add tests for sym

pyproject.toml

@@ -12,12 +12,20 @@ dependencies = [
     "numpy>=1.26.4",
     "scipy>=1.13.0",
     "beartype>=0.18.2",
+    "pytest>=8.1.1",
 ]
 readme = "README.md"
 requires-python = ">= 3.10"
 
-[project.scripts]
-hello = "cytnx_torch:hello"
+[tool.rye.scripts]
+clean = "git clean -f"
+"coverage:run" = "coverage run -m pytest test"
+"coverage:xml" = "coverage xml"
+"coverage:html" = "coverage html"
+"coverage:report" = "coverage report"
+"coverage:open" = "open htmlcov/index.html"
+"coverage:github" = { chain = ["coverage:run", "coverage:xml", "coverage:report"]}
+"test:all" = "pytest test"
 
 [build-system]
 requires = ["hatchling"]
@@ -43,10 +51,6 @@ dev-dependencies = [
 [tool.hatch.metadata]
 allow-direct-references = true
 
-[tool.hatch.build.targets.wheel]
-packages = ["src/cytnx_torch"]
-
-
 [tool.black]
 line-length = 88
 

requirements.lock

@@ -11,6 +11,7 @@
 beartype==0.18.2
 filelock==3.13.4
 fsspec==2024.3.1
+iniconfig==2.0.0
 jinja2==3.1.3
 markupsafe==2.1.5
 mpmath==1.3.0
@@ -28,7 +29,10 @@ nvidia-cusparse-cu12==12.1.0.106
 nvidia-nccl-cu12==2.19.3
 nvidia-nvjitlink-cu12==12.4.127
 nvidia-nvtx-cu12==12.1.105
+packaging==24.0
 pillow==10.3.0
+pluggy==1.4.0
+pytest==8.1.1
 scipy==1.13.0
 sympy==1.12
 torch==2.2.2

src/cytnx_torch/symmetry.py

@@ -0,0 +1,77 @@
+import numpy as np
+from dataclasses import dataclass, field
+from beartype.typing import List
+from abc import abstractmethod
+
+
+@dataclass(frozen=True)
+class Symmetry:
+    label: str = field(default="")
+
+
+# trait
+@dataclass(frozen=True)
+class AbelianSym(Symmetry):
+
+    @abstractmethod
+    def combine_rule(self, A: int, B: int) -> int:
+        raise NotImplementedError("not implement for abstract type trait.")
+
+    @abstractmethod
+    def check_qnums(self, qnums: List[int]) -> bool:
+        raise NotImplementedError("not implement for abstract type trait.")
+
+    def combine_qnums(self, qnums_a: List[int], qnums_b: List[int]) -> List[int]:
+        mesh_b, mesh_a = np.meshgrid(qnums_b, qnums_a)
+        return self.combine_rule(mesh_a.flatten(), mesh_b.flatten())
+
+
+@dataclass(frozen=True)
+class U1(AbelianSym):
+    """
+    Unitary Symmetry class.
+    The U1 symmetry can have quantum number represent as arbitrary unsigned integer.
+
+    Fusion rule for combine two quantum number:
+
+        q1 + q2
+
+    """
+
+    def combine_rule(self, A: int, B: int) -> int:
+        return A + B
+
+    def __str__(self) -> str:
+        return f"U1 label={self.label}"
+
+    def check_qnums(self, qnums: List[int]) -> bool:
+        return True
+
+
+@dataclass(frozen=True)
+class Zn(AbelianSym):
+    """
+    Z(n) Symmetry class.
+    The Z(n) symmetry can have integer quantum number, with n > 1.
+
+    Fusion rule for combine two quantum number:
+
+        (q1 + q2)%n
+    """
+
+    n: int = field(default=2)
+
+    def __post_init__(self):
+        if self.n < 2:
+            raise ValueError(
+                "Symmetry.Zn", "[ERROR] discrete symmetry Zn must have n >= 2."
+            )
+
+    def combine_rule(self, A: int, B: int) -> int:
+        return (A + B) % self.n
+
+    def __str__(self):
+        return f"Z{self.n} label={self.label}"
+
+    def check_qnums(self, qnums: List[int]) -> bool:
+        return np.all([(q >= 0) and (q < self.n) for q in qnums])

test/test_sym.py

@@ -0,0 +1,28 @@
+import numpy as np
+from cytnx_torch.symmetry import U1, Zn
+
+
+def test_U1():
+
+    s1 = U1(label="a")
+
+    q1 = [0, 1, 2]
+    q2 = [0, -1, 3]
+
+    assert s1.check_qnums(q1)
+    assert s1.check_qnums(q2)
+
+    assert np.all(s1.combine_qnums(q1, q2) == [0, -1, 3, 1, 0, 4, 2, 1, 5])
+
+
+def test_Zn():
+
+    s1 = Zn(label="x", n=3)
+
+    q1 = [0, 1, 2]
+    q2 = [2, 0]
+
+    assert s1.check_qnums(q1)
+    assert s1.check_qnums(q2)
+
+    assert np.all(s1.combine_qnums(q1, q2) == [2, 0, 0, 1, 1, 2])