e6data-python-connector 2.3.12rc11__tar.gz → 2.3.12rc31__tar.gz

{e6data_python_connector-2.3.12rc11/e6data_python_connector.egg-info → e6data_python_connector-2.3.12rc31}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: e6data-python-connector
-Version: 2.3.12rc11
+Version: 2.3.12rc31
 Summary: Client for the e6data distributed SQL Engine.
 Home-page: https://github.com/e6x-labs/e6data-python-connector
 Author: e6data, Inc.
@@ -39,7 +39,7 @@ Dynamic: summary
 
 # e6data Python Connector
 
-![version](https://img.shields.io/badge/version-2.3.12rc11-blue.svg)
+![version](https://img.shields.io/badge/version-2.3.12rc31-blue.svg)
 
 ## Introduction
 
@@ -125,6 +125,7 @@ The `Connection` class supports the following parameters:
 | `auto_resume` | bool | No | True | Automatically resume cluster if suspended |
 | `grpc_options` | dict | No | None | Additional gRPC configuration options |
 | `debug` | bool | No | False | Enable debug logging for troubleshooting |
+| `require_fastbinary` | bool | No | True | Require fastbinary module for Thrift deserialization. Set to False to use pure Python implementation if system dependencies cannot be installed |
 
 #### Secure Connection Example
 
@@ -158,6 +159,51 @@ conn = Connection(
 )
 ```
 
+#### Handling Missing System Dependencies
+
+The e6data connector uses the `fastbinary` module (from Apache Thrift) for optimal performance when deserializing data. This module requires system-level dependencies (`python3-devel` and `gcc-c++`) to be installed.
+
+**Default Behavior (Recommended):**
+By default, the connector requires `fastbinary` to be available. If it's not found, the connection will fail immediately with a clear error message:
+
+```python
+conn = Connection(
+    host=host,
+    port=port,
+    username=username,
+    password=password,
+    database=database
+)
+# Raises exception if fastbinary is not available
+```
+
+**Fallback to Pure Python:**
+If you cannot install system dependencies (e.g., in restricted environments, serverless platforms, or containers without build tools), you can disable the `fastbinary` requirement. The connector will fall back to a pure Python implementation with a performance penalty:
+
+```python
+conn = Connection(
+    host=host,
+    port=port,
+    username=username,
+    password=password,
+    database=database,
+    require_fastbinary=False  # Allow operation without fastbinary
+)
+# Logs warning but continues with pure Python implementation
+```
+
+**When to use `require_fastbinary=False`:**
+- Running in AWS Lambda or other serverless environments
+- Docker containers built without compilation tools
+- Restricted environments where system packages cannot be installed
+- Development/testing environments where performance is not critical
+
+**Performance Impact:**
+- With `fastbinary`: Optimal performance for data deserialization
+- Without `fastbinary` (pure Python): ~2-3x slower deserialization, but otherwise fully functional
+
+**Note:** It's strongly recommended to install system dependencies when possible for best performance. The `require_fastbinary=False` option should only be used when system dependencies cannot be installed.
+
 ### Perform a Queries & Get Results
 
 ```python
@@ -691,5 +737,6 @@ python your_script.py
 | 456 Strategy Error | Automatic blue-green failover will handle this |
 | Memory issues with large results | Use `fetchall_buffer()` instead of `fetchall()` |
 | gRPC message size errors | Configure `grpc_options` with appropriate message size limits |
+| fastbinary import error | Install system dependencies (`python3-devel`, `gcc-c++`) or set `require_fastbinary=False` |
 
 See [TECH_DOC.md](TECH_DOC.md) for detailed technical documentation.

{e6data_python_connector-2.3.12rc11 → e6data_python_connector-2.3.12rc31}/README.md

@@ -1,6 +1,6 @@
 # e6data Python Connector
 
-![version](https://img.shields.io/badge/version-2.3.12rc11-blue.svg)
+![version](https://img.shields.io/badge/version-2.3.12rc31-blue.svg)
 
 ## Introduction
 
@@ -86,6 +86,7 @@ The `Connection` class supports the following parameters:
 | `auto_resume` | bool | No | True | Automatically resume cluster if suspended |
 | `grpc_options` | dict | No | None | Additional gRPC configuration options |
 | `debug` | bool | No | False | Enable debug logging for troubleshooting |
+| `require_fastbinary` | bool | No | True | Require fastbinary module for Thrift deserialization. Set to False to use pure Python implementation if system dependencies cannot be installed |
 
 #### Secure Connection Example
 
@@ -119,6 +120,51 @@ conn = Connection(
 )
 ```
 
+#### Handling Missing System Dependencies
+
+The e6data connector uses the `fastbinary` module (from Apache Thrift) for optimal performance when deserializing data. This module requires system-level dependencies (`python3-devel` and `gcc-c++`) to be installed.
+
+**Default Behavior (Recommended):**
+By default, the connector requires `fastbinary` to be available. If it's not found, the connection will fail immediately with a clear error message:
+
+```python
+conn = Connection(
+    host=host,
+    port=port,
+    username=username,
+    password=password,
+    database=database
+)
+# Raises exception if fastbinary is not available
+```
+
+**Fallback to Pure Python:**
+If you cannot install system dependencies (e.g., in restricted environments, serverless platforms, or containers without build tools), you can disable the `fastbinary` requirement. The connector will fall back to a pure Python implementation with a performance penalty:
+
+```python
+conn = Connection(
+    host=host,
+    port=port,
+    username=username,
+    password=password,
+    database=database,
+    require_fastbinary=False  # Allow operation without fastbinary
+)
+# Logs warning but continues with pure Python implementation
+```
+
+**When to use `require_fastbinary=False`:**
+- Running in AWS Lambda or other serverless environments
+- Docker containers built without compilation tools
+- Restricted environments where system packages cannot be installed
+- Development/testing environments where performance is not critical
+
+**Performance Impact:**
+- With `fastbinary`: Optimal performance for data deserialization
+- Without `fastbinary` (pure Python): ~2-3x slower deserialization, but otherwise fully functional
+
+**Note:** It's strongly recommended to install system dependencies when possible for best performance. The `require_fastbinary=False` option should only be used when system dependencies cannot be installed.
+
 ### Perform a Queries & Get Results
 
 ```python
@@ -652,5 +698,6 @@ python your_script.py
 | 456 Strategy Error | Automatic blue-green failover will handle this |
 | Memory issues with large results | Use `fetchall_buffer()` instead of `fetchall()` |
 | gRPC message size errors | Configure `grpc_options` with appropriate message size limits |
+| fastbinary import error | Install system dependencies (`python3-devel`, `gcc-c++`) or set `require_fastbinary=False` |
 
 See [TECH_DOC.md](TECH_DOC.md) for detailed technical documentation.

{e6data_python_connector-2.3.12rc11 → e6data_python_connector-2.3.12rc31}/e6data_python_connector/datainputstream.py

@@ -12,20 +12,27 @@ from e6data_python_connector.e6x_vector.ttypes import Chunk, Vector, VectorType
 from e6data_python_connector.constants import ZONE
 from e6data_python_connector.date_time_utils import floor_div, floor_mod, timezone_from_offset
 
+# Try to import fastbinary - it's optional but provides better performance
+_fastbinary_available = False
 try:
     from thrift.protocol import fastbinary
+    _fastbinary_available = True
 except ImportError:
-    raise Exception(
-        """
-        Failed to import fastbinary. 
-        Did you install system dependencies?
-        Please verify https://github.com/e6x-labs/e6data-python-connector#dependencies
-        """
-    )
+    pass  # Will check require_fastbinary flag at connection creation time
 
 _logger = logging.getLogger(__name__)
 
 
+def is_fastbinary_available():
+    """
+    Check if fastbinary module is available.
+
+    Returns:
+        bool: True if fastbinary is available, False otherwise
+    """
+    return _fastbinary_available
+
+
 def _binary_to_decimal128(binary_data, scale=None):
     """
     Convert binary data to Decimal128.
@@ -504,6 +511,16 @@ def read_values_from_array(query_columns_description: list, dis: DataInputStream
 
 
 def read_rows_from_chunk(query_columns_description: list, buffer):
+    """
+    Read rows from a Thrift-encoded chunk buffer.
+
+    Args:
+        query_columns_description: List of column descriptions
+        buffer: Thrift-encoded binary buffer
+
+    Returns:
+        List of rows
+    """
     # Create a transport and protocol instance for deserialization
     transport = TTransport.TMemoryBuffer(buffer)
     protocol = TBinaryProtocol.TBinaryProtocolAccelerated(transport)

{e6data_python_connector-2.3.12rc11 → e6data_python_connector-2.3.12rc31}/e6data_python_connector/e6data_grpc.py

@@ -26,7 +26,7 @@ from e6data_python_connector.cluster_manager import ClusterManager
 from e6data_python_connector.strategy import _get_grpc_header as _get_strategy_header
 from e6data_python_connector.common import DBAPITypeObject, ParamEscaper, DBAPICursor
 from e6data_python_connector.constants import *
-from e6data_python_connector.datainputstream import get_query_columns_info, read_rows_from_chunk
+from e6data_python_connector.datainputstream import get_query_columns_info, read_rows_from_chunk, is_fastbinary_available
 from e6data_python_connector.server import e6x_engine_pb2_grpc, e6x_engine_pb2
 from e6data_python_connector.typeId import *
 
@@ -335,6 +335,7 @@ class Connection(object):
             scheme: str = 'e6data',
             grpc_options: dict = None,
             debug: bool = False,
+            require_fastbinary: bool = True,
     ):
         """
         Parameters
@@ -368,6 +369,10 @@ class Connection(object):
                 - keepalive_time_ms: This parameter defines the time, in milliseconds, Default to 30 seconds
             debug: bool, Optional
                 Flag to enable debug logging for blue-green deployment strategy changes
+            require_fastbinary: bool, Optional
+                Flag to require fastbinary module for Thrift deserialization. If True (default),
+                raises an exception if fastbinary is not available. If False, logs a warning
+                and continues with pure Python implementation (with reduced performance).
         """
         if not username or not password:
             raise ValueError("username or password cannot be empty.")
@@ -387,6 +392,29 @@ class Connection(object):
 
         self._auto_resume = auto_resume
 
+        # Store require_fastbinary flag
+        self._require_fastbinary = require_fastbinary
+
+        # Check fastbinary availability at connection creation time
+        if not is_fastbinary_available():
+            if require_fastbinary:
+                raise Exception(
+                    """
+                    Failed to import fastbinary.
+                    Did you install system dependencies?
+                    Please verify https://github.com/e6x-labs/e6data-python-connector#dependencies
+
+                    To continue without fastbinary (with reduced performance), set require_fastbinary=False
+                    in the connection parameters.
+                    """
+                )
+            else:
+                logger.warning(
+                    "fastbinary module is not available. Using pure Python implementation. "
+                    "Performance may be degraded. To enable fastbinary, install system dependencies: "
+                    "https://github.com/e6x-labs/e6data-python-connector#dependencies"
+                )
+
         self._grpc_options = grpc_options
         if self._grpc_options is None:
             self._grpc_options = dict()
@@ -853,56 +881,6 @@ class Connection(object):
         )
         return dry_run_response.dryrunValue
 
-    def get_logical_plan(self, query, catalog=None, schema=None):
-        """
-        Gets the logical execution plan for a query without executing it.
-        This method performs syntax, parsing, and plan compilation, then returns the logical plan.
-
-        Args:
-            query (str): The SQL query to get the plan for.
-            catalog (str, optional): Catalog name. Uses connection's catalog if not provided.
-            schema (str, optional): Schema/database name. Uses connection's database if not provided.
-
-        Returns:
-            dict: A dictionary with:
-                - 'success' (bool): True if plan generation succeeded, False otherwise
-                - 'plan' (str, optional): The logical plan string if successful
-                - 'error' (str, optional): Error message if plan generation failed
-
-        Example:
-            >>> result = conn.get_logical_plan("SELECT * FROM table")
-            >>> if result['success']:
-            ...     print(f"Plan: {result['plan']}")
-            >>> else:
-            ...     print(f"Error: {result['error']}")
-        """
-        catalog_to_use = catalog if catalog is not None else self.catalog_name
-        schema_to_use = schema if schema is not None else self.database
-        
-        try:
-            dry_run_request = e6x_engine_pb2.DryRunRequestV2(
-                sessionId=self.get_session_id,
-                schema=schema_to_use,
-                catalog=catalog_to_use,
-                queryString=query
-            )
-            
-            dry_run_response = self._client.dryRunV2(
-                dry_run_request,
-                metadata=_get_grpc_header(cluster=self.cluster_name, strategy=_get_active_strategy())
-            )
-            
-            return {
-                'success': True,
-                'plan': dry_run_response.dryrunValue
-            }
-            
-        except Exception as e:
-            return {
-                'success': False,
-                'error': str(e)
-            } 
-
     def get_tables(self, catalog, database):
         """
         Retrieves the list of tables from the specified catalog and database.
@@ -1514,7 +1492,10 @@ class Cursor(DBAPICursor):
         if not buffer or len(buffer) == 0:
             return None
         # one batch retrieves the predefined set of rows
-        return read_rows_from_chunk(self._query_columns_description, buffer)
+        return read_rows_from_chunk(
+            self._query_columns_description,
+            buffer
+        )
 
     def fetchall(self):
         """

{e6data_python_connector-2.3.12rc11 → e6data_python_connector-2.3.12rc31/e6data_python_connector.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: e6data-python-connector
-Version: 2.3.12rc11
+Version: 2.3.12rc31
 Summary: Client for the e6data distributed SQL Engine.
 Home-page: https://github.com/e6x-labs/e6data-python-connector
 Author: e6data, Inc.
@@ -39,7 +39,7 @@ Dynamic: summary
 
 # e6data Python Connector
 
-![version](https://img.shields.io/badge/version-2.3.12rc11-blue.svg)
+![version](https://img.shields.io/badge/version-2.3.12rc31-blue.svg)
 
 ## Introduction
 
@@ -125,6 +125,7 @@ The `Connection` class supports the following parameters:
 | `auto_resume` | bool | No | True | Automatically resume cluster if suspended |
 | `grpc_options` | dict | No | None | Additional gRPC configuration options |
 | `debug` | bool | No | False | Enable debug logging for troubleshooting |
+| `require_fastbinary` | bool | No | True | Require fastbinary module for Thrift deserialization. Set to False to use pure Python implementation if system dependencies cannot be installed |
 
 #### Secure Connection Example
 
@@ -158,6 +159,51 @@ conn = Connection(
 )
 ```
 
+#### Handling Missing System Dependencies
+
+The e6data connector uses the `fastbinary` module (from Apache Thrift) for optimal performance when deserializing data. This module requires system-level dependencies (`python3-devel` and `gcc-c++`) to be installed.
+
+**Default Behavior (Recommended):**
+By default, the connector requires `fastbinary` to be available. If it's not found, the connection will fail immediately with a clear error message:
+
+```python
+conn = Connection(
+    host=host,
+    port=port,
+    username=username,
+    password=password,
+    database=database
+)
+# Raises exception if fastbinary is not available
+```
+
+**Fallback to Pure Python:**
+If you cannot install system dependencies (e.g., in restricted environments, serverless platforms, or containers without build tools), you can disable the `fastbinary` requirement. The connector will fall back to a pure Python implementation with a performance penalty:
+
+```python
+conn = Connection(
+    host=host,
+    port=port,
+    username=username,
+    password=password,
+    database=database,
+    require_fastbinary=False  # Allow operation without fastbinary
+)
+# Logs warning but continues with pure Python implementation
+```
+
+**When to use `require_fastbinary=False`:**
+- Running in AWS Lambda or other serverless environments
+- Docker containers built without compilation tools
+- Restricted environments where system packages cannot be installed
+- Development/testing environments where performance is not critical
+
+**Performance Impact:**
+- With `fastbinary`: Optimal performance for data deserialization
+- Without `fastbinary` (pure Python): ~2-3x slower deserialization, but otherwise fully functional
+
+**Note:** It's strongly recommended to install system dependencies when possible for best performance. The `require_fastbinary=False` option should only be used when system dependencies cannot be installed.
+
 ### Perform a Queries & Get Results
 
 ```python
@@ -691,5 +737,6 @@ python your_script.py
 | 456 Strategy Error | Automatic blue-green failover will handle this |
 | Memory issues with large results | Use `fetchall_buffer()` instead of `fetchall()` |
 | gRPC message size errors | Configure `grpc_options` with appropriate message size limits |
+| fastbinary import error | Install system dependencies (`python3-devel`, `gcc-c++`) or set `require_fastbinary=False` |
 
 See [TECH_DOC.md](TECH_DOC.md) for detailed technical documentation.

{e6data_python_connector-2.3.12rc11 → e6data_python_connector-2.3.12rc31}/setup.py

@@ -12,7 +12,7 @@
 
 import setuptools
 
-VERSION = (2, 3, 12,'rc11')
+VERSION = (2, 3, 12, 'rc31',)
 
 
 def get_long_desc():