using mypy for type checking

.github/workflows/test_pytest.yml

@@ -1,4 +1,4 @@
-name: 🧬
+name: 🌱💚🧬
 
 on:
   push:
@@ -19,5 +19,9 @@ jobs:
           python-version-file: "pyproject.toml"
       - name: Install the project
         run: uv sync --all-extras --dev
-      - name: Run tests
-        run: uv run pytest -s
+      - name: Run uv for formatting (via ruff), type checking (via mypy) and testing (via pytest)
+        run: |
+          uv run ruff format
+          uv run ruff check
+          uv run mypy */*.py
+          uv run pytest -s

.vscode/settings.json

@@ -0,0 +1,5 @@
+{
+    "cSpell.words": [
+        "ndarray"
+    ]
+}

README.md

@@ -14,8 +14,9 @@ cd gb/
 pip install uv
 pip install ruff
 uv init .
-uv add pytest numpy progressbar2 matplotlib polars scipy statsmodels  click
-ruff format
-ruff check
-pytest -s # or uv run pytest
+uv add pytest mypy numpy progressbar2 matplotlib polars scipy statsmodels  click
+uv run ruff format
+uv run ruff check
+uv run mypy */*.py
+uv run pytest -s
 ```

data/error.py

@@ -1,70 +1,82 @@
+"""
+Error classes
+"""
+
 from typing import Self
 
 
-class Error:
-	code: list[int]
-	message: list[str]
-
-	def __init__(self, code: int, message: str) -> None:
-		if isinstance(code, list):
-			self.code = code
-			self.message = message
-		else:
-			self.code = [code]
-			self.message = [message]
-
-	def __getitem__(self, index) -> int:
-		return self.code[index]
-
-	def __repr__(self) -> str:
-		n = len(self.code)
-		out = []
-		for i in range(n):
-			out.append(str(self.code[i]) + ': ' + self.message[i])
-		return '< ' + '|'.join(out) + ' > @ ' + str(hex(id(self)))
-
-	def __str__(self) -> str:
-		n = len(self.code)
-		out = []
-		out.append('')
-		out.append('@@@@@@@@@@@@@@@')
-		out.append('@@@ Error/s @@@')
-		out.append('@@@@@@@@@@@@@@@')
-		for i in range(n):
-			out.append(f'- {self.code[i]}:\t {self.message[i]}')
-		return '\n'.join(out)
-
-	def __eq__(self, error) -> bool:
-		return (self.code == error.code) and (self.message == error.message)
-
-	def chain(self, errors: Self) -> None:
-		codes = errors.code.copy()
-		messages = errors.message.copy()
-		codes.extend(self.code)
-		messages.extend(self.message)
-		self.code = codes
-		self.message = messages
+class Error(BaseException):
+    """
+    Parent error class
+    """
+
+    errors: list[Self]
+
+    def __init__(self: Self) -> None: ...
+    def __str__(self: Self) -> str:
+        if hasattr(self, 'errors'):
+            error_names = []
+            for e in self.errors:
+                error_names.append(str(type(e)).rsplit('.', maxsplit=-1)[-1].split("'")[0])
+            return ' | '.join(error_names)
+        else:
+            error_name = str(type(self)).rsplit('.', maxsplit=-1)[-1].split("'")[0]
+            return error_name
+
+    def chain(self: Self, other_error: Self) -> None:
+        """
+        Chain errors by putting them on a list, i.e. errors: list[Error]
+        """
+        if hasattr(self, 'errors') and hasattr(other_error, 'errors'):
+            self.errors.extend(other_error.errors)
+        elif hasattr(self, 'errors'):
+            self.errors.append(other_error)
+        elif hasattr(other_error, 'errors'):
+            self.errors = [self]
+            self.errors.append(other_error)
+        else:
+            self.errors = [self, other_error]
+
+
+class LogicError(Error):
+    """
+    Error in the logic of the code.
+    This means an unexpected error, i.e. an error we did not anticipate.
+    This is something we need to correctly classify or make a new error class for.
+    """
+
+
+class IncompatibleParameters(Error):
+    """
+    Input parameters are incompatible, e.g.
+    - the number of entries and number of loci should be non-zero and positive,
+    - two vectors/matrices/arrays which should be the same size are not
+    """
+
+
+class RandomSamplingError(Error):
+    """
+    Error in defining the parameters for the sampling distribution or
+    there are more items requested than there are items to choose from when sampling without replacement.
+    """
+
+
+class AlleleFreqOverUnderflow(Error):
+    """
+    Allele frequencies per locus are less than zero or higher than one.
+    """
 
 
 def test_error():
-	error1 = Error(101, 'Error test')
-	error2 = Error(102, 'Another error')
-	error3 = Error(103, 'Yet another error')
-	print(error1)
-	print(error2)
-	print(error3)
-
-	error1.chain(error2)
-	print(error1)
-	assert len(error1.code) == 2
-	error1.chain(error3)
-	assert len(error1.code) == 3
-	error1
-	print(error1)
-	type(error1)
-
-	assert error1 == Error(
-		[103, 102, 101], ['Yet another error', 'Another error', 'Error test']
-	)
-	assert error2 == Error(102, 'Another error')
-	assert error3 == Error(103, 'Yet another error')
+    error1 = Error()
+    error2 = LogicError()
+    error3 = IncompatibleParameters()
+    assert str(error1) == 'Error'
+    assert str(error2) == 'LogicError'
+    assert str(error3) == 'IncompatibleParameters'
+    error1.chain(error3)
+    assert str(error1) == 'Error | IncompatibleParameters'
+    error2.chain(error3)
+    assert str(error2) == 'LogicError | IncompatibleParameters'
+    error1.chain(error2)
+    assert str(error1) == 'Error | IncompatibleParameters | LogicError | IncompatibleParameters'

data/genotype.py

@@ -1,39 +1,39 @@
-from data.type_aliases import VectorU64, VectorF64, VectorSTR, MatrixF64, MatrixB
+import numpy as np
 # from data.error import Error
 
 
 class Genomes:
-	n: int
-	p: int
-	m: int
-	a: int
-	b: VectorF64
-	entries: VectorSTR
-	chromosomes: VectorSTR
-	positions: VectorU64
-	alleles: VectorSTR
-	X: MatrixF64
-	mask: MatrixB
+    n: int
+    p: int
+    m: int
+    a: int
+    b: np.ndarray
+    entries: np.ndarray
+    chromosomes: np.ndarray
+    positions: np.ndarray
+    alleles: np.ndarray
+    X: np.ndarray
+    mask: np.ndarray
 
-	def __init__(self) -> None:
-		self.n = 1
-		self.p = 1
-		self.m = 1
-		self.a = 1
-		self.b = [1.00]
-		self.entries = ['dummy_entry_1']
-		self.chromosomes = ['chromosome_1']
-		self.positions = [123_456_789]
-		self.alleles = ['allele 1|allele_2|allele_3\tallele_2']
-		self.X = [[1.00]]
-		self.mask = [[False]]
+    def __init__(self) -> None:
+        self.n = 1
+        self.p = 1
+        self.m = 1
+        self.a = 1
+        self.b = np.array([1.00])
+        self.entries = np.array(['dummy_entry_1'])
+        self.chromosomes = np.array(['chromosome_1'])
+        self.positions = np.array([123_456_789])
+        self.alleles = np.array(['allele 1|allele_2|allele_3\tallele_2'])
+        self.X = np.array([[1.00]])
+        self.mask = np.array([[False]])
 
 
 def test_genotype():
-	genotypes = Genomes()
-	print(genotypes.X)
-	assert genotypes.n == 1
-	assert genotypes.n == genotypes.p
-	assert genotypes.p == genotypes.m
-	assert genotypes.m == genotypes.a
-	assert len(genotypes.b) == 1
+    genotypes = Genomes()
+    print(genotypes.X)
+    assert genotypes.n == 1
+    assert genotypes.n == genotypes.p
+    assert genotypes.p == genotypes.m
+    assert genotypes.m == genotypes.a
+    assert len(genotypes.b) == 1

data/phenotype.py

@@ -1,25 +1,25 @@
-from data.type_aliases import VectorSTR, MatrixF64, MatrixB
+import numpy as np
 # from data.error import Error
 
 
 class Phenomes:
-	n: int
-	q: int
-	entries: VectorSTR
-	traits: VectorSTR
-	Y: MatrixF64
-	mask: MatrixB
+    n: int
+    q: int
+    entries: np.ndarray
+    traits: np.ndarray
+    Y: np.ndarray
+    mask: np.ndarray
 
-	def __init__(self) -> None:
-		self.n = 1
-		self.q = 1
-		self.entries = ['dummy_entry_1']
-		self.traits = ['dummy_trait_1']
-		self.Y = [[0.00]]
-		self.mask = [[False]]
+    def __init__(self) -> None:
+        self.n = 1
+        self.q = 1
+        self.entries = np.array(['dummy_entry_1'])
+        self.traits = np.array(['dummy_trait_1'])
+        self.Y = np.array([[0.00]])
+        self.mask = np.array([[False]])
 
 
 def test_phenomes():
-	Y = Phenomes()
-	assert Y.n == 1
-	assert Y.n == Y.q
+    Y = Phenomes()
+    assert Y.n == 1
+    assert Y.n == Y.q