datafun-streaming 0.2.0__tar.gz → 0.3.0__tar.gz

{datafun_streaming-0.2.0 → datafun_streaming-0.3.0}/CHANGELOG.md

@@ -13,6 +13,25 @@ and this project adheres to **[Semantic Versioning](https://semver.org/spec/v2.0
 
 ---
 
+## [0.3.0] - 2026-05-08
+
+### Added
+
+- `datafun_streaming.storage.duckdb_sql` - pure SQL string builder functions
+  (`build_create_table_sql`, `build_clear_table_sql`, `build_insert_sql`)
+  with no database connection required, fully testable in isolation
+- Tests for all three SQL builder functions in `tests/test_duckdb_sql.py`
+
+### Changed
+
+- `upsert_row` now requires caller-supplied `allowed_tables: frozenset[str]`
+  parameter - removes the module-level placeholder allowlist and gives
+  callers full control over which tables are permitted
+- Removed `_ALLOWED_TABLE_NAMES` placeholder constant from `duckdb_utils.py`
+- pytest `minversion` updated to `9.0`
+
+---
+
 ## [0.2.0] - 2026-05-08
 
 ### Changed
@@ -109,7 +128,8 @@ git push origin :refs/tags/vX.Z.Y
 
 ## Links
 
-[Unreleased]: https://github.com/denisecase/datafun-streaming/compare/v0.2.0...HEAD
+[Unreleased]: https://github.com/denisecase/datafun-streaming/compare/v0.3.0...HEAD
+[0.3.0]: https://github.com/denisecase/datafun-streaming/releases/tag/v0.3.0
 [0.2.0]: https://github.com/denisecase/datafun-streaming/releases/tag/v0.2.0
 [0.1.0]: https://github.com/denisecase/datafun-streaming/releases/tag/v0.1.0
 

{datafun_streaming-0.2.0 → datafun_streaming-0.3.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: datafun-streaming
-Version: 0.2.0
+Version: 0.3.0
 Summary: Utilities for streaming data analytics with Kafka and DuckDB.
 Project-URL: Homepage, https://github.com/denisecase/datafun-streaming
 Project-URL: Repository, https://github.com/denisecase/datafun-streaming
@@ -43,7 +43,7 @@ Description-Content-Type: text/markdown
 [![PyPI](https://img.shields.io/pypi/v/datafun-streaming?logo=pypi&label=pypi)](https://pypi.org/project/datafun-streaming/)
 [![Docs Site](https://img.shields.io/badge/docs-site-blue?logo=github)](https://denisecase.github.io/datafun-streaming/)
 [![Repo](https://img.shields.io/badge/repo-GitHub-black?logo=github)](https://github.com/denisecase/datafun-streaming)
-[![Python 3.15+](https://img.shields.io/badge/python-3.15%2B-blue?logo=python)](https://github.com/denisecase/datafun-streaming/blob/main/pyproject.toml)
+[![Python 3.14+](https://img.shields.io/badge/python-3.14%2B-blue?logo=python)](https://github.com/denisecase/datafun-streaming/blob/main/pyproject.toml)
 [![License: MIT](https://img.shields.io/badge/license-MIT-blue.svg)](LICENSE)
 
 [![CI](https://github.com/denisecase/datafun-streaming/actions/workflows/ci-python-zensical.yml/badge.svg?branch=main)](https://github.com/denisecase/datafun-streaming/actions/workflows/ci-python-zensical.yml)

{datafun_streaming-0.2.0 → datafun_streaming-0.3.0}/README.md

@@ -4,7 +4,7 @@
 [![PyPI](https://img.shields.io/pypi/v/datafun-streaming?logo=pypi&label=pypi)](https://pypi.org/project/datafun-streaming/)
 [![Docs Site](https://img.shields.io/badge/docs-site-blue?logo=github)](https://denisecase.github.io/datafun-streaming/)
 [![Repo](https://img.shields.io/badge/repo-GitHub-black?logo=github)](https://github.com/denisecase/datafun-streaming)
-[![Python 3.15+](https://img.shields.io/badge/python-3.15%2B-blue?logo=python)](https://github.com/denisecase/datafun-streaming/blob/main/pyproject.toml)
+[![Python 3.14+](https://img.shields.io/badge/python-3.14%2B-blue?logo=python)](https://github.com/denisecase/datafun-streaming/blob/main/pyproject.toml)
 [![License: MIT](https://img.shields.io/badge/license-MIT-blue.svg)](LICENSE)
 
 [![CI](https://github.com/denisecase/datafun-streaming/actions/workflows/ci-python-zensical.yml/badge.svg?branch=main)](https://github.com/denisecase/datafun-streaming/actions/workflows/ci-python-zensical.yml)

{datafun_streaming-0.2.0 → datafun_streaming-0.3.0}/pyproject.toml

@@ -110,7 +110,7 @@ reportCallIssue = "none"
 
 [tool.pytest.ini_options]
 # WHY: Consistent test discovery and coverage visibility.
-minversion = "7.0"
+minversion = "9.0"
 testpaths = ["tests"]
 addopts = "--cov=datafun_streaming --cov-report=term-missing --cov-fail-under=50"
 
@@ -226,5 +226,5 @@ packages = ["src/datafun_streaming"]  # REQ.PACKAGES: Discovery rooted at src/.
 [tool.hatch.version]
 # WHY: Version derived from git tags at build time, no manual source file edits.
 source = "vcs"
-fallback-version = "0.2.0"              # Used when no git tags present (fresh clone, CI).
+fallback-version = "0.3.0"              # Used when no git tags present (fresh clone, CI).
 tag-pattern = "^(?:v)?(?P<version>\\d+\\.\\d+\\.\\d+)$"

{datafun_streaming-0.2.0 → datafun_streaming-0.3.0}/src/datafun_streaming/_version.py

@@ -18,7 +18,7 @@ version_tuple: tuple[int | str, ...]
 commit_id: str | None
 __commit_id__: str | None
 
-__version__ = version = '0.2.0'
-__version_tuple__ = version_tuple = (0, 2, 0)
+__version__ = version = '0.3.0'
+__version_tuple__ = version_tuple = (0, 3, 0)
 
 __commit_id__ = commit_id = None

datafun_streaming-0.3.0/src/datafun_streaming/storage/duckdb_sql.py

@@ -0,0 +1,83 @@
+"""src/datafun_streaming/storage/duckdb_sql.py.
+
+Pure SQL string builders for DuckDB streaming tables.
+
+These functions build SQL statements as strings.
+They require no database connection and have no side effects.
+All use module-level constant table names, never raw user input.
+
+Author: Denise Case
+Date: 2026-05
+"""
+
+# === EXPORTS ===
+
+__all__ = [
+    "build_create_table_sql",
+    "build_clear_table_sql",
+    "build_insert_sql",
+]
+
+
+# === DEFINE SQL BUILDER FUNCTIONS ===
+
+
+def build_create_table_sql(table_name: str, fieldnames: list[str]) -> str:
+    """Build a DuckDB CREATE TABLE IF NOT EXISTS statement.
+
+    All columns are declared as VARCHAR.
+    Use this to initialize tables before writing records.
+
+    Arguments:
+        table_name: The table to create.
+        fieldnames: The field names to include as VARCHAR columns.
+
+    Returns:
+        A CREATE TABLE IF NOT EXISTS SQL string.
+
+    Example:
+        sql = build_create_table_sql("sales", ["order_id", "region_id"])
+        connection.execute(sql)
+    """
+    columns = ", ".join(f"{field} VARCHAR" for field in fieldnames)
+    return f"CREATE TABLE IF NOT EXISTS {table_name} ({columns})"  # noqa: S608 - caller responsible for validated table name
+
+
+def build_clear_table_sql(table_name: str) -> str:
+    """Build a DuckDB DELETE statement to clear all rows from a table.
+
+    Use this at the start of a consumer run to reset the table.
+
+    Arguments:
+        table_name: The table to clear.
+
+    Returns:
+        A DELETE FROM SQL string.
+
+    Example:
+        sql = build_clear_table_sql("sales")
+        connection.execute(sql)
+    """
+    return f"DELETE FROM {table_name}"  # noqa: S608 - caller responsible for validated table name
+
+
+def build_insert_sql(table_name: str, fieldnames: list[str]) -> str:
+    """Build a DuckDB INSERT statement with ? parameter placeholders.
+
+    The returned SQL expects one ? per field, passed as a list of values
+    to connection.execute().
+
+    Arguments:
+        table_name: The table to insert into.
+        fieldnames: The field names to insert.
+
+    Returns:
+        An INSERT INTO SQL string with ? placeholders.
+
+    Example:
+        sql = build_insert_sql("sales", ["order_id", "region_id"])
+        connection.execute(sql, ["S001", "US-MO"])
+    """
+    columns = ", ".join(fieldnames)
+    placeholders = ", ".join("?" for _ in fieldnames)
+    return f"INSERT INTO {table_name} ({columns}) VALUES ({placeholders})"  # noqa: S608 - caller responsible for validated table name

{datafun_streaming-0.2.0 → datafun_streaming-0.3.0}/src/datafun_streaming/storage/duckdb_utils.py

@@ -45,16 +45,6 @@ _DUCKDB_TYPE_MAP: dict[type, str] = {
     bool: "BOOLEAN",
 }
 
-# === DECLARE SQL-SAFE TABLE NAMES ===
-
-_ALLOWED_TABLE_NAMES: frozenset[str] = frozenset(
-    {
-        "valid_table",
-        "rejected_table",
-    }
-)
-
-
 # === DEFINE FUNCTIONS ===