dagua 0.0.0__tar.gz

dagua-0.0.0/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 John Mark Taylor
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

dagua-0.0.0/PKG-INFO

@@ -0,0 +1,58 @@
+Metadata-Version: 2.4
+Name: dagua
+Version: 0.0.0
+Summary: GPU-accelerated differentiable graph layout engine built on PyTorch
+Author: John Mark Taylor
+License-Expression: MIT
+Project-URL: Homepage, https://github.com/johnmarktaylor91/dagua
+Project-URL: Repository, https://github.com/johnmarktaylor91/dagua
+Project-URL: Issues, https://github.com/johnmarktaylor91/dagua/issues
+Keywords: graph,layout,visualization,pytorch,dag,graphviz,hierarchical
+Classifier: Development Status :: 2 - Pre-Alpha
+Classifier: Intended Audience :: Developers
+Classifier: Intended Audience :: Science/Research
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.9
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Topic :: Scientific/Engineering :: Visualization
+Classifier: Topic :: Software Development :: Libraries :: Python Modules
+Requires-Python: >=3.9
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: torch>=1.9
+Provides-Extra: render
+Requires-Dist: matplotlib>=3.5; extra == "render"
+Provides-Extra: graphviz
+Requires-Dist: graphviz>=0.20; extra == "graphviz"
+Provides-Extra: dev
+Requires-Dist: matplotlib>=3.5; extra == "dev"
+Requires-Dist: pytest>=7.0; extra == "dev"
+Requires-Dist: pytest-cov; extra == "dev"
+Requires-Dist: ruff>=0.4; extra == "dev"
+Requires-Dist: mypy; extra == "dev"
+Requires-Dist: pip-audit; extra == "dev"
+Requires-Dist: pre-commit; extra == "dev"
+Provides-Extra: test
+Requires-Dist: pytest>=7.0; extra == "test"
+Requires-Dist: matplotlib>=3.5; extra == "test"
+Dynamic: license-file
+
+# dagua
+
+GPU-accelerated differentiable graph layout engine built on PyTorch.
+
+**DAG + agua.** Directed acyclic graphs + water. Named after the Dagua River in Colombia — a river flows downhill (like a DAG), never cycles back (acyclic), and finds its own path through the landscape (like gradient descent finding optimal node positions).
+
+## Why?
+
+Graphviz has dominated graph visualization for 30 years but has hard scaling limits. No existing Python package provides pip-installable, hierarchical (Sugiyama-style) graph layout. Dagua fills this gap: `pip install dagua`, pure Python + PyTorch, GPU-accelerated, hierarchical layout with composable constraints.
+
+## Status
+
+Pre-alpha. Under active development.
+
+## License
+
+MIT

dagua-0.0.0/README.md

@@ -0,0 +1,17 @@
+# dagua
+
+GPU-accelerated differentiable graph layout engine built on PyTorch.
+
+**DAG + agua.** Directed acyclic graphs + water. Named after the Dagua River in Colombia — a river flows downhill (like a DAG), never cycles back (acyclic), and finds its own path through the landscape (like gradient descent finding optimal node positions).
+
+## Why?
+
+Graphviz has dominated graph visualization for 30 years but has hard scaling limits. No existing Python package provides pip-installable, hierarchical (Sugiyama-style) graph layout. Dagua fills this gap: `pip install dagua`, pure Python + PyTorch, GPU-accelerated, hierarchical layout with composable constraints.
+
+## Status
+
+Pre-alpha. Under active development.
+
+## License
+
+MIT

dagua-0.0.0/dagua/__init__.py

@@ -0,0 +1,9 @@
+"""Dagua: GPU-accelerated differentiable graph layout engine built on PyTorch."""
+
+__version__ = "0.0.0"
+
+# Public API will be re-exported here once modules are implemented:
+#   from dagua.elements import Node, Edge, Cluster
+#   from dagua.graph import Graph
+#   from dagua.style import NodeStyle, EdgeStyle, ClusterStyle
+#   from dagua.defaults import set_default_device, set_default_theme

dagua-0.0.0/dagua/defaults.py

@@ -0,0 +1,9 @@
+"""Module-level defaults — device, theme.
+
+Minimal global state. Prefer passing these explicitly to Graph; these exist
+only as convenience to avoid repeating device='cuda' on every call.
+
+Usage:
+    dagua.set_default_device('cuda')
+    dagua.set_default_theme('dark')
+"""

dagua-0.0.0/dagua/elements.py

@@ -0,0 +1,5 @@
+"""Node, Edge, Cluster dataclasses — pure structural data.
+
+These hold topology and identity, not visual styling (that's style.py).
+Elements are the atoms that Graph holds collections of.
+"""

dagua-0.0.0/dagua/graph.py

@@ -0,0 +1,6 @@
+"""Graph class — the central orchestrator.
+
+Holds nodes, edges, and clusters. Maintains ID→index mapping.
+Provides layout() and render() orchestration methods.
+from_* classmethods delegate to io.py for construction from various formats.
+"""

dagua-0.0.0/dagua/io.py

@@ -0,0 +1,8 @@
+"""Graph construction and export utilities.
+
+from_edges, from_edge_index, from_networkx, from_dict — thin converters
+that build Graph instances from various input formats.
+to_dot — export to Graphviz DOT format.
+
+Graph.from_* classmethods are thin wrappers over functions here.
+"""

dagua-0.0.0/dagua/layout/__init__.py

@@ -0,0 +1,4 @@
+"""Layout subpackage — differentiable graph layout via PyTorch optimization.
+
+Re-exports the public API: layout(), individual constraint classes.
+"""

dagua-0.0.0/dagua/layout/constraints.py

@@ -0,0 +1,14 @@
+"""Composable constraint loss functions.
+
+Each constraint is a callable: (pos, graph_data) -> scalar loss.
+
+Built-in constraints:
+  - DAG: enforce top-to-bottom ordering for directed edges
+  - Repel: prevent node overlap via pairwise repulsion
+  - Attract: pull connected nodes together (edge attraction)
+  - Overlap: hard overlap penalty (complements projection.py)
+  - Cluster: group related nodes spatially
+  - Align: rank/axis alignment for specific node groups
+
+Users can write custom constraints in ~3 lines by following this protocol.
+"""

dagua-0.0.0/dagua/layout/engine.py

@@ -0,0 +1,8 @@
+"""Core optimization loop — the heart of dagua (~200 lines of PyTorch).
+
+Takes edge_index, node_sizes, groups as tensors (NOT a Graph object).
+Initializes positions as learnable parameters, runs optimization with
+composite loss from constraints, returns detached position tensor.
+
+Headless design: independently testable without the Graph abstraction.
+"""

dagua-0.0.0/dagua/layout/projection.py

@@ -0,0 +1,5 @@
+"""Hard overlap resolution via projected gradient descent.
+
+After each optimizer step, projects positions to satisfy hard constraints
+(e.g., no node overlap). This is the "projection" in projected gradient descent.
+"""

dagua-0.0.0/dagua/layout/schedule.py

@@ -0,0 +1,6 @@
+"""Annealing schedules for constraint weights.
+
+Controls how constraint weights evolve during optimization — e.g., start with
+strong DAG ordering, then gradually increase repulsion. Enables curriculum-style
+layout optimization.
+"""

dagua-0.0.0/dagua/render/__init__.py

@@ -0,0 +1,6 @@
+"""Render subpackage — multiple output backends.
+
+Re-exports: render(), to_svg().
+Renderers accept structured data (positions, elements, styles), not Graph objects.
+Graph.render() is a thin wrapper that extracts data and calls the appropriate renderer.
+"""

dagua-0.0.0/dagua/render/graphviz.py

@@ -0,0 +1,5 @@
+"""Optional Graphviz passthrough — neato -n2 with fixed positions.
+
+Writes a .dot file with pre-computed positions, renders via Graphviz's neato
+engine with -n2 flag (use given positions). Requires graphviz Python package.
+"""

dagua-0.0.0/dagua/render/mpl.py

@@ -0,0 +1,5 @@
+"""Matplotlib renderer — batched rendering via PatchCollection + LineCollection.
+
+Default renderer. Requires matplotlib (optional dependency).
+Renders nodes as patches, edges as line collections, labels as text.
+"""

dagua-0.0.0/dagua/render/svg.py

@@ -0,0 +1,5 @@
+"""Direct SVG string output — zero external dependencies.
+
+Generates SVG markup directly. Jupyter-friendly (displays inline).
+Useful when matplotlib is not available or not desired.
+"""

dagua-0.0.0/dagua/routing.py

@@ -0,0 +1,5 @@
+"""Bezier edge routing — computes control points after layout.
+
+Tier 1: heuristic routing (orthogonal + bezier curves).
+Tier 2 (future): differentiable routing integrated into the optimization loop.
+"""

dagua-0.0.0/dagua/style.py

@@ -0,0 +1,5 @@
+"""NodeStyle, EdgeStyle, ClusterStyle — visual presentation data.
+
+Themes, palettes, auto-coloring. Separate from elements.py because styling
+concerns grow independently of structural concerns.
+"""

dagua-0.0.0/dagua/utils.py

@@ -0,0 +1,4 @@
+"""Shared utilities — text measurement, graph topology helpers.
+
+Small pure functions used across modules. No heavy dependencies.
+"""

dagua-0.0.0/dagua.egg-info/PKG-INFO

@@ -0,0 +1,58 @@
+Metadata-Version: 2.4
+Name: dagua
+Version: 0.0.0
+Summary: GPU-accelerated differentiable graph layout engine built on PyTorch
+Author: John Mark Taylor
+License-Expression: MIT
+Project-URL: Homepage, https://github.com/johnmarktaylor91/dagua
+Project-URL: Repository, https://github.com/johnmarktaylor91/dagua
+Project-URL: Issues, https://github.com/johnmarktaylor91/dagua/issues
+Keywords: graph,layout,visualization,pytorch,dag,graphviz,hierarchical
+Classifier: Development Status :: 2 - Pre-Alpha
+Classifier: Intended Audience :: Developers
+Classifier: Intended Audience :: Science/Research
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.9
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Topic :: Scientific/Engineering :: Visualization
+Classifier: Topic :: Software Development :: Libraries :: Python Modules
+Requires-Python: >=3.9
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: torch>=1.9
+Provides-Extra: render
+Requires-Dist: matplotlib>=3.5; extra == "render"
+Provides-Extra: graphviz
+Requires-Dist: graphviz>=0.20; extra == "graphviz"
+Provides-Extra: dev
+Requires-Dist: matplotlib>=3.5; extra == "dev"
+Requires-Dist: pytest>=7.0; extra == "dev"
+Requires-Dist: pytest-cov; extra == "dev"
+Requires-Dist: ruff>=0.4; extra == "dev"
+Requires-Dist: mypy; extra == "dev"
+Requires-Dist: pip-audit; extra == "dev"
+Requires-Dist: pre-commit; extra == "dev"
+Provides-Extra: test
+Requires-Dist: pytest>=7.0; extra == "test"
+Requires-Dist: matplotlib>=3.5; extra == "test"
+Dynamic: license-file
+
+# dagua
+
+GPU-accelerated differentiable graph layout engine built on PyTorch.
+
+**DAG + agua.** Directed acyclic graphs + water. Named after the Dagua River in Colombia — a river flows downhill (like a DAG), never cycles back (acyclic), and finds its own path through the landscape (like gradient descent finding optimal node positions).
+
+## Why?
+
+Graphviz has dominated graph visualization for 30 years but has hard scaling limits. No existing Python package provides pip-installable, hierarchical (Sugiyama-style) graph layout. Dagua fills this gap: `pip install dagua`, pure Python + PyTorch, GPU-accelerated, hierarchical layout with composable constraints.
+
+## Status
+
+Pre-alpha. Under active development.
+
+## License
+
+MIT

dagua-0.0.0/dagua.egg-info/SOURCES.txt

@@ -0,0 +1,30 @@
+LICENSE
+README.md
+pyproject.toml
+dagua/__init__.py
+dagua/defaults.py
+dagua/elements.py
+dagua/graph.py
+dagua/io.py
+dagua/routing.py
+dagua/style.py
+dagua/utils.py
+dagua.egg-info/PKG-INFO
+dagua.egg-info/SOURCES.txt
+dagua.egg-info/dependency_links.txt
+dagua.egg-info/requires.txt
+dagua.egg-info/top_level.txt
+dagua/layout/__init__.py
+dagua/layout/constraints.py
+dagua/layout/engine.py
+dagua/layout/projection.py
+dagua/layout/schedule.py
+dagua/render/__init__.py
+dagua/render/graphviz.py
+dagua/render/mpl.py
+dagua/render/svg.py
+tests/test_elements.py
+tests/test_graph.py
+tests/test_io.py
+tests/test_routing.py
+tests/test_style.py

dagua-0.0.0/dagua.egg-info/dependency_links.txt

@@ -0,0 +1 @@
+

dagua-0.0.0/dagua.egg-info/requires.txt

@@ -0,0 +1,20 @@
+torch>=1.9
+
+[dev]
+matplotlib>=3.5
+pytest>=7.0
+pytest-cov
+ruff>=0.4
+mypy
+pip-audit
+pre-commit
+
+[graphviz]
+graphviz>=0.20
+
+[render]
+matplotlib>=3.5
+
+[test]
+pytest>=7.0
+matplotlib>=3.5

dagua-0.0.0/dagua.egg-info/top_level.txt

@@ -0,0 +1 @@
+dagua

dagua-0.0.0/pyproject.toml

@@ -0,0 +1,107 @@
+[build-system]
+requires = ["setuptools>=64"]
+build-backend = "setuptools.build_meta"
+
+[project]
+name = "dagua"
+version = "0.0.0"
+description = "GPU-accelerated differentiable graph layout engine built on PyTorch"
+readme = "README.md"
+license = "MIT"
+requires-python = ">=3.9"
+authors = [
+    {name = "John Mark Taylor"},
+]
+keywords = ["graph", "layout", "visualization", "pytorch", "dag", "graphviz", "hierarchical"]
+classifiers = [
+    "Development Status :: 2 - Pre-Alpha",
+    "Intended Audience :: Developers",
+    "Intended Audience :: Science/Research",
+    "Programming Language :: Python :: 3",
+    "Programming Language :: Python :: 3.9",
+    "Programming Language :: Python :: 3.10",
+    "Programming Language :: Python :: 3.11",
+    "Programming Language :: Python :: 3.12",
+    "Topic :: Scientific/Engineering :: Visualization",
+    "Topic :: Software Development :: Libraries :: Python Modules",
+]
+dependencies = [
+    "torch>=1.9",
+]
+
+[project.optional-dependencies]
+render = ["matplotlib>=3.5"]
+graphviz = ["graphviz>=0.20"]
+dev = [
+    "matplotlib>=3.5",
+    "pytest>=7.0",
+    "pytest-cov",
+    "ruff>=0.4",
+    "mypy",
+    "pip-audit",
+    "pre-commit",
+]
+test = [
+    "pytest>=7.0",
+    "matplotlib>=3.5",
+]
+
+[project.urls]
+Homepage = "https://github.com/johnmarktaylor91/dagua"
+Repository = "https://github.com/johnmarktaylor91/dagua"
+Issues = "https://github.com/johnmarktaylor91/dagua/issues"
+
+[tool.setuptools.packages.find]
+include = ["dagua*"]
+
+[tool.pytest.ini_options]
+testpaths = ["tests"]
+markers = [
+    "smoke: quick sanity checks",
+    "slow: tests that take >10s",
+    "gpu: tests requiring CUDA",
+]
+
+[tool.coverage.run]
+source = ["dagua"]
+branch = true
+
+[tool.coverage.report]
+show_missing = true
+skip_empty = true
+
+[tool.coverage.html]
+directory = "tests/test_outputs/reports/coverage_html"
+
+[tool.coverage.lcov]
+output = "tests/test_outputs/coverage.lcov"
+
+[tool.mypy]
+python_version = "3.11"
+warn_return_any = false
+warn_unused_configs = true
+ignore_missing_imports = true
+check_untyped_defs = false
+disallow_untyped_defs = false
+
+[tool.ruff]
+line-length = 100
+target-version = "py39"
+
+[tool.ruff.lint]
+select = ["E", "F", "W", "I"]
+
+[tool.semantic_release]
+version_toml = ["pyproject.toml:project.version"]
+version_variables = ["dagua/__init__.py:__version__"]
+branch = "main"
+major_on_zero = true
+build_command = "pip install build && python -m build"
+commit_message = "chore(release): {version}"
+
+[tool.semantic_release.remote]
+type = "github"
+token = { env = "GH_TOKEN" }
+
+[tool.semantic_release.publish]
+upload_to_vcs_release = true

dagua-0.0.0/setup.cfg

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

dagua-0.0.0/tests/test_elements.py

@@ -0,0 +1 @@
+"""Tests for Node, Edge, Cluster dataclasses."""

dagua-0.0.0/tests/test_graph.py

@@ -0,0 +1 @@
+"""Tests for Graph class — construction, ID mapping, orchestration."""

dagua-0.0.0/tests/test_io.py

@@ -0,0 +1 @@
+"""Tests for io.py — from_edges, from_edge_index, from_networkx, to_dot."""

dagua-0.0.0/tests/test_routing.py

@@ -0,0 +1 @@
+"""Tests for bezier edge routing."""

dagua-0.0.0/tests/test_style.py

@@ -0,0 +1 @@
+"""Tests for NodeStyle, EdgeStyle, ClusterStyle, themes."""