Fix CI: install dependencies via Poetry instead of bare pip

[Copilot]

The workflow was failing with ModuleNotFoundError: No module named 'requests' during flake8 because only flake8 and pytest were installed via pip, bypassing Poetry and leaving runtime dependencies (requests) absent.

Changes

• Install Poetry in CI before the dependency step
• Replace pip install flake8 pytest with poetry install --with dev, which installs all runtime and dev dependencies from pyproject.toml
• Prefix flake8 and pytest invocations with poetry run to execute inside the Poetry virtual environment
• Remove requirements.txt fallback — redundant given pyproject.toml is the source of truth

- name: Install Poetry
  run: pip install poetry
- name: Install dependencies
  run: poetry install --with dev
- name: Lint with flake8
  run: |
    poetry run flake8 . --count --select=E9,F63,F7,F82 --show-source --statistics
    poetry run flake8 . --count --exit-zero --max-complexity=10 --max-line-length=127 --statistics
- name: Test with pytest
  run: |
    poetry run pytest

Original prompt

Create a fix for the failing GitHub Actions workflow in AKB0700/deepl-python.

Context:

• Failing job URL: https://github.com/AKB0700/deepl-python/actions/runs/22592189871/job/65458803978
• The job fails during the flake8 step with:
  • ModuleNotFoundError: No module named 'requests'
  • originating from deepl/http_client.py importing requests.

Repository findings:

• The repository uses Poetry, as indicated by pyproject.toml with [tool.poetry] and build backend poetry.core.masonry.api.
• pyproject.toml declares runtime dependency requests = ">=2.32.5" under [tool.poetry.dependencies], plus dev dependencies including flake8 and pytest under [tool.poetry.group.dev.dependencies].
• Current workflow .github/workflows/python-package.yml installs only flake8 and pytest via pip and optionally installs requirements.txt if present; it does not install Poetry dependencies from pyproject.toml, so requests is missing.

Task:

• Update .github/workflows/python-package.yml to correctly install project dependencies using Poetry.
• Ensure lint (flake8) and tests (pytest) run within the Poetry environment.
• Keep the existing matrix for Python versions (3.9, 3.10, 3.11) unless a change is required.
• Prefer minimal, standard changes suitable for CI.

Acceptance criteria:

• Workflow installs dependencies from pyproject.toml (including requests).
• flake8 and pytest run successfully in CI.
• No dependency on requirements.txt is required.

Files involved:

• pyproject.toml
• .github/workflows/python-package.yml

This pull request was created from Copilot chat.

---

💬 We'd love your input! Share your thoughts on Copilot coding agent in our 2 minute survey.

[Copilot]

Pull request overview

Updates the GitHub Actions “Python package” workflow to use Poetry for dependency installation so lint/test steps run with the full set of project (runtime + dev) dependencies defined in pyproject.toml, addressing the CI failure caused by missing runtime deps during flake8.

Changes:

• Install Poetry in the workflow before dependency installation.
• Replace direct pip install flake8 pytest (and requirements.txt fallback) with poetry install --with dev.
• Run flake8 and pytest via poetry run to ensure they execute inside the Poetry environment.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

CI installs Poetry with an unpinned pip install poetry. This makes the workflow non-reproducible and can break the matrix unexpectedly when Poetry releases a version incompatible with one of the Python versions. Elsewhere in this repo CI pins Poetry (e.g., .github/workflows/run_ci.yml installs poetry==2.1.2) and upgrades pip first; consider doing the same here and using python -m pip ... to ensure Poetry is installed into the matrix Python environment.

Suggested change

	run: pip install poetry
	run: |
	python -m pip install --upgrade pip
	python -m pip install "poetry==2.1.2"