catalogkit-lineage 0.1.2__tar.gz

catalogkit_lineage-0.1.2/PKG-INFO

@@ -0,0 +1,114 @@
+Metadata-Version: 2.4
+Name: catalogkit-lineage
+Version: 0.1.2
+Summary: Build project-level SQL lineage artifacts from dbt manifests or SQL folders.
+Author: ClearMetric Labs
+License-Expression: Apache-2.0
+Project-URL: Homepage, https://github.com/Clearmetric-Labs/CatalogKit
+Project-URL: Source, https://github.com/Clearmetric-Labs/CatalogKit
+Project-URL: Issues, https://github.com/Clearmetric-Labs/CatalogKit/issues
+Keywords: lineage,dbt,sql,catalog,sqlglot
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Developers
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3 :: Only
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Classifier: Topic :: Database
+Classifier: Topic :: Software Development :: Libraries :: Python Modules
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+Requires-Dist: catalogkit-core>=0.1.2
+Requires-Dist: sqlglot>=25.0.0
+Provides-Extra: dev
+Requires-Dist: pytest>=7.0.0; extra == "dev"
+Provides-Extra: release
+Requires-Dist: build>=1.2.2; extra == "release"
+Requires-Dist: twine>=5.1.1; extra == "release"
+
+# catalogkit-lineage
+
+`catalogkit-lineage` builds project-level SQL lineage from either:
+
+- a dbt `manifest.json` with compiled SQL available, or
+- a folder of `.sql` files
+
+It is a headless static-analysis tool:
+
+- input: one dbt manifest path or one SQL folder
+- output: a deterministic `LineageMap` plus the shared `CatalogArtifact`
+- no warehouse credentials
+- no dbt execution
+- no AI key
+
+## Install
+
+```bash
+python -m pip install catalogkit-lineage
+```
+
+## Imports
+
+```python
+from catalogkit.lineage import (
+    build_catalog_artifact,
+    build_lineage_map,
+    build_openlineage_export,
+    render_json,
+    render_text,
+    trace_downstream,
+    trace_upstream,
+)
+```
+
+For local development:
+
+```bash
+python -m pip install -e ../catalogkit-core
+python -m pip install -e ".[dev,release]"
+```
+
+## Quickstart
+
+Manifest input:
+
+```bash
+catalogkit-lineage --dialect postgres ./examples/jaffle_shop/manifest.json
+catalogkit-lineage --dialect postgres --format json ./examples/jaffle_shop/manifest.json
+catalogkit-lineage --dialect postgres --format openlineage ./examples/jaffle_shop/manifest.json
+```
+
+Folder input:
+
+```bash
+catalogkit-lineage --dialect postgres ./examples/sql_folder
+catalogkit-lineage --dialect postgres --upstream customers_report.customer_lifetime_value ./examples/sql_folder
+catalogkit-lineage --dialect postgres --downstream orders_base.amount ./examples/sql_folder
+```
+
+## Output Contract
+
+`catalogkit-lineage` exposes a module-specific `LineageMap` with:
+
+- `version`
+- `summary`
+- `nodes`
+- `edges`
+- `warnings`
+
+For CatalogKit composition, the package also exposes a shared
+`CatalogArtifact` builder backed by `catalogkit-core`.
+
+The shared core artifact contains:
+
+- `version`
+- `nodes`
+- `edges`
+- `warnings`
+
+## Contract Docs
+
+- [`../catalogkit-core/docs/contract.md`](../catalogkit-core/docs/contract.md)
+- [`docs/limitations.md`](docs/limitations.md)

catalogkit_lineage-0.1.2/README.md

@@ -0,0 +1,84 @@
+# catalogkit-lineage
+
+`catalogkit-lineage` builds project-level SQL lineage from either:
+
+- a dbt `manifest.json` with compiled SQL available, or
+- a folder of `.sql` files
+
+It is a headless static-analysis tool:
+
+- input: one dbt manifest path or one SQL folder
+- output: a deterministic `LineageMap` plus the shared `CatalogArtifact`
+- no warehouse credentials
+- no dbt execution
+- no AI key
+
+## Install
+
+```bash
+python -m pip install catalogkit-lineage
+```
+
+## Imports
+
+```python
+from catalogkit.lineage import (
+    build_catalog_artifact,
+    build_lineage_map,
+    build_openlineage_export,
+    render_json,
+    render_text,
+    trace_downstream,
+    trace_upstream,
+)
+```
+
+For local development:
+
+```bash
+python -m pip install -e ../catalogkit-core
+python -m pip install -e ".[dev,release]"
+```
+
+## Quickstart
+
+Manifest input:
+
+```bash
+catalogkit-lineage --dialect postgres ./examples/jaffle_shop/manifest.json
+catalogkit-lineage --dialect postgres --format json ./examples/jaffle_shop/manifest.json
+catalogkit-lineage --dialect postgres --format openlineage ./examples/jaffle_shop/manifest.json
+```
+
+Folder input:
+
+```bash
+catalogkit-lineage --dialect postgres ./examples/sql_folder
+catalogkit-lineage --dialect postgres --upstream customers_report.customer_lifetime_value ./examples/sql_folder
+catalogkit-lineage --dialect postgres --downstream orders_base.amount ./examples/sql_folder
+```
+
+## Output Contract
+
+`catalogkit-lineage` exposes a module-specific `LineageMap` with:
+
+- `version`
+- `summary`
+- `nodes`
+- `edges`
+- `warnings`
+
+For CatalogKit composition, the package also exposes a shared
+`CatalogArtifact` builder backed by `catalogkit-core`.
+
+The shared core artifact contains:
+
+- `version`
+- `nodes`
+- `edges`
+- `warnings`
+
+## Contract Docs
+
+- [`../catalogkit-core/docs/contract.md`](../catalogkit-core/docs/contract.md)
+- [`docs/limitations.md`](docs/limitations.md)

catalogkit_lineage-0.1.2/pyproject.toml

@@ -0,0 +1,65 @@
+[build-system]
+requires = ["setuptools>=61.0", "wheel"]
+build-backend = "setuptools.build_meta"
+
+[project]
+name = "catalogkit-lineage"
+dynamic = ["version"]
+description = "Build project-level SQL lineage artifacts from dbt manifests or SQL folders."
+readme = "README.md"
+requires-python = ">=3.10"
+license = "Apache-2.0"
+authors = [
+    {name = "ClearMetric Labs"},
+]
+keywords = [
+    "lineage",
+    "dbt",
+    "sql",
+    "catalog",
+    "sqlglot",
+]
+classifiers = [
+    "Development Status :: 3 - Alpha",
+    "Intended Audience :: Developers",
+    "Programming Language :: Python :: 3",
+    "Programming Language :: Python :: 3 :: Only",
+    "Programming Language :: Python :: 3.10",
+    "Programming Language :: Python :: 3.11",
+    "Programming Language :: Python :: 3.12",
+    "Programming Language :: Python :: 3.13",
+    "Topic :: Database",
+    "Topic :: Software Development :: Libraries :: Python Modules",
+]
+dependencies = [
+    "catalogkit-core>=0.1.2",
+    "sqlglot>=25.0.0",
+]
+
+[project.optional-dependencies]
+dev = [
+    "pytest>=7.0.0",
+]
+release = [
+    "build>=1.2.2",
+    "twine>=5.1.1",
+]
+
+[project.scripts]
+catalogkit-lineage = "catalogkit.lineage.cli:main"
+
+[project.urls]
+Homepage = "https://github.com/Clearmetric-Labs/CatalogKit"
+Source = "https://github.com/Clearmetric-Labs/CatalogKit"
+Issues = "https://github.com/Clearmetric-Labs/CatalogKit/issues"
+
+[tool.setuptools.dynamic]
+version = {attr = "catalogkit.lineage._version.__version__"}
+
+[tool.setuptools.package-dir]
+"" = "src"
+
+[tool.setuptools.packages.find]
+where = ["src"]
+include = ["catalogkit.lineage*"]
+namespaces = true

catalogkit_lineage-0.1.2/setup.cfg

@@ -0,0 +1,4 @@
+[egg_info]
+tag_build = 
+tag_date = 0
+

catalogkit_lineage-0.1.2/src/catalogkit/lineage/__init__.py

@@ -0,0 +1,36 @@
+"""Public package surface for catalogkit-lineage."""
+
+from __future__ import annotations
+
+from catalogkit.core import CatalogArtifact
+
+from ._version import __version__
+from .api import (
+    build_catalog_artifact,
+    build_lineage_map,
+    build_openlineage_export,
+    render_json,
+    render_text,
+    trace_downstream,
+    trace_upstream,
+)
+from .errors import LineageContractError, LineageError, LineageInputError
+from .models import LineageMap, LineageSummary, TraversalResult
+
+__all__ = [
+    "__version__",
+    "build_catalog_artifact",
+    "build_lineage_map",
+    "build_openlineage_export",
+    "CatalogArtifact",
+    "LineageContractError",
+    "LineageError",
+    "LineageInputError",
+    "LineageMap",
+    "LineageSummary",
+    "render_json",
+    "render_text",
+    "trace_downstream",
+    "trace_upstream",
+    "TraversalResult",
+]

catalogkit_lineage-0.1.2/src/catalogkit/lineage/__main__.py

@@ -0,0 +1,8 @@
+"""Module entrypoint for catalogkit-lineage."""
+
+from __future__ import annotations
+
+from .cli import main
+
+if __name__ == "__main__":
+    raise SystemExit(main())

catalogkit_lineage-0.1.2/src/catalogkit/lineage/_version.py

@@ -0,0 +1,3 @@
+"""Package version for catalogkit-lineage."""
+
+__version__ = "0.1.2"

catalogkit_lineage-0.1.2/src/catalogkit/lineage/api.py

@@ -0,0 +1,71 @@
+"""Public API for catalogkit-lineage."""
+
+from __future__ import annotations
+
+from pathlib import Path
+from typing import Any
+
+from catalogkit.core import CatalogArtifact
+
+from .build import (
+    build_catalog_artifact_from_project,
+    build_lineage_map_from_project,
+    build_openlineage_export_from_project,
+    trace_downstream_from_project,
+    trace_upstream_from_project,
+)
+from .loaders import load_project
+from .models import LineageMap, TraversalResult
+from .render.json import render_json
+from .render.text import render_text
+
+
+def build_lineage_map(path: str | Path, *, dialect: str) -> LineageMap:
+    """Build the public catalogkit-lineage artifact for one project input."""
+    project = load_project(path, dialect=dialect)
+    return build_lineage_map_from_project(project, dialect=dialect)
+
+
+def build_catalog_artifact(path: str | Path, *, dialect: str) -> CatalogArtifact:
+    """Build the shared catalog artifact for CatalogKit composition."""
+    project = load_project(path, dialect=dialect)
+    return build_catalog_artifact_from_project(project, dialect=dialect)
+
+
+def trace_upstream(
+    path: str | Path,
+    *,
+    dialect: str,
+    selection: str,
+) -> TraversalResult:
+    """Trace upstream column lineage for one selected dataset column."""
+    project = load_project(path, dialect=dialect)
+    return trace_upstream_from_project(project, dialect=dialect, selection=selection)
+
+
+def trace_downstream(
+    path: str | Path,
+    *,
+    dialect: str,
+    selection: str,
+) -> TraversalResult:
+    """Trace downstream column lineage for one selected dataset column."""
+    project = load_project(path, dialect=dialect)
+    return trace_downstream_from_project(project, dialect=dialect, selection=selection)
+
+
+def build_openlineage_export(path: str | Path, *, dialect: str) -> dict[str, Any]:
+    """Build an OpenLineage-compatible export view for one project input."""
+    project = load_project(path, dialect=dialect)
+    return build_openlineage_export_from_project(project, dialect=dialect)
+
+
+__all__ = [
+    "build_catalog_artifact",
+    "build_lineage_map",
+    "build_openlineage_export",
+    "render_json",
+    "render_text",
+    "trace_downstream",
+    "trace_upstream",
+]