Move to pyproject.toml, python=3.11, Ruff pre-commit

.gitattributes

@@ -1,3 +1,3 @@
-*.t linguist-detectable=false
+*.ctl linguist-detectable=false
 *.html linguist-detectable=false
-
+*.t linguist-detectable=false

.github/workflows/test.yml

@@ -7,7 +7,7 @@ jobs:
     strategy:
       matrix:
         os: [ubuntu-latest, macos-latest]
-        python-version: [3.9, 3.11]
+        python-version: [3.11]
     name: Python ${{ matrix.python-version }} (${{ matrix.os }})
     steps:
       - uses: actions/checkout@v2
@@ -19,10 +19,8 @@ jobs:
           conda-channels: bioconda
       - name: Install
         run: |
-          conda install paml datrie
-          pip install --upgrade pip setuptools wheel
-          pip install pytest
-          pip install .
+          conda env create -f environment.yml
+          pip install '.[dev]'
       - name: Test
         run: |
           pytest -s --slow

.pre-commit-config.yaml

@@ -1,11 +1,10 @@
 repos:
--   repo: https://github.com/pre-commit/pre-commit-hooks
-    rev: v4.4.0
-    hooks:
-    -   id: check-yaml
-    -   id: end-of-file-fixer
-    -   id: trailing-whitespace
--   repo: https://github.com/psf/black
-    rev: 23.3.0
-    hooks:
-    -   id: black
+- repo: https://github.com/astral-sh/ruff-pre-commit
+  # Ruff version.
+  rev: v0.1.5
+  hooks:
+    # Run the linter.
+    - id: ruff
+      args: [ --fix ]
+    # Run the formatter.
+    - id: ruff-format

README.md

@@ -28,7 +28,7 @@ If the conda package manager is already installed, skip this step, otherwise:
 
 An `x86_64` Miniconda installation is required in order to install Vespasian.
 
-- If your Mac has an Intel processor, skip this step. Otherwise:
+- If using a Mac with an Intel processor, skip this step. Otherwise:
 
   ```bash
   arch -x86_64 zsh
@@ -45,7 +45,7 @@ An `x86_64` Miniconda installation is required in order to install Vespasian.
 
 ### Installing Vespasian
 
-- If using a Mac with an Intel processor, skip this step. Otherwise, run:
+- If using a Mac has an Intel processor, skip this step. Otherwise:
 
   ```bash
   arch -x86_64 zsh
@@ -68,10 +68,10 @@ An `x86_64` Miniconda installation is required in order to install Vespasian.
 ### Development install
 
 ```bash
-conda create -y -n vespasian-dev python=3.11 paml==4.10.6 pytest black pre-commit -c conda-forge -c bioconda
+conda create -y -n vespasian-dev python=3.11 paml==4.10.6 -c conda-forge -c bioconda
 conda activate vespasian-dev
 git clone https://github.com/bede/vespasian
-pip install --editable ./vespasian
+pip install --editable './vespasian[dev]'
 ```
 
 

pyproject.toml

@@ -0,0 +1,38 @@
+[build-system]
+requires = ["flit_core >=3.2,<4"]
+build-backend = "flit_core.buildapi"
+
+[project]
+name = "vespasian"
+authors = [{name = "Bede Constantinides", email = "bedeabc@gmail.com"}]
+license = {file = "LICENSE"}
+readme = "README.md"
+dynamic = ["version", "description"]
+requires-python = ">=3.11"
+dependencies = [
+  "biopython>=1.78",
+  "treeswift==1.1.14",
+  "tqdm",
+  "argh",
+  "pyyaml",
+  "parmap",
+  "snakemake",
+  "pandas>=1.2.4",
+  "numpy>=1.20.2",
+  "scipy>=1.6.2"
+]
+
+[project.scripts]
+vespasian = "vespasian.cli:main"
+
+[project.optional-dependencies]
+dev = [
+    "pytest>=7.4.3",
+    "pre-commit>=3.5.0",
+    "ruff>=0.1.5",
+    "flit>=3.9.0"
+]
+
+[tool.ruff]
+# Ignore forward annotation syntax errors and ambiguous variable name
+ignore = ["F722", "E741"]

vespasian/__init__.py

@@ -1 +1,2 @@
+"""Genome scale evolutionary hypothesis testing"""
 __version__ = "0.5.3"

vespasian/cli.py

@@ -3,7 +3,6 @@
 import warnings
 
 from math import ceil
-from pprint import pprint
 
 import argh
 

vespasian/util.py

@@ -1,16 +1,17 @@
-import os
 import shutil
 
-import tqdm
 
-from Bio import AlignIO, SeqIO
+from Bio import SeqIO
 
 from vespasian import vespasian
 
 
 def convert_phylip(fasta_path, phylip_path):
     """Convert fasta to sequential phylip format"""
-    chunkify_string = lambda x, n: [x[i : i + n] for i in range(0, len(x), n)]
+
+    def chunkify_string(string, chunksize):
+        return [string[i : i + chunksize] for i in range(0, len(string), chunksize)]
+
     records = list(SeqIO.parse(fasta_path, "fasta"))
     indent_len = max(len(r.id) for r in records) + 2
     indent_fmt = "\n" + " " * indent_len
@@ -27,7 +28,7 @@ def convert_phylip(fasta_path, phylip_path):
 
 def reformat_environments(codeml_root):
     """Reformat vespasian codeml environments for use with legacy codeml_reader"""
-    codeml_envs = vespa.list_codeml_dirs(codeml_root)
+    codeml_envs = vespasian.list_codeml_dirs(codeml_root)
     for env in codeml_envs:
         convert_phylip(f"{env}/align.fa", f"{env}/align.phy")
         shutil.copy(f"{env}/tree.nwk", f"{env}/tree")

vespasian/vespasian.py

@@ -14,11 +14,10 @@
 import parmap
 import treeswift
 
-import numpy as np
 import pandas as pd
 
 from Bio import AlignIO, SeqIO
-from scipy.stats import chi2, chisquare
+from scipy.stats import chi2
 
 from vespasian import util
 
@@ -645,7 +644,6 @@ def filter_results(results):
                 r["model"],
             )  # Tuple-keyed dict of family-tree-model
             # print(key)
-            # pprint(r)
             if r["lnl"] > names_lnls[key]:
                 names_lnls[key] = r["lnl"]
                 names_records[key] = r