chore: bump update versions

.github/workflows/publish-docs.yml

@@ -21,7 +21,7 @@ jobs:
     - name: Set up Python
       uses: actions/setup-python@v5
       with:
-        python-version: '3.9'
+        python-version: '3.12'
     - name: Install dependencies
       run: python -m pip install .[dev]
     - name: Build

.github/workflows/publish-package.yml

@@ -21,7 +21,7 @@ jobs:
     - name: Set up Python
       uses: actions/setup-python@a26af69be951a213d495a4c3e4e4022e16d87065 # pin@v5.6.0
       with:
-        python-version: '3.10'
+        python-version: '3.12'
     - name: Install dependencies
       run: python -m pip install -r requirements.txt .[dev]
     - name: Build package

.github/workflows/python-package.yml

@@ -14,10 +14,10 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - uses: actions/checkout@v4
-      - name: Set up Python 3.9
+      - name: Set up Python 3.12
         uses: actions/setup-python@v5
         with:
-            python-version: "3.9"
+            python-version: "3.12"
       - name: Install dependencies
         run: python -m pip install .[qa]
       - name: Linting by ruff
@@ -30,7 +30,7 @@ jobs:
     runs-on: ubuntu-latest
     strategy:
       matrix:
-        python-version: [3.9, '3.10', '3.11', '3.12']
+        python-version: ['3.10', '3.11', '3.12', '3.13']
 
     steps:
     - uses: actions/checkout@v4

CLAUDE.md

@@ -0,0 +1,73 @@
+# CLAUDE.md
+
+This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
+
+## Project Overview
+
+locopy is a Python library for ETL processing with Amazon Redshift (`COPY`/`UNLOAD`) and Snowflake (`COPY INTO`). It wraps boto3 for S3 operations and is DB-API 2.0 adapter agnostic (tested with psycopg2, pg8000, snowflake-connector-python). Supports Python 3.10-3.14.
+
+## Common Commands
+
+### Install for development
+```bash
+pip install .[dev,psycopg2,pg8000,snowflake]
+# Test data setup (needed for tests):
+cp tests/data/.locopyrc ~/.locopyrc
+cp tests/data/.locopy-sfrc ~/.locopy-sfrc
+```
+
+### Run tests
+```bash
+make not_integration              # Unit tests only (default CI target)
+make coverage                     # All tests with coverage
+pytest tests/test_utility.py      # Single test file
+pytest tests/test_utility.py::test_find_column_type -v  # Single test
+pytest -m 'not integration'       # Skip integration tests (same as make not_integration)
+```
+
+### Lint and format
+```bash
+ruff check                        # Lint
+ruff check --fix                  # Lint with auto-fix
+ruff format --check               # Check formatting
+ruff format                       # Auto-format
+```
+
+### Build docs
+```bash
+make sphinx
+```
+
+### Dependency version bumps (edgetest)
+Edgetest config is in `pyproject.toml` under `[edgetest.envs.core]`. It bumps upper bounds for boto3, PyYAML, pandas, numpy. The CI workflow runs weekly and creates PRs with updated `pyproject.toml` and `requirements.txt`. The lockfile is generated via `uv pip compile --output-file=requirements.txt pyproject.toml`.
+
+## Architecture
+
+```
+locopy/
+├── database.py    # Database - base class for DB connections (connect, execute, to_dataframe)
+├── s3.py          # S3 - boto3 wrapper for upload/download/delete on S3 buckets
+├── redshift.py    # Redshift(S3, Database) - multiple inheritance, adds COPY/UNLOAD + load_and_copy/unload_and_copy
+├── snowflake.py   # Snowflake(S3, Database) - multiple inheritance, adds COPY INTO + internal stage support
+├── utility.py     # Helpers: file splitting, compression, YAML config reading, column type detection
+├── errors.py      # Custom exception hierarchy: LocopyError, DBError, S3Error (each with sub-exceptions)
+├── logger.py      # Logging setup
+└── _version.py    # Single source of version (__version__)
+```
+
+**Key inheritance pattern:** Both `Redshift` and `Snowflake` use multiple inheritance from `S3` and `Database`. The `S3` class handles AWS session/credentials and file transfer. `Database` handles DB connection lifecycle and query execution. The subclasses override `connect()` to set up both S3 and DB connections.
+
+**Column type detection:** `utility.py` has `find_column_type` as a `@singledispatch` function with separate implementations for pandas (`find_column_type_pandas`) and polars (`find_column_type_polars`) DataFrames. When bumping pandas/polars versions, watch for dtype representation changes (e.g., pandas 3.0 changed string dtype from `object` to `StringDtype` and datetime resolution from `ns` to `us`).
+
+**Version** is defined in `locopy/_version.py` and read dynamically by setuptools via `pyproject.toml` (`[tool.setuptools.dynamic]`).
+
+## Code Style
+
+- Linter/formatter: **ruff** (config in `pyproject.toml`). Pre-commit hooks enforce ruff + trailing whitespace + debug statements.
+- Docstring convention: **numpy style** (`[tool.ruff.lint.pydocstyle] convention = "numpy"`)
+- Relative imports are banned (`ban-relative-imports = "all"`)
+- Target Python version: 3.12 (ruff target)
+
+## Test Markers
+
+- `@pytest.mark.integration` - Integration tests requiring real DB/S3 connections (skipped in CI unit test runs)

locopy/_version.py

@@ -14,4 +14,4 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-__version__ = "0.6.8"
+__version__ = "0.7.0"

locopy/database.py

@@ -17,7 +17,7 @@
 """Database Module."""
 
 import time
-from typing import Dict, Generator, List, Optional, Union
+from typing import Dict, Generator, List
 
 import pandas
 import polars
@@ -73,8 +73,8 @@ class Database:
     def __init__(
         self,
         dbapi: object,
-        config_yaml: Optional[str] = None,
-        **kwargs: Union[str, int],
+        config_yaml: str | None = None,
+        **kwargs: str | int,
     ) -> None:
         self.dbapi = dbapi
         self.connection = kwargs or {}
@@ -205,8 +205,8 @@ def column_names(self) -> List[str]:
             return [column[0].lower() for column in self.cursor.description]
 
     def to_dataframe(
-        self, df_type: str = "pandas", size: Optional[int] = None
-    ) -> Optional[Union[pandas.DataFrame, polars.DataFrame]]:
+        self, df_type: str = "pandas", size: int | None = None
+    ) -> pandas.DataFrame | polars.DataFrame | None:
         """Return a dataframe of the last query results.
 
         Parameters
@@ -244,7 +244,7 @@ def to_dataframe(
         elif df_type == "polars":
             return polars.DataFrame(fetched, schema=columns, orient="row")
 
-    def to_dict(self) -> Generator[Dict[str, Union[str, int, float]], None, None]:
+    def to_dict(self) -> Generator[Dict[str, str | int | float], None, None]:
         """Generate dictionaries of rows.
 
         Yields
@@ -254,7 +254,7 @@ def to_dict(self) -> Generator[Dict[str, Union[str, int, float]], None, None]:
         """
         columns = self.column_names()
         for row in self.cursor:
-            yield dict(zip(columns, row))
+            yield dict(zip(columns, row, strict=False))
 
     def _is_connected(self) -> bool:
         """Check the connection and cursor class arrtributes are initalized.

locopy/s3.py

@@ -21,7 +21,7 @@
 """
 
 import os
-from typing import List, Optional, Tuple
+from typing import List, Tuple
 
 from boto3 import Session
 from boto3.s3.transfer import TransferConfig
@@ -88,7 +88,7 @@ class S3:
     """
 
     def __init__(
-        self, profile: Optional[str] = None, kms_key: Optional[str] = None, **kwargs
+        self, profile: str | None = None, kms_key: str | None = None, **kwargs
     ) -> None:
         self.profile = profile
         self.kms_key = kms_key
@@ -153,7 +153,7 @@ def _generate_s3_path(self, bucket: str, key: str) -> str:
         """
         return f"s3://{bucket}/{key}"
 
-    def _generate_unload_path(self, bucket: str, folder: Optional[str]) -> str:
+    def _generate_unload_path(self, bucket: str, folder: str | None) -> str:
         """Return the S3 file URL.
 
         If a valid (not None) folder is provided, returns in the format s3://bucket/folder.
@@ -226,7 +226,7 @@ def upload_to_s3(self, local: str, bucket: str, key: str) -> None:
             raise S3UploadError("Error uploading to S3.") from e
 
     def upload_list_to_s3(
-        self, local_list: List[str], bucket: str, folder: Optional[str] = None
+        self, local_list: List[str], bucket: str, folder: str | None = None
     ) -> List[str]:
         """
         Upload a list of files to a S3 bucket.
@@ -300,7 +300,7 @@ def download_from_s3(self, bucket: str, key: str, local: str) -> None:
             raise S3DownloadError("Error downloading from S3.") from e
 
     def download_list_from_s3(
-        self, s3_list: List[str], local_path: Optional[str] = None
+        self, s3_list: List[str], local_path: str | None = None
     ) -> List[str]:
         """
         Download a list of files from s3.

locopy/utility.py

@@ -27,7 +27,7 @@
 from collections import OrderedDict
 from functools import singledispatch
 from itertools import cycle
-from typing import Dict, List, Union
+from typing import Dict, List
 
 import pandas as pd
 import polars as pl
@@ -47,7 +47,7 @@
 
 
 def write_file(
-    data: List[List[Union[str, int, float]]],
+    data: List[List[str | int | float]],
     delimiter: str,
     filepath: str,
     mode: str = "w",
@@ -225,7 +225,7 @@ def concatenate_files(
         raise LocopyConcatError("Error concateneating files.") from e
 
 
-def read_config_yaml(config_yaml: Union[str, object]) -> Dict[str, Union[str, int]]:
+def read_config_yaml(config_yaml: str | object) -> Dict[str, str | int]:
     """Read a configuration YAML file.
 
     Populate the database connection attributes, and validate required ones.
@@ -357,13 +357,13 @@ def check_column_type_pyarrow(pa_dtype):
             datatype = check_column_type_pyarrow(data.dtype.pyarrow_dtype)
             column_type.append(datatype)
         else:
-            if (data.dtype in ["datetime64[ns]", "M8[ns]"]) or (
-                re.match(r"(datetime64\[ns\,\W)([a-zA-Z]+)(\])", str(data.dtype))
-            ):
+            if pd.api.types.is_datetime64_any_dtype(data.dtype):
                 column_type.append("timestamp")
             elif str(data.dtype).lower().startswith("bool"):
                 column_type.append("boolean")
-            elif str(data.dtype).startswith("object"):
+            elif str(data.dtype).startswith("object") or isinstance(
+                data.dtype, pd.StringDtype
+            ):
                 data_type = validate_float_object(data) or validate_date_object(data)
                 if not data_type:
                     column_type.append("varchar")
@@ -376,7 +376,7 @@ def check_column_type_pyarrow(pa_dtype):
             else:
                 column_type.append("varchar")
         logger.info("Parsing column %s to %s", column, column_type[-1])
-    return OrderedDict(zip(list(dataframe.columns), column_type))
+    return OrderedDict(zip(list(dataframe.columns), column_type, strict=False))
 
 
 @find_column_type.register(pl.DataFrame)
@@ -467,7 +467,7 @@ def validate_float_object(column):
             else:
                 column_type.append(data_type)
         logger.info("Parsing column %s to %s", column, column_type[-1])
-    return OrderedDict(zip(list(dataframe.columns), column_type))
+    return OrderedDict(zip(list(dataframe.columns), column_type, strict=False))
 
 
 class ProgressPercentage: