywords for these methods of creating a DBAPI connection will raise an exception
+release (0.9.0), unrecognized arguments/keywords for these methods of creating a DBAPI connection will raise an exception
 instead of being passed as ClickHouse server settings. This is in conjunction with some refactoring in Client construction.
 The supported method of passing ClickHouse server settings is to prefix such arguments/query parameters with`ch_`.  
 
+## 0.8.0, 2024-09-26
+### Experimental Feature - "New" JSON/Dynamic/Variant DataTypes
+#### Usage Notes
+- JSON data can be inserted as either a Python dictionary or a JSON string containing a JSON object `{}`.  Other
+forms of JSON data are not supported
+- Valid formats for the JSON type are 'native', which returns a Python dictionary, or 'string', which returns a JSON string
+- Any value can be inserted into a Variant column, and ClickHouse will try to correctly determine the correct Variant
+Type for the value, based on its String representation.
+- More complete documentation for the new types will be provided in the future.
+
+#### Known limitations:
+- Each of these types must be enabled in the ClickHouse settings before using.  The "new" JSON type is available started
+with the 24.8 release
+- Returned JSON objects will only return the `max_dynamic_paths` number of elements (which defaults to 1024).  This
+will be fixed in a future release.
+- Inserts into `Dynamic` columns will always be the String representation of the Python value.  This will be fixed
+in a future release.
+- The implementation for the new types has not been optimized in C code, so performance may be somewhat slower than for
+simpler, established data types.
+
+This is the first time that a new `clickhouse_connect` features has been labeled "experimental", but these new
+datatypes are complex and still experimental in ClickHouse server.  Current test coverage for these types is also
+quite limited.  Please don't hesitate to report issues with the new types.
+
+### Bug Fixes
+- When operating ClickHouse Server in `strict` TLS mode, HTTPS connections [require](https://github.com/ClickHouse/poco/blob/master/NetSSL_OpenSSL/include/Poco/Net/Context.h#L84-L89) a client certificate even if that
+certificate is not used for authentication.  A new client parameter `tls_mode='strict'` can be used in this situation where
+username/password authentication is being used with client certificates.  Other valid values for the new `tls_mode` setting
+are `'proxy'` when TLS termination occurs at a proxy, and `'mutual'` to specify mutual TLS authentication is used by
+the ClickHouse server.  If `tls_mode` is not set, and a client certificate and key are provided, `mutual` is assumed.
+- The server timezone was not being used for parameter binding if parameters were sent as a list instead of a dictionary.
+This should fully fix the reopened https://github.com/ClickHouse/clickhouse-connect/issues/377.
+- String port numbers (such as from environmental variables) are now correctly interpreted to determine the correct interface/protocol.
+Fixes https://github.com/ClickHouse/clickhouse-connect/issues/395
+- Insert commands with a `SELECT FROM ... LIMIT 0` will no longer raise an exception.  Closes https://github.com/ClickHouse/clickhouse-connect/issues/389.
+
+### Improvements
+- Some low level errors for problems with Native format inserts and queries now include the relevant column name in the
+error message.  Thanks to [Angus Holder](https://github.com/angusholder) for the PR!
+- There is a new intermediate buffer for HTTP streaming/chunked queries.  The buffer will store raw data from the HTTP request
+until it is actually requested in a stream.  This allows some lag between reading the data from ClickHouse and processing
+the same data.  Previously, if processing the data stream fell 30 seconds behind the ClickHouse HTTP writes to the stream,
+the ClickHouse server would close the connection, aborting the query and stream processing.  This will now be mitigated by
+storing the data stream in the new intermediate buffer.  By default, this buffer is set to 10 megabytes, but for slow
+processing of large queries where memory is not an issue, the buffer size can be increasing using the new `common` setting
+`http_buffer_size`.  This is a fix in some cases of https://github.com/ClickHouse/clickhouse-connect/issues/399, but note that
+slow processing of large queries will still cause connection and processing failures if the data cannot be buffered.
+- It is now possible to correctly bind `DateTime64` type parameters when calling Client `query` methods through one of two approaches:
+  - Wrap the Python `datetime.datetime` value in the new DT64Param class, e.g.
+  ```python
+    query = 'SELECT {p1:DateTime64(3)}'  # Server side binding with dictionary
+    parameters={'p1': DT64Param(dt_value)}
+
+    query = 'SELECT %s as string, toDateTime64(%s,6) as dateTime' # Client side binding with list 
+    parameters=['a string', DT64Param(datetime.now())]
+  ```
+  - If using a dictionary of parameter values, append the string `_64` to the parameter name
+  ```python
+    query = 'SELECT {p1:DateTime64(3)}, {a1:Array(DateTime(3))}'  # Server side binding with dictionary
+
+    parameters={'p1_64': dt_value, 'a1_64': [dt_value1, dt_value2]}
+  ```
+  This closes https://github.com/ClickHouse/clickhouse-connect/issues/396, see also the similar issue https://github.com/ClickHouse/clickhouse-connect/issues/212
+
+
 ## 0.7.19, 2024-08-23
 ### Bug Fix
 - Insertion of large strings was triggering an exception. This has been fixed.

clickhouse_connect/__version__.py

@@ -1 +1 @@
-version = '0.7.19'
+version = '0.8.0'

clickhouse_connect/cc_sqlalchemy/datatypes/base.py

@@ -5,7 +5,7 @@
 
 from clickhouse_connect.datatypes.base import ClickHouseType, TypeDef, EMPTY_TYPE_DEF
 from clickhouse_connect.datatypes.registry import parse_name, type_map
-from clickhouse_connect.driver.query import str_query_value
+from clickhouse_connect.driver.binding import str_query_value
 
 logger = logging.getLogger(__name__)
 

clickhouse_connect/cc_sqlalchemy/ddl/custom.py

@@ -1,7 +1,7 @@
 from sqlalchemy.sql.ddl import DDL
 from sqlalchemy.exc import ArgumentError
 
-from clickhouse_connect.driver.query import quote_identifier
+from clickhouse_connect.driver.binding import quote_identifier
 
 
 #  pylint: disable=too-many-ancestors,abstract-method
@@ -31,7 +31,7 @@ def __init__(self, name: str, engine: str = None, zoo_path: str = None, shard_na
         super().__init__(stmt)
 
 
-#  pylint: disable=too-many-ancestors,abstract-method
+# pylint: disable=too-many-ancestors,abstract-method
 class DropDatabase(DDL):
     """
     Alternative DDL statement for built in SqlAlchemy DropSchema DDL class

clickhouse_connect/cc_sqlalchemy/dialect.py

@@ -8,7 +8,7 @@
 from clickhouse_connect.cc_sqlalchemy.sql.ddlcompiler import ChDDLCompiler
 from clickhouse_connect.cc_sqlalchemy import ischema_names, dialect_name
 from clickhouse_connect.cc_sqlalchemy.sql.preparer import ChIdentifierPreparer
-from clickhouse_connect.driver.query import quote_identifier, format_str
+from clickhouse_connect.driver.binding import quote_identifier, format_str
 
 
 # pylint: disable=too-many-public-methods,no-self-use,unused-argument

clickhouse_connect/cc_sqlalchemy/sql/__init__.py

@@ -2,7 +2,7 @@
 
 from sqlalchemy import Table
 
-from clickhouse_connect.driver.query import quote_identifier
+from clickhouse_connect.driver.binding import quote_identifier
 
 
 def full_table(table_name: str, schema: Optional[str] = None) -> str:

clickhouse_connect/cc_sqlalchemy/sql/ddlcompiler.py

@@ -2,7 +2,7 @@
 from sqlalchemy.sql.compiler import DDLCompiler
 
 from clickhouse_connect.cc_sqlalchemy.sql import  format_table
-from clickhouse_connect.driver.query import quote_identifier
+from clickhouse_connect.driver.binding import quote_identifier
 
 
 class ChDDLCompiler(DDLCompiler):

clickhouse_connect/cc_sqlalchemy/sql/preparer.py

@@ -1,6 +1,6 @@
 from sqlalchemy.sql.compiler import IdentifierPreparer
 
-from clickhouse_connect.driver.query import quote_identifier
+from clickhouse_connect.driver.binding import quote_identifier
 
 
 class ChIdentifierPreparer(IdentifierPreparer):

clickhouse_connect/common.py

@@ -81,3 +81,6 @@ def _init_common(name: str, options: Sequence[Any], default: Any):
 _init_common('use_protocol_version', (True, False), True)
 
 _init_common('max_error_size', (), 1024)
+
+# HTTP raw data buffer for streaming queries.  This should not be reduced below 64KB to ensure compatibility with LZ4 compression
+_init_common('http_buffer_size', (), 10 * 1024 * 1024)

clickhouse_connect/datatypes/__init__.py

@@ -4,4 +4,6 @@
 import clickhouse_connect.datatypes.special
 import clickhouse_connect.datatypes.string
 import clickhouse_connect.datatypes.temporal
+import clickhouse_connect.datatypes.dynamic
 import clickhouse_connect.datatypes.registry
+import clickhouse_connect.datatypes.postinit

clickhouse_connect/datatypes/base.py

@@ -3,7 +3,7 @@
 
 from abc import ABC
 from math import log
-from typing import NamedTuple, Dict, Type, Any, Sequence, MutableSequence, Optional, Union, Collection
+from typing import NamedTuple, Dict, Type, Any, Sequence, MutableSequence, Union, Collection
 
 from clickhouse_connect.driver.common import array_type, int_size, write_array, write_uint64, low_card_version
 from clickhouse_connect.driver.context import BaseQueryContext
@@ -94,6 +94,10 @@ def name(self):
             name = f'{wrapper}({name})'
         return name
 
+    @property
+    def insert_name(self):
+        return self.name
+
     def data_size(self, sample: Sequence) -> int:
         if self.low_card:
             values = set(sample)
@@ -104,10 +108,13 @@ def data_size(self, sample: Sequence) -> int:
             d_size += 1
         return d_size
 
-    def _data_size(self, _sample: Collection) -> int:
+    def _data_size(self, sample: Collection) -> int:
         if self.byte_size:
             return self.byte_size
-        return 0
+        total = 0
+        for x in sample:
+            total += len(str(x))
+        return total / len(sample) + 1
 
     def write_column_prefix(self, dest: bytearray):
         """
@@ -119,7 +126,7 @@ def write_column_prefix(self, dest: bytearray):
         if self.low_card:
             write_uint64(low_card_version, dest)
 
-    def read_column_prefix(self, source: ByteSource):
+    def read_column_prefix(self, source: ByteSource, _ctx: QueryContext):
         """
         Read the low cardinality version.  Like the write method, this has to happen immediately for container classes
         :param source: The native protocol binary read buffer
@@ -139,7 +146,7 @@ def read_column(self, source: ByteSource, num_rows: int, ctx: QueryContext) -> S
         :param ctx: QueryContext for query specific settings
         :return: The decoded column data as a sequence and the updated location pointer
         """
-        self.read_column_prefix(source)
+        self.read_column_prefix(source, ctx)
         return self.read_column_data(source, num_rows, ctx)
 
     def read_column_data(self, source: ByteSource, num_rows: int, ctx: QueryContext) -> Sequence:
@@ -274,18 +281,11 @@ def _write_column_low_card(self, column: Sequence, dest: bytearray, ctx: InsertC
         write_uint64(len(index), dest)
         self._write_column_binary(index, dest, ctx)
         write_uint64(len(keys), dest)
-        write_array(array_type(1 << ix_type, False), keys, dest, ctx)
+        write_array(array_type(1 << ix_type, False), keys, dest, ctx.column_name)
 
     def _active_null(self, _ctx: QueryContext) -> Any:
         return None
 
-    def _first_value(self, column: Sequence) -> Optional[Any]:
-        if self.nullable:
-            return next((x for x in column if x is not None), None)
-        if len(column):
-            return column[0]
-        return None
-
 
 EMPTY_TYPE_DEF = TypeDef()
 NULLABLE_TYPE_DEF = TypeDef(wrappers=('Nullable',))
@@ -338,7 +338,7 @@ def _finalize_column(self, column: Sequence, ctx: QueryContext) -> Sequence:
     def _write_column_binary(self, column: Union[Sequence, MutableSequence], dest: bytearray, ctx: InsertContext):
         if len(column) and self.nullable:
             column = [0 if x is None else x for x in column]
-        write_array(self._array_type, column, dest, ctx)
+        write_array(self._array_type, column, dest, ctx.column_name)
 
     def _active_null(self, ctx: QueryContext):
         if ctx.as_pandas and ctx.use_extended_dtypes: