scgraph 3.5.0__tar.gz → 3.6.0b2__tar.gz

scgraph-3.6.0b2/CMakeLists.txt

@@ -0,0 +1,48 @@
+cmake_minimum_required(VERSION 3.15)
+
+# -----------------------------
+# Define The Project
+# -----------------------------
+project(scgraph LANGUAGES CXX)
+
+# -----------------------------
+# Find Packages
+# -----------------------------
+find_package(Python 3.10 REQUIRED COMPONENTS Interpreter Development.Module)
+find_package(nanobind CONFIG REQUIRED)
+
+# -----------------------------
+# Set C++ standard
+# -----------------------------
+set(CMAKE_CXX_STANDARD 20)
+set(CMAKE_CXX_STANDARD_REQUIRED ON)
+
+# -----------------------------
+# OPTIMIZATION FLAGS
+# -----------------------------
+set(CMAKE_BUILD_TYPE Release)
+if(MSVC)
+    set(CMAKE_CXX_FLAGS_RELEASE "/O2 /DNDEBUG")
+else()
+    set(CMAKE_CXX_FLAGS_RELEASE "-O3 -DNDEBUG -fno-math-errno -fomit-frame-pointer")
+endif()
+
+# -----------------------------
+# Graph module
+# -----------------------------
+nanobind_add_module(
+    cpp
+    STABLE_ABI
+    scgraph/cpp/bindings/graph_bindings.cpp
+    scgraph/cpp/src/graph.cpp
+    scgraph/cpp/src/graph_utils.cpp
+    scgraph/cpp/src/graph_reducer.cpp
+    scgraph/cpp/src/contraction_hierarchies.cpp
+    scgraph/cpp/src/transit_node_routing.cpp
+)
+install(TARGETS cpp DESTINATION scgraph)
+
+# -----------------------------
+# Install Python package files
+# -----------------------------
+install(DIRECTORY scgraph DESTINATION . FILES_MATCHING PATTERN "*.py")

{scgraph-3.5.0 → scgraph-3.6.0b2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: scgraph
-Version: 3.5.0
+Version: 3.6.0b2
 Summary: Determine an approximate distance and route between two points on earth.
 Author-Email: Connor Makowski <conmak@mit.edu>
 License-Expression: MIT
@@ -11,7 +11,7 @@ Project-URL: Homepage, https://github.com/connor-makowski/scgraph
 Project-URL: Bug Tracker, https://github.com/connor-makowski/scgraph/issues
 Project-URL: Documentation, https://connor-makowski.github.io/scgraph/scgraph.html
 Requires-Python: >=3.11
-Requires-Dist: geokdtree==1.2.4
+Requires-Dist: geokdtree>=1.3.0
 Requires-Dist: bmsspy<3.0.0,>=2.2.1
 Requires-Dist: requests<3.0.0,>=2.32.0
 Provides-Extra: dev
@@ -68,18 +68,33 @@ If you use SCGraph in your research, please cite:
 pip install scgraph
 ```
 
-A C++ extension is compiled automatically if a C++ compiler is available (~10x speedup on core algorithms). To verify: `from scgraph.utils import cpp_check; cpp_check()`.
+### Pure Python Installation (Skip Binary Builds)
 
-To skip the C++ build:
+If you want to install `scgraph` in pure Python mode without compiling or using binary extensions (or in environments without a C++ compiler):
 
-```bash
-# macOS / Linux / WSL2
-export SKBUILD_CMAKE_ARGS="-DSKIP_CPP_BUILD=ON" && pip install scgraph
-# Windows (PowerShell)
-$env:SKBUILD_CMAKE_ARGS="-DSKIP_CPP_BUILD=ON"; pip install scgraph
-# Windows (CMD)
-set SKBUILD_CMAKE_ARGS=-DSKIP_CPP_BUILD=ON && pip install scgraph
-```
+- **Linux / macOS / WSL2**:
+  ```bash
+  SCGRAPH_NO_BUILD=1 pip install --no-binary scgraph scgraph
+  ```
+  Or with `uv`:
+  ```bash
+  SCGRAPH_NO_BUILD=1 uv pip install --no-binary scgraph scgraph
+  ```
+
+- **Windows (PowerShell)**:
+  ```powershell
+  $env:SCGRAPH_NO_BUILD="1"
+  pip install --no-binary scgraph scgraph
+  ```
+
+- **Windows (CMD)**:
+  ```cmd
+  set SCGRAPH_NO_BUILD=1
+  pip install --no-binary scgraph scgraph
+  ```
+
+#### Automatic Fallback
+When building from source, if a C++ compiler is not available or compilation fails, `scgraph` will automatically fall back to a pure Python implementation.
 
 ---
 
@@ -200,9 +215,10 @@ All algorithms are available on `Graph` objects and accessible from `GeoGraph` v
 
 | `algorithm_fn` | Description | Time Complexity |
 |---|---|---|
-| `'dijkstra'` | Standard Dijkstra; general purpose, non-negative weights (default) | O((n+m) log n) |
+| `'dijkstra'` | Standard Dijkstra; general purpose, non-negative weights | O((n+m) log n) |
 | `'bidirectional_dijkstra'` | Bidirectional Dijkstra; simultaneous search from origin and destination | O((n+m) log n) |
 | `'dijkstra_buckets'` | Dijkstra with buckets (Dial's algorithm); efficient for non-negative weights (ideally >= 1) | O(n+m+W) |
+| `'bidirectional_buckets'` | Bidirectional Dijkstra (default for geographs) with buckets (Dial's algorithm); efficient for non-negative weights | O(n+m+W) |
 | `'dijkstra_negative'` | Dijkstra with cycle detection; supports negative weights | O(n·m) |
 | `'a_star'` | A* with optional heuristic; faster than Dijkstra with a good heuristic | O((n+m) log n) |
 | `'bellman_ford'` | Bellman-Ford; supports negative weights, slower than Dijkstra | O(n·m) |
@@ -215,7 +231,7 @@ All algorithms are available on `Graph` objects and accessible from `GeoGraph` v
 
 | Scenario | Recommended Approach |
 |---|---|
-| Single query | `dijkstra` (default) or `bidirectional_dijkstra` |
+| Single query | `dijkstra`, `bidirectional_buckets` or `bidirectional_dijkstra` |
 | Sparse networks (road/rail/maritime) | Reduce with `geograph.reduce()` for 2–4x faster queries |
 | Weights generally >= 1 | `dijkstra_buckets` |
 | Repeated queries from one origin | `cached_shortest_path` |
@@ -410,15 +426,10 @@ output = marnet_geograph.get_shortest_path(
 
 | `node_addition_type` | Description |
 |---|---|
-| `'kdclosest'` | Closest node via KD-tree (default, fastest) |
-| `'closest'` | Closest node via brute force |
-| `'quadrant'` | Closest node in each of 4 quadrants |
+| `'closest'` / `'kdclosest'` | Closest node via GeoKDTree (default) |
+| `'quadrant'` / `'kdquadrant'` | Closest node in each quadrant via GeoKDTree |
 | `'all'` | All nodes within the bound |
 
-| `node_addition_math` | Description |
-|---|---|
-| `'euclidean'` | Fast planar distance (default) |
-| `'haversine'` | Accurate great-circle distance |
 
 ---
 

{scgraph-3.5.0 → scgraph-3.6.0b2}/README.md

@@ -40,18 +40,33 @@ If you use SCGraph in your research, please cite:
 pip install scgraph
 ```
 
-A C++ extension is compiled automatically if a C++ compiler is available (~10x speedup on core algorithms). To verify: `from scgraph.utils import cpp_check; cpp_check()`.
-
-To skip the C++ build:
-
-```bash
-# macOS / Linux / WSL2
-export SKBUILD_CMAKE_ARGS="-DSKIP_CPP_BUILD=ON" && pip install scgraph
-# Windows (PowerShell)
-$env:SKBUILD_CMAKE_ARGS="-DSKIP_CPP_BUILD=ON"; pip install scgraph
-# Windows (CMD)
-set SKBUILD_CMAKE_ARGS=-DSKIP_CPP_BUILD=ON && pip install scgraph
-```
+### Pure Python Installation (Skip Binary Builds)
+
+If you want to install `scgraph` in pure Python mode without compiling or using binary extensions (or in environments without a C++ compiler):
+
+- **Linux / macOS / WSL2**:
+  ```bash
+  SCGRAPH_NO_BUILD=1 pip install --no-binary scgraph scgraph
+  ```
+  Or with `uv`:
+  ```bash
+  SCGRAPH_NO_BUILD=1 uv pip install --no-binary scgraph scgraph
+  ```
+
+- **Windows (PowerShell)**:
+  ```powershell
+  $env:SCGRAPH_NO_BUILD="1"
+  pip install --no-binary scgraph scgraph
+  ```
+
+- **Windows (CMD)**:
+  ```cmd
+  set SCGRAPH_NO_BUILD=1
+  pip install --no-binary scgraph scgraph
+  ```
+
+#### Automatic Fallback
+When building from source, if a C++ compiler is not available or compilation fails, `scgraph` will automatically fall back to a pure Python implementation.
 
 ---
 
@@ -172,9 +187,10 @@ All algorithms are available on `Graph` objects and accessible from `GeoGraph` v
 
 | `algorithm_fn` | Description | Time Complexity |
 |---|---|---|
-| `'dijkstra'` | Standard Dijkstra; general purpose, non-negative weights (default) | O((n+m) log n) |
+| `'dijkstra'` | Standard Dijkstra; general purpose, non-negative weights | O((n+m) log n) |
 | `'bidirectional_dijkstra'` | Bidirectional Dijkstra; simultaneous search from origin and destination | O((n+m) log n) |
 | `'dijkstra_buckets'` | Dijkstra with buckets (Dial's algorithm); efficient for non-negative weights (ideally >= 1) | O(n+m+W) |
+| `'bidirectional_buckets'` | Bidirectional Dijkstra (default for geographs) with buckets (Dial's algorithm); efficient for non-negative weights | O(n+m+W) |
 | `'dijkstra_negative'` | Dijkstra with cycle detection; supports negative weights | O(n·m) |
 | `'a_star'` | A* with optional heuristic; faster than Dijkstra with a good heuristic | O((n+m) log n) |
 | `'bellman_ford'` | Bellman-Ford; supports negative weights, slower than Dijkstra | O(n·m) |
@@ -187,7 +203,7 @@ All algorithms are available on `Graph` objects and accessible from `GeoGraph` v
 
 | Scenario | Recommended Approach |
 |---|---|
-| Single query | `dijkstra` (default) or `bidirectional_dijkstra` |
+| Single query | `dijkstra`, `bidirectional_buckets` or `bidirectional_dijkstra` |
 | Sparse networks (road/rail/maritime) | Reduce with `geograph.reduce()` for 2–4x faster queries |
 | Weights generally >= 1 | `dijkstra_buckets` |
 | Repeated queries from one origin | `cached_shortest_path` |
@@ -382,15 +398,10 @@ output = marnet_geograph.get_shortest_path(
 
 | `node_addition_type` | Description |
 |---|---|
-| `'kdclosest'` | Closest node via KD-tree (default, fastest) |
-| `'closest'` | Closest node via brute force |
-| `'quadrant'` | Closest node in each of 4 quadrants |
+| `'closest'` / `'kdclosest'` | Closest node via GeoKDTree (default) |
+| `'quadrant'` / `'kdquadrant'` | Closest node in each quadrant via GeoKDTree |
 | `'all'` | All nodes within the bound |
 
-| `node_addition_math` | Description |
-|---|---|
-| `'euclidean'` | Fast planar distance (default) |
-| `'haversine'` | Accurate great-circle distance |
 
 ---
 

{scgraph-3.5.0 → scgraph-3.6.0b2}/pyproject.toml

@@ -4,15 +4,20 @@ target-version = ['py312']
 exclude = '/.*(migrations|__pycache__|geographs).*/'
 
 [build-system]
-requires = ["scikit_build_core", "nanobind"]
+requires = ["scikit-build-core", "nanobind>=3.1.0"]
 build-backend = "scikit_build_core.build"
 
 [tool.scikit-build]
 build.verbose = false
 build-dir = "build/{wheel_tag}"
 
+[[tool.scikit-build.overrides]]
+if.env.SCGRAPH_NO_BUILD = true
+wheel.cmake = false
+
 [[tool.scikit-build.overrides]]
 if.failed = true
+if.not.env.SCGRAPH_REQUIRE_CPP = true
 wheel.cmake = false
 
 [tool.scikit-build.sdist]
@@ -30,7 +35,7 @@ include = [
 
 [project]
 name = "scgraph"
-version = "3.5.0"
+version = "3.6.0b2"
 description = "Determine an approximate distance and route between two points on earth."
 authors = [
     {name="Connor Makowski", email="conmak@mit.edu"}
@@ -44,7 +49,7 @@ classifiers = [
     "Operating System :: OS Independent",
 ]
 dependencies = [
-    "geokdtree==1.2.4",
+    "geokdtree>=1.3.0",
     "bmsspy>=2.2.1,<3.0.0",
     "requests>=2.32.0,<3.0.0"
 ]
@@ -65,6 +70,7 @@ dev = [
 [tool.pytest.ini_options]
 testpaths = ["test"]
 pythonpath = ["test"]
+python_files = ["*_test.py"]
 
 [tool.uv]
 no-build-isolation-package = []
@@ -72,4 +78,17 @@ no-build-isolation-package = []
 [project.urls]
 "Homepage" = "https://github.com/connor-makowski/scgraph"
 "Bug Tracker" = "https://github.com/connor-makowski/scgraph/issues"
-"Documentation" = "https://connor-makowski.github.io/scgraph/scgraph.html"
+"Documentation" = "https://connor-makowski.github.io/scgraph/scgraph.html"
+
+[tool.cibuildwheel]
+build = "cp311-* cp312-* cp313-* cp314-*"
+skip = "*-win32 *_i686 pp*"
+test-command = "pytest {project}/test"
+test-requires = ["pytest>=9.0.3"]
+
+[tool.cibuildwheel.linux]
+manylinux-x86_64-image = "manylinux_2_28"
+manylinux-aarch64-image = "manylinux_2_28"
+
+[tool.cibuildwheel.windows]
+repair-wheel-command = ""

{scgraph-3.5.0 → scgraph-3.6.0b2}/scgraph/__init__.py

@@ -1,4 +1,4 @@
-"""
+r"""
 # SCGraph
 [![PyPI version](https://badge.fury.io/py/scgraph.svg)](https://badge.fury.io/py/scgraph)
 [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
@@ -41,18 +41,33 @@ If you use SCGraph in your research, please cite:
 pip install scgraph
 ```
 
-A C++ extension is compiled automatically if a C++ compiler is available (~10x speedup on core algorithms). To verify: `from scgraph.utils import cpp_check; cpp_check()`.
+### Pure Python Installation (Skip Binary Builds)
 
-To skip the C++ build:
+If you want to install `scgraph` in pure Python mode without compiling or using binary extensions (or in environments without a C++ compiler):
 
-```bash
-# macOS / Linux / WSL2
-export SKBUILD_CMAKE_ARGS="-DSKIP_CPP_BUILD=ON" && pip install scgraph
-# Windows (PowerShell)
-$env:SKBUILD_CMAKE_ARGS="-DSKIP_CPP_BUILD=ON"; pip install scgraph
-# Windows (CMD)
-set SKBUILD_CMAKE_ARGS=-DSKIP_CPP_BUILD=ON && pip install scgraph
-```
+- **Linux / macOS / WSL2**:
+  ```bash
+  SCGRAPH_NO_BUILD=1 pip install --no-binary scgraph scgraph
+  ```
+  Or with `uv`:
+  ```bash
+  SCGRAPH_NO_BUILD=1 uv pip install --no-binary scgraph scgraph
+  ```
+
+- **Windows (PowerShell)**:
+  ```powershell
+  $env:SCGRAPH_NO_BUILD="1"
+  pip install --no-binary scgraph scgraph
+  ```
+
+- **Windows (CMD)**:
+  ```cmd
+  set SCGRAPH_NO_BUILD=1
+  pip install --no-binary scgraph scgraph
+  ```
+
+#### Automatic Fallback
+When building from source, if a C++ compiler is not available or compilation fails, `scgraph` will automatically fall back to a pure Python implementation.
 
 ---
 
@@ -173,9 +188,10 @@ All algorithms are available on `Graph` objects and accessible from `GeoGraph` v
 
 | `algorithm_fn` | Description | Time Complexity |
 |---|---|---|
-| `'dijkstra'` | Standard Dijkstra; general purpose, non-negative weights (default) | O((n+m) log n) |
+| `'dijkstra'` | Standard Dijkstra; general purpose, non-negative weights | O((n+m) log n) |
 | `'bidirectional_dijkstra'` | Bidirectional Dijkstra; simultaneous search from origin and destination | O((n+m) log n) |
 | `'dijkstra_buckets'` | Dijkstra with buckets (Dial's algorithm); efficient for non-negative weights (ideally >= 1) | O(n+m+W) |
+| `'bidirectional_buckets'` | Bidirectional Dijkstra (default for geographs) with buckets (Dial's algorithm); efficient for non-negative weights | O(n+m+W) |
 | `'dijkstra_negative'` | Dijkstra with cycle detection; supports negative weights | O(n·m) |
 | `'a_star'` | A* with optional heuristic; faster than Dijkstra with a good heuristic | O((n+m) log n) |
 | `'bellman_ford'` | Bellman-Ford; supports negative weights, slower than Dijkstra | O(n·m) |
@@ -188,7 +204,7 @@ All algorithms are available on `Graph` objects and accessible from `GeoGraph` v
 
 | Scenario | Recommended Approach |
 |---|---|
-| Single query | `dijkstra` (default) or `bidirectional_dijkstra` |
+| Single query | `dijkstra`, `bidirectional_buckets` or `bidirectional_dijkstra` |
 | Sparse networks (road/rail/maritime) | Reduce with `geograph.reduce()` for 2–4x faster queries |
 | Weights generally >= 1 | `dijkstra_buckets` |
 | Repeated queries from one origin | `cached_shortest_path` |
@@ -383,15 +399,10 @@ output = marnet_geograph.get_shortest_path(
 
 | `node_addition_type` | Description |
 |---|---|
-| `'kdclosest'` | Closest node via KD-tree (default, fastest) |
-| `'closest'` | Closest node via brute force |
-| `'quadrant'` | Closest node in each of 4 quadrants |
+| `'closest'` / `'kdclosest'` | Closest node via GeoKDTree (default) |
+| `'quadrant'` / `'kdquadrant'` | Closest node in each quadrant via GeoKDTree |
 | `'all'` | All nodes within the bound |
 
-| `node_addition_math` | Description |
-|---|---|
-| `'euclidean'` | Fast planar distance (default) |
-| `'haversine'` | Accurate great-circle distance |
 
 ---
 
@@ -432,7 +443,7 @@ The cache location defaults to the platform-appropriate directory:
 |---|---|
 | Linux | `~/.cache/scgraph/` |
 | macOS | `~/Library/Caches/scgraph/` |
-| Windows | `%LOCALAPPDATA%\\scgraph\\` |
+| Windows | `%LOCALAPPDATA%\scgraph\` |
 
 ### `load_geograph` Parameters
 
@@ -817,6 +828,7 @@ Originally inspired by [searoute](https://github.com/genthalili/searoute-py), in
 
 """
 
+
 from scgraph.graph_reducer import algorithm
 
 try:
@@ -829,4 +841,3 @@ except ImportError:
 
 from scgraph.geograph import GeoGraph
 from scgraph.grid import GridGraph
-

{scgraph-3.5.0 → scgraph-3.6.0b2}/scgraph/contraction_hierarchies.py

@@ -1,7 +1,11 @@
 import json
-from heapq import heappush, heappop
+from heapq import heappush, heappop, heapify
 from typing import Callable, Any, Optional, Union
-from scgraph.graph_utils import GraphUtils, GraphModifiers
+from scgraph.graph_utils import (
+    GraphUtils,
+    GraphModifiers,
+    normalize_node_input,
+)
 
 
 class CHGraphIO:
@@ -505,7 +509,10 @@ class CHGraphAlgorithms:
             return neighbors
 
     def search(
-        self, origin_id: int, destination_id: int, length_only: bool = False
+        self,
+        origin_id: int | set[int] | dict[int, int | float],
+        destination_id: int | set[int] | dict[int, int | float],
+        length_only: bool = False,
     ) -> dict[str, Any]:
         """
         Function:
@@ -515,11 +522,11 @@ class CHGraphAlgorithms:
         Required Arguments:
 
         - `origin_id`
-            - Type: int
-            - What: The id of the origin node
+            - Type: int | set[int] | dict[int, int | float]
+            - What: The id(s) of the origin node(s) (with optional starting distances)
         - `destination_id`
-            - Type: int
-            - What: The id of the destination node
+            - Type: int | set[int] | dict[int, int | float]
+            - What: The id(s) of the destination node(s) (with optional exit distances)
 
         Optional Arguments:
 
@@ -531,28 +538,39 @@ class CHGraphAlgorithms:
         Returns:
 
         - A dictionary with the following keys:
-            - `path`: A list of node ids representing the shortest path (omitted if length_only=True)
+            - `path`: A list of node ids representing the shortest path (empty if length_only=True)
             - `length`: The total length of the shortest path
         """
-        if origin_id == destination_id:
-            return (
-                {"length": 0}
-                if length_only
-                else {"path": [origin_id], "length": 0}
-            )
+        origin_dict = normalize_node_input(origin_id)
+        destination_dict = normalize_node_input(destination_id)
+
+        overlap = set(origin_dict.keys()) & set(destination_dict.keys())
+        best_dist = float("inf")
+        meeting_node = -1
+        if overlap:
+            for node in overlap:
+                dist = origin_dict[node] + destination_dict[node]
+                if dist < best_dist:
+                    best_dist = dist
+                    meeting_node = node
 
         # Forward search state
-        forward_distances = {origin_id: 0}
-        forward_parent = {origin_id: -1}
-        forward_open_leaves = [(0, origin_id)]
+        forward_distances = {node: dist for node, dist in origin_dict.items()}
+        forward_parent = {node: -1 for node in origin_dict}
+        forward_open_leaves = [
+            (dist, node) for node, dist in origin_dict.items()
+        ]
+        heapify(forward_open_leaves)
 
         # Backward search state
-        backward_distances = {destination_id: 0}
-        backward_parent = {destination_id: -1}
-        backward_open_leaves = [(0, destination_id)]
-
-        best_dist = float("inf")
-        meeting_node = -1
+        backward_distances = {
+            node: dist for node, dist in destination_dict.items()
+        }
+        backward_parent = {node: -1 for node in destination_dict}
+        backward_open_leaves = [
+            (dist, node) for node, dist in destination_dict.items()
+        ]
+        heapify(backward_open_leaves)
 
         while forward_open_leaves or backward_open_leaves:
             # Forward step
@@ -688,7 +706,7 @@ class CHGraphAlgorithms:
             raise Exception("No path found between origin and destination")
 
         if length_only:
-            return {"length": best_dist}
+            return {"path": [], "length": best_dist}
 
         path = self.__reconstruct_ch_path__(
             origin_id,
@@ -925,7 +943,10 @@ class CHGraph(
         return self.original_graph
 
     def get_shortest_path(
-        self, origin_id: int, destination_id: int, **kwargs: Any
+        self,
+        origin_id: int | set[int] | dict[int, int | float],
+        destination_id: int | set[int] | dict[int, int | float],
+        **kwargs: Any,
     ) -> dict[str, Any]:
         """
         Function:
@@ -935,11 +956,11 @@ class CHGraph(
         Required Arguments:
 
         - `origin_id`
-            - Type: int
-            - What: The id of the origin node
+            - Type: int | set[int] | dict[int, int | float]
+            - What: The origin node id(s), optionally with starting distances
         - `destination_id`
-            - Type: int
-            - What: The id of the destination node
+            - Type: int | set[int] | dict[int, int | float]
+            - What: The destination node id(s), optionally with exit distances
 
         Returns:
 
@@ -947,4 +968,4 @@ class CHGraph(
             - `path`: A list of node ids representing the shortest path
             - `length`: The total length of the shortest path
         """
-        return self.search(origin_id, destination_id)
+        return self.search(origin_id, destination_id, **kwargs)