sibi-dst 2025.8.5__py3-none-any.whl → 2025.8.6__py3-none-any.whl

sibi_dst/df_helper/_df_helper.py

@@ -241,8 +241,8 @@ class DfHelper(ManagedResource):
 
     # ---------- sinks ----------
     def save_to_parquet(self, df: dd.DataFrame, parquet_filename: str, **kwargs):
-        fs: AbstractFileSystem = kwargs.get("fs", self.fs)
-        path: str = kwargs.get("parquet_storage_path")
+        fs: AbstractFileSystem = kwargs.pop("fs", self.fs)
+        path: str = kwargs.pop("parquet_storage_path")
         if not fs:
             raise ValueError("A filesystem (fs) must be provided to save the parquet file.")
         if not path:
@@ -268,7 +268,7 @@ class DfHelper(ManagedResource):
         if hasattr(df, "npartitions") and df.npartitions == 1 and not len(df.head(1)):
             self.logger.warning("Cannot write to ClickHouse; DataFrame is empty.")
             return
-        with ClickHouseWriter(debug=self.debug, logger=self.logger, fs=self.fs, verbose=self.verbose, **credentials) as writer:
+        with ClickHouseWriter(debug=self.debug, logger=self.logger, verbose=self.verbose, **credentials) as writer:
             writer.save_to_clickhouse(df)
             self.logger.debug("Save to ClickHouse completed.")
 

sibi_dst/utils/clickhouse_writer.py

@@ -91,7 +91,7 @@ class ClickHouseWriter(ManagedResource):
             return
 
         # lazily fill missing values per-partition (no global compute)
-        df = df.map_partitions(self._fill_missing_partition, meta=df)
+        df = df.map_partitions(type(self)._fill_missing_partition, meta=df._meta)
 
         # (re)create table
         ow = self.overwrite if overwrite is None else bool(overwrite)
@@ -201,23 +201,21 @@ class ClickHouseWriter(ManagedResource):
 
     # ------------- missing values (lazy) -------------
 
-    def _fill_missing_partition(self, pdf: pd.DataFrame) -> pd.DataFrame:
-        # Fill by dtype family; leave real NaT for datetimes so Nullable(DateTime) accepts NULL
+    @staticmethod
+    def _fill_missing_partition(pdf: pd.DataFrame) -> pd.DataFrame:
+        # (unchanged body)
         for col in pdf.columns:
             s = pdf[col]
             if pd.api.types.is_integer_dtype(s.dtype):
-                # pandas nullable IntX supports NA → fill where needed
                 if pd.api.types.is_extension_array_dtype(s.dtype):
                     pdf[col] = s.fillna(pd.NA)
                 else:
                     pdf[col] = s.fillna(0)
             elif pd.api.types.is_bool_dtype(s.dtype):
-                # boolean pandas extension supports NA, ClickHouse uses UInt8; keep NA → Nullable
                 pdf[col] = s.fillna(pd.NA)
             elif pd.api.types.is_float_dtype(s.dtype):
                 pdf[col] = s.fillna(0.0)
             elif pd.api.types.is_datetime64_any_dtype(s.dtype):
-                # keep NaT; ClickHouse Nullable(DateTime) will take NULL
                 pass
             else:
                 pdf[col] = s.fillna("")
@@ -264,238 +262,3 @@ class ClickHouseWriter(ManagedResource):
             if hasattr(self._tlocal, "client"):
                 delattr(self._tlocal, "client")
 
-# from concurrent.futures import ThreadPoolExecutor
-# from typing import ClassVar, Dict
-#
-# import clickhouse_connect
-# import pandas as pd
-# from clickhouse_driver import Client
-# import dask.dataframe as dd
-#
-# from . import ManagedResource
-#
-#
-# class ClickHouseWriter(ManagedResource):
-#     """
-#     Provides functionality to write a Dask DataFrame to a ClickHouse database using
-#     a specified schema. This class handles the creation of tables, schema generation,
-#     data transformation, and data insertion. It ensures compatibility between Dask
-#     data types and ClickHouse types.
-#
-#     :ivar clickhouse_host: Host address of the ClickHouse database.
-#     :type clickhouse_host: str
-#     :ivar clickhouse_port: Port of the ClickHouse database.
-#     :type clickhouse_port: int
-#     :ivar clickhouse_dbname: Name of the database to connect to in ClickHouse.
-#     :type clickhouse_dbname: str
-#     :ivar clickhouse_user: Username for database authentication.
-#     :type clickhouse_user: str
-#     :ivar clickhouse_password: Password for database authentication.
-#     :type clickhouse_password: str
-#     :ivar clickhouse_table: Name of the table to store the data in.
-#     :type clickhouse_table: str
-#     :ivar logger: Logger instance for logging messages.
-#     :type logger: logging.Logger
-#     :ivar client: Instance of the ClickHouse database client.
-#     :type client: clickhouse_connect.Client or None
-#     :ivar df: Dask DataFrame to be written into ClickHouse.
-#     :type df: dask.dataframe.DataFrame
-#     :ivar order_by: Field or column name to use for table ordering.
-#     :type order_by: str
-#     """
-#     dtype_to_clickhouse:  ClassVar[Dict[str, str]] = {
-#         'int64': 'Int64',
-#         'int32': 'Int32',
-#         'float64': 'Float64',
-#         'float32': 'Float32',
-#         'bool': 'UInt8',
-#         'datetime64[ns]': 'DateTime',
-#         'object': 'String',
-#         'category': 'String',
-#     }
-#     df: dd.DataFrame
-#
-#     def __init__(self, **kwargs):
-#         super().__init__(**kwargs)
-#         self.clickhouse_host = kwargs.setdefault('host', "localhost")
-#         self.clickhouse_port = kwargs.setdefault('port', 8123)
-#         self.clickhouse_dbname = kwargs.setdefault('database', 'sibi_data')
-#         self.clickhouse_user = kwargs.setdefault('user', 'default')
-#         self.clickhouse_password = kwargs.setdefault('password', '')
-#         self.clickhouse_table = kwargs.setdefault('table', 'test_sibi_table')
-#
-#         #self.logger = logger or Logger.default_logger(logger_name=self.__class__.__name__)
-#         self.client = None
-#         self.order_by = kwargs.setdefault('order_by', 'id')
-#
-#     def save_to_clickhouse(self, df, **kwargs):
-#         self.df = df.copy()
-#         self.order_by = kwargs.setdefault('order_by', self.order_by)
-#         if len(self.df.head().index) == 0:
-#             self.logger.debug("Dataframe is empty")
-#             return
-#         self._handle_missing_values()
-#         self._connect()
-#         self._drop_table()
-#         self._create_table_from_dask()
-#         self._write_data()
-#
-#     def _connect(self):
-#         try:
-#             self.client = clickhouse_connect.get_client(
-#                 host=self.clickhouse_host,
-#                 port=self.clickhouse_port,
-#                 database=self.clickhouse_dbname,
-#                 user=self.clickhouse_user,
-#                 password=self.clickhouse_password
-#             )
-#             self.logger.debug("Connected to ClickHouse")
-#         except Exception as e:
-#             self.logger.error(e)
-#             raise
-#
-#     @staticmethod
-#     def _generate_clickhouse_schema(dask_dtypes, dtype_map):
-#         schema = []
-#         for col, dtype in dask_dtypes.items():
-#             # Handle pandas nullable types explicitly
-#             if isinstance(dtype, pd.Int64Dtype):  # pandas nullable Int64
-#                 clickhouse_type = 'Int64'
-#             elif isinstance(dtype, pd.Float64Dtype):  # pandas nullable Float64
-#                 clickhouse_type = 'Float64'
-#             elif isinstance(dtype, pd.BooleanDtype):  # pandas nullable Boolean
-#                 clickhouse_type = 'UInt8'
-#             elif isinstance(dtype, pd.DatetimeTZDtype) or 'datetime' in str(dtype):  # Nullable datetime
-#                 clickhouse_type = 'Nullable(DateTime)'
-#             elif isinstance(dtype, pd.StringDtype):  # pandas nullable String
-#                 clickhouse_type = 'String'
-#             else:
-#                 # Default mapping using the provided dtype_map
-#                 clickhouse_type = dtype_map.get(str(dtype), 'String')
-#             schema.append(f"`{col}` {clickhouse_type}")
-#         return ', '.join(schema)
-#
-#     def _drop_table(self):
-#         if self.client:
-#             self.client.command('DROP TABLE IF EXISTS {}'.format(self.clickhouse_table))
-#             self.logger.debug(f"Dropped table {self.clickhouse_table}")
-#
-#     def _create_table_from_dask(self, engine=None):
-#         if engine is None:
-#             engine = f"ENGINE = MergeTree() order by {self.order_by}"
-#         dtypes = self.df.dtypes
-#         clickhouse_schema = self._generate_clickhouse_schema(dtypes, self.dtype_to_clickhouse)
-#         create_table_sql = f"CREATE TABLE IF NOT EXISTS {self.clickhouse_table} ({clickhouse_schema}) {engine};"
-#         self.logger.debug(f"Creating table SQL:{create_table_sql}")
-#         if self.client:
-#             self.client.command(create_table_sql)
-#             self.logger.debug("Created table '{}'".format(self.clickhouse_table))
-#
-#     def _handle_missing_values(self):
-#         """
-#         Handle missing values in the Dask DataFrame before writing to ClickHouse.
-#         """
-#         self.logger.debug("Checking for missing values...")
-#         missing_counts = self.df.isnull().sum().compute()
-#         self.logger.debug(f"Missing values per column:\n{missing_counts}")
-#
-#         # Replace missing values based on column types
-#         def replace_missing_values(df):
-#             for col in df.columns:
-#                 if pd.api.types.is_integer_dtype(df[col]):
-#                     df[col] = df[col].fillna(0)  # Replace NA with 0 for integers
-#                 elif pd.api.types.is_float_dtype(df[col]):
-#                     df[col] = df[col].fillna(0.0)  # Replace NA with 0.0 for floats
-#                 elif pd.api.types.is_bool_dtype(df[col]):
-#                     df[col] = df[col].fillna(False)  # Replace NA with False for booleans
-#                 else:
-#                     df[col] = df[col].fillna('')  # Replace NA with empty string for other types
-#             return df
-#
-#         # Apply replacement
-#         self.df = replace_missing_values(self.df)
-#         self.logger.debug("Missing values replaced.")
-#
-#     def _write_data(self):
-#         """
-#         Writes the Dask DataFrame to a ClickHouse table partition by partition.
-#         """
-#         if len(self.df.index) == 0:
-#             self.logger.debug("No data found. Nothing written.")
-#             return
-#
-#         for i, partition in enumerate(self.df.to_delayed()):
-#             try:
-#                 # Compute the current partition into a pandas DataFrame
-#                 df = partition.compute()
-#
-#                 if df.empty:
-#                     self.logger.debug(f"Partition {i} is empty. Skipping...")
-#                     continue
-#
-#                 self.logger.debug(f"Writing partition {i} with {len(df)} rows to ClickHouse.")
-#
-#                 # Write the partition to the ClickHouse table
-#                 self.client.insert_df(self.clickhouse_table, df)
-#             except Exception as e:
-#                 self.logger.error(f"Error writing partition {i}: {e}")
-#
-#     def _write_data_multi_not_working_yet(self):
-#         """
-#         Writes the Dask DataFrame to a ClickHouse table partition by partition.
-#         Ensures a separate client instance is used per thread to avoid session conflicts.
-#         """
-#         if len(self.df.index) == 0:
-#             self.logger.debug("No data found. Nothing written.")
-#             return
-#
-#         def create_client():
-#             client = Client(
-#                 host=self.clickhouse_host,
-#                 port=self.clickhouse_port,
-#                 database=self.clickhouse_dbname,
-#                 user=self.clickhouse_user,
-#                 password=self.clickhouse_password
-#             )
-#             """
-#             Create a new instance of the ClickHouse client for each thread.
-#             This avoids session conflicts during concurrent writes.
-#             """
-#             return client
-#
-#         def write_partition(partition, index):
-#             """
-#             Write a single partition to ClickHouse using a separate client instance.
-#             """
-#             try:
-#                 self.logger.debug(f"Starting to process partition {index}")
-#                 client = create_client()  # Create a new client for the thread
-#
-#                 # Compute the Dask partition into a Pandas DataFrame
-#                 df = partition.compute()
-#                 if df.empty:
-#                     self.logger.debug(f"Partition {index} is empty. Skipping...")
-#                     return
-#
-#                 # Convert DataFrame to list of tuples
-#                 data = [tuple(row) for row in df.to_numpy()]
-#                 columns = df.columns.tolist()
-#
-#                 # Perform the insert
-#                 self.logger.debug(f"Writing partition {index} with {len(df)} rows to ClickHouse.")
-#                 client.execute(f"INSERT INTO {self.clickhouse_table} ({', '.join(columns)}) VALUES", data)
-#
-#             except Exception as e:
-#                 self.logger.error(f"Error writing partition {index}: {e}")
-#             finally:
-#                 if 'client' in locals() and hasattr(client, 'close'):
-#                     client.close()
-#                     self.logger.debug(f"Closed client for partition {index}")
-#
-#         try:
-#             # Get delayed partitions and enumerate them
-#             partitions = self.df.to_delayed()
-#             with ThreadPoolExecutor() as executor:
-#                 executor.map(write_partition, partitions, range(len(partitions)))
-#         except Exception as e:
-#             self.logger.error(f"Error during multi-partition write: {e}")

sibi_dst/utils/storage_config.py

@@ -6,13 +6,13 @@ from .storage_manager import StorageManager
 from .credentials import ConfigManager
 
 class StorageConfig:
-    def __init__(self, config:ConfigManager, depots:dict=None):
+    def __init__(self, config:ConfigManager, depots:dict=None, clear_existing=False, write_mode="full-access"):
         self.conf = config
         self.depots = depots
         self._initialize_storage()
         self.storage_manager = StorageManager(self.base_storage, self.filesystem_type, self.filesystem_options)
         if self.depots is not None:
-            self.depot_paths, self.depot_names = self.storage_manager.rebuild_depot_paths(depots)
+            self.depot_paths, self.depot_names = self.storage_manager.rebuild_depot_paths(depots, clear_existing=clear_existing, write_mode=write_mode)
         else:
             self.depot_paths = None
             self.depot_names = None

sibi_dst/utils/storage_manager.py

@@ -83,7 +83,7 @@ class StorageManager:
                 self.fs.rm(sub_path, recursive=True)
             self.fs.mkdirs(sub_path, exist_ok=True)
 
-    def rebuild_depot_paths(self, depots, clear_existing=False):
+    def rebuild_depot_paths(self, depots, clear_existing=False, write_mode="full-access"):
         """
         Rebuilds depot_paths (dictionary) and depot_name (SimpleNamespace).
         Handles clear_existing scenario by resetting directories when required.
@@ -96,7 +96,8 @@ class StorageManager:
             depot_path = self.join_paths(self.storage_path, depot)
             if self.debug:
                 print(f"Rebuilding depot at: {depot_path}")
-            self.setup_directories(depot_path, sub_directories, clear_existing=clear_existing)
+            if write_mode == "full-access":
+                self.setup_directories(depot_path, sub_directories, clear_existing=clear_existing)
 
         # Generate depot_paths dictionary
         self.depot_paths = {

{sibi_dst-2025.8.5.dist-info → sibi_dst-2025.8.6.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sibi-dst
-Version: 2025.8.5
+Version: 2025.8.6
 Summary: Data Science Toolkit
 Author: Luis Valverde
 Author-email: lvalverdeb@gmail.com

{sibi_dst-2025.8.5.dist-info → sibi_dst-2025.8.6.dist-info}/RECORD

@@ -2,7 +2,7 @@ sibi_dst/__init__.py,sha256=D01Z2Ds4zES8uz5Zp7qOWD0EcfCllWgew7AWt2X1SQg,445
 sibi_dst/df_helper/__init__.py,sha256=CyDXtFhRnMrycktxNO8jGGkP0938QiScl56kMZS1Sf8,578
 sibi_dst/df_helper/_artifact_updater_async.py,sha256=0lUwel-IkmKewRnmMv9GtuT-P6SivkIKtgOHvKchHlc,8462
 sibi_dst/df_helper/_artifact_updater_threaded.py,sha256=M5GNZismOqMmBrcyfolP1DPv87VILQf_P18is_epn50,7238
-sibi_dst/df_helper/_df_helper.py,sha256=g1ftfSMO40l60EJWRLE0DDZvbIowrqvG1GMf2zXqYGw,12957
+sibi_dst/df_helper/_df_helper.py,sha256=TS8nQV6QExSz5rNh94zmawNOvQ6eBEzsAcJkiiKXAb0,12945
 sibi_dst/df_helper/_parquet_artifact.py,sha256=tqYOjwxHV1MsADmn-RNFuVI_RrEvvmCJHZieRcsVXuc,12334
 sibi_dst/df_helper/_parquet_reader.py,sha256=tFq0OQVczozbKZou93vscokp2R6O2DIJ1zHbZqVjagc,3069
 sibi_dst/df_helper/backends/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
@@ -39,7 +39,7 @@ sibi_dst/tests/test_data_wrapper_class.py,sha256=6uFmZR2DxnxQz49L5jT2ehlKvlLnpUH
 sibi_dst/utils/__init__.py,sha256=vShNCOMPw8KKwlb4tq5XGrpjqakJ_OE8YDc_xDAWAxI,1302
 sibi_dst/utils/base.py,sha256=IyObjZ7AaE-YjVU0RLIXNCnQKWwzi5NH2I6D1KfcIyk,8716
 sibi_dst/utils/business_days.py,sha256=dP0Xj4FhTBIvZZrZYLOHZl5zOpDAgWkD4p_1a7BOT7I,8461
-sibi_dst/utils/clickhouse_writer.py,sha256=pE-igxddDdxekJywsaWQqKlGXIvLjPMpoFBUJ24t9Tw,20255
+sibi_dst/utils/clickhouse_writer.py,sha256=NngJyJpx2PjUQWsX0YmwCuGdeViK77Wi3HmYqHz3jTc,9544
 sibi_dst/utils/credentials.py,sha256=cHJPPsmVyijqbUQIq7WWPe-lIallA-mI5RAy3YUuRME,1724
 sibi_dst/utils/data_from_http_source.py,sha256=AcpKNsqTgN2ClNwuhgUpuNCx62r5_DdsAiKY8vcHEBA,1867
 sibi_dst/utils/data_utils.py,sha256=7bLidEjppieNoozDFb6OuRY0W995cxg4tiGAlkGfePI,7768
@@ -54,8 +54,8 @@ sibi_dst/utils/manifest_manager.py,sha256=9y4cV-Ig8O-ekhApp_UObTY-cTsl-bGnvKIThI
 sibi_dst/utils/parquet_saver.py,sha256=aYBlijqPAn-yuJXhmaRIteAN_IAQZvPh8I8Os2TLGgI,4861
 sibi_dst/utils/periods.py,sha256=8eTGi-bToa6_a8Vwyg4fkBPryyzft9Nzy-3ToxjqC8c,1434
 sibi_dst/utils/phone_formatter.py,sha256=oeM22nLjhObENrpItCNeVpkYS4pXRm5hSxdk0M4nvwU,4580
-sibi_dst/utils/storage_config.py,sha256=uaCBF8rgCeYkk-lxVSCjsic8O8HJKAu455MR-OBliCo,4325
-sibi_dst/utils/storage_manager.py,sha256=yyZqT8XjTf4MKFrfznCmxXxOYz_TiWgtQhzqPoXR9So,6569
+sibi_dst/utils/storage_config.py,sha256=DLtP5jKVM0mdFdgRw6LQfRqyavMjJcCVU7GhsUCRH78,4427
+sibi_dst/utils/storage_manager.py,sha256=La1NY79bhRAmHWXp7QcXJZtbHoRboJMgoXOSXbIl1SA,6643
 sibi_dst/utils/update_planner.py,sha256=smlMHpr1p8guZnP5SyzCe6RsC-XkPOJWIsdeospUyb0,11471
 sibi_dst/utils/webdav_client.py,sha256=D9J5d1f1qQwHGm5FE5AMVpOPwcU5oD7K8JZoKGP8NpM,5811
 sibi_dst/v2/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
@@ -78,6 +78,6 @@ sibi_dst/v2/df_helper/core/_params_config.py,sha256=DYx2drDz3uF-lSPzizPkchhy-kxR
 sibi_dst/v2/df_helper/core/_query_config.py,sha256=Y8LVSyaKuVkrPluRDkQoOwuXHQxner1pFWG3HPfnDHM,441
 sibi_dst/v2/utils/__init__.py,sha256=6H4cvhqTiFufnFPETBF0f8beVVMpfJfvUs6Ne0TQZNY,58
 sibi_dst/v2/utils/log_utils.py,sha256=rfk5VsLAt-FKpv6aPTC1FToIPiyrnHAFFBAkHme24po,4123
-sibi_dst-2025.8.5.dist-info/METADATA,sha256=ADWrf_9UI4NiTWslrJ0LgfmHTTdxSSCIc0AaP-mqSQg,2610
-sibi_dst-2025.8.5.dist-info/WHEEL,sha256=Nq82e9rUAnEjt98J6MlVmMCZb-t9cYE2Ir1kpBmnWfs,88
-sibi_dst-2025.8.5.dist-info/RECORD,,
+sibi_dst-2025.8.6.dist-info/METADATA,sha256=fFvtxHyXl8FCryToGB4H91n_NZ3hzJNRos0O2FUNVBQ,2610
+sibi_dst-2025.8.6.dist-info/WHEEL,sha256=Nq82e9rUAnEjt98J6MlVmMCZb-t9cYE2Ir1kpBmnWfs,88
+sibi_dst-2025.8.6.dist-info/RECORD,,