Add new files and modules

README.md

@@ -1,59 +1,32 @@
-# CMI-DAIR Template Python Repository
-
-This is a template repository. Below is a checklist of things you should do to use it:
-
-- [x] Run `setup_template.py` to set up the repository.
-- [ ] Rewrite this `README` file.
-- [ ] Install the `pre-commit` hooks.
-- [x] Update the `LICENSE` file to your desired license and set the year.
-- [ ] Update the supported versions in `SECURITY.md` or, if not relevant, delete this file.
-- [ ] Remove the placeholder src and test files, these are there merely to show how the CI works.
-- [ ] Grant third-party app permissions (e.g. Codecov) [here](https://github.com/organizations/cmi-dair/settings/installations), if necessary.
-- [ ] Either generate a `CODECOV_TOKEN` secret [here](https://github.com/cmi-dair/flowdump/blob/main/.github/workflows/python_tests.yaml) (if its a private repository) or remove the line `token: ${{ secrets.CODECOV_TOKEN }}`
-- [ ] API docs website: After the first successful build, go to the `Settings` tab of your repository, scroll down to the `GitHub Pages` section, and select `gh-pages` as the source. This will generate a link to your API docs.
-- [ ] Update stability badge in `README.md` to reflect the current state of the project. A list of stability badges to copy can be found [here](https://github.com/orangemug/stability-badges). The [node documentation](https://nodejs.org/docs/latest-v20.x/api/documentation.html#documentation_stability_index) can be used as a reference for the stability levels.
-
-# Project name
+# CLI-OAI
 
 [![Build](https://github.com/cmi-dair/cli-oai/actions/workflows/test.yaml/badge.svg?branch=main)](https://github.com/cmi-dair/cli-oai/actions/workflows/test.yaml?query=branch%3Amain)
 [![codecov](https://codecov.io/gh/cmi-dair/cli-oai/branch/main/graph/badge.svg?token=22HWWFWPW5)](https://codecov.io/gh/cmi-dair/cli-oai)
 [![Ruff](https://img.shields.io/endpoint?url=https://raw.githubusercontent.com/astral-sh/ruff/main/assets/badge/v2.json)](https://github.com/astral-sh/ruff)
 ![stability-wip](https://img.shields.io/badge/stability-work_in_progress-lightgrey.svg)
-[![MIT License](https://img.shields.io/badge/license-MIT-blue.svg)](https://github.com/cmi-dair/cli-oai/blob/main/LICENSE)
+[![MIT License](https://img.shields.io/badge/license-LGPL_2.1-blue.svg)](https://github.com/cmi-dair/cli-oai/blob/main/LICENSE)
 [![pages](https://img.shields.io/badge/api-docs-blue)](https://cmi-dair.github.io/cli-oai)
 
-What problem does this tool solve?
+CLI-OAI is a Python command-line interface for interacting with the OpenAI API.
 
-## Features
+## Usage
 
-- A few
-- Cool
-- Things
+Before running oai, make sure the environment variable OPENAI_API_KEY is set to
+your API key. To use the CLI, run `oai --help` in your terminal. See the `--help`
+function of the subcommands for more information on each command.
 
 ## Installation
 
-Install this package via :
-
-```sh
-pip install oai
-```
-
-Or get the newest development version via:
+Get the newest development version via:
 
 ```sh
-pip install git+https://github.com/cmi-dair/cli-oai
+poetry add git+https://github.com/cmi-dair/cli-oai
 ```
 
-## Quick start
-
-Short tutorial, maybe with a
+## Contributing
 
-```Python
-import oai
-
-oai.short_example()
-```
+Please see the [contributing guidelines](CONTRIBUTING.md) for guidelines on contributing to this project.
 
-## Links or References
+## License
 
-- [https://www.wikipedia.de](https://www.wikipedia.de)
+This project is licensed under the terms of the [L-GPLv2.1 license](LICENSE).

pyproject.toml

@@ -9,13 +9,20 @@ packages = [{include = "oai", from = "src"}]
 
 [tool.poetry.dependencies]
 python = "^3.11"
+pydantic-settings = "^2.1.0"
+pydantic = "^2.5.2"
+ffmpeg-python = "^0.2.0"
+openai = "^1.3.7"
+requests = "^2.31.0"
 
 [tool.poetry.group.dev.dependencies]
 pytest = "^7.4.3"
 mypy = "^1.7.1"
 pre-commit = "^3.5.0"
 pytest-cov = "^4.1.0"
 ruff = "^0.1.7"
+pytest-mock = "^3.12.0"
+pytest-asyncio = "^0.23.2"
 
 [tool.poetry.group.docs.dependencies]
 pdoc = "^14.1.0"
@@ -28,43 +35,48 @@ pythonpath = [
   "src"
 ]
 
+[tool.poetry.scripts]
+oai = 'oai.__main__:main'
+
 [tool.mypy]
 ignore_missing_imports = true
 
 [tool.ruff]
 exclude = [
-  ".bzr",
-  ".direnv",
-  ".eggs",
-  ".git",
-  ".git-rewrite",
-  ".hg",
-  ".mypy_cache",
-  ".nox",
-  ".pants.d",
-  ".pytype",
-  ".ruff_cache",
-  ".svn",
-  ".tox",
-  ".venv",
-  "__pypackages__",
-  "_build",
-  "buck-out",
-  "build",
-  "dist",
-  "node_modules",
-  "venv"
+    ".bzr",
+    ".direnv",
+    ".eggs",
+    ".git",
+    ".git-rewrite",
+    ".hg",
+    ".mypy_cache",
+    ".nox",
+    ".pants.d",
+    ".pytype",
+    ".ruff_cache",
+    ".svn",
+    ".tox",
+    ".venv",
+    "__pypackages__",
+    "_build",
+    "buck-out",
+    "build",
+    "dist",
+    "node_modules",
+    "venv",
 ]
 line-length = 88
 indent-width = 4
-src = ["src"]
+src=["src"]
+
 target-version = "py311"
 
 [tool.ruff.lint]
-select = ["ANN", "D", "E", "F", "I"]
+select = ["ALL"]
 ignore = [
-  "ANN101",  # self should not be annotated.
-  "ANN102"  # cls should not be annotated.
+  "ANN101", # Self should never be type annotated.
+  "ANN102", # cls should never be type annotated.
+  "B008", # Allow function call in arguments; this is common in FastAPI.
 ]
 fixable = ["ALL"]
 unfixable = []
@@ -80,8 +92,16 @@ line-ending = "auto"
 convention = "google"
 
 [tool.ruff.per-file-ignores]
-"tests/**/*.py" = []
-
-[build-system]
-requires = ["poetry-core>=1.2.0"]
-build-backend = "poetry.core.masonry.api"
+"tests/**/*.py" = [
+    "S101", # Allow assets
+    "ARG", # Unused arguments are common in tests (fixtures).
+    "FBT", #Allow positional booleans
+    "SLF001", # Allow private member access.
+    "INP001", # No need for namespace packages in tests.
+]
+"src/**/models.py" = [
+    "A003", # Allow id as a field name.
+]
+"src/**/schemas.py" = [
+    "A003", # Allow id as a field name.
+]

src/oai/__main__.py

@@ -0,0 +1,13 @@
+"""Entry point of the OAI package."""
+import asyncio
+
+from oai.cli import parser
+
+
+def main() -> None:
+    """Entry point for the CLI."""
+    asyncio.run(parser.parse_args())
+
+
+if __name__ == "__main__":
+    main()

src/oai/cli/__init__.py

@@ -0,0 +1 @@
+"""Modules for the Command Line Interface."""

src/oai/cli/commands.py

@@ -0,0 +1,101 @@
+"""Contains the core business logic of the OpenAI CLI."""
+import asyncio
+import logging
+import pathlib
+import tempfile
+
+import ffmpeg
+
+from oai import openai_api
+from oai.cli import utils
+from oai.core import config
+
+settings = config.get_settings()
+logger = logging.getLogger(settings.LOGGER_NAME)
+MAX_FILE_SIZE = 24_500_000  # Max size is 25MB, but we leave some room for error.
+
+
+async def speech_to_text(
+    filename: pathlib.Path,
+    model: str,
+    *,
+    clip: bool = False,
+) -> str:
+    """Transcribes audio files with OpenAI's TTS models.
+
+    Args:
+        filename: The file to transcribe. Can be any format that ffmpeg supports.
+        model: The transcription model to use.
+        voice: The voice to use.
+        clip: Whether to clip the file if it is too large, defaults to False.
+    """
+    with tempfile.TemporaryDirectory() as temp_dir:
+        temp_file = pathlib.Path(temp_dir) / "temp.mp3"
+        ffmpeg.input(filename).output(str(temp_file)).overwrite_output().run()
+
+        if clip:
+            files = utils.clip_audio(temp_file, temp_dir, MAX_FILE_SIZE)
+        else:
+            files = [temp_file]
+
+        stt = openai_api.SpeechToText()
+        transcription_promises = [stt.run(filename, model=model) for filename in files]
+        transcriptions = await asyncio.gather(*transcription_promises)
+
+        return " ".join(transcriptions)
+
+
+async def text_to_speech(
+    text: str,
+    output_file: str,
+    model: str,
+    voice: str,
+) -> bytes:
+    """Converts text to speech with OpenAI's TTS models.
+
+    Args:
+        text: The text to convert to speech.
+        output_file: The name of the output file.
+        model: The model to use.
+        voice: The voice to use.
+    """
+    tts = openai_api.TextToSpeech()
+    return await tts.run(text, output_file, model=model, voice=voice)
+
+
+async def image_generation(  # noqa: PLR0913
+    prompt: str,
+    output_base_name: str,
+    model: str,
+    width: int,
+    height: int,
+    quality: str,
+    n: int,
+) -> None:
+    """Generates an image from text with OpenAI's Image Generation models.
+
+    Args:
+        prompt: The text to generate an image from.
+        output_base_name: The base name of the output file.
+        model: The model to use.
+        width: The width of the generated image. Defaults to 1024.
+        height: The height of the generated image. Defaults to 1024.
+        quality: The quality of the generated image. Defaults to "standard".
+        n: The number of images to generate. Defaults to 1.
+
+    Returns:
+        bytes: The generated image as bytes.
+    """
+    image_generation = openai_api.ImageGeneration()
+    urls = await image_generation.run(
+        prompt,
+        model=model,
+        width=width,
+        height=height,
+        quality=quality,
+        n=n,
+    )
+
+    for index, url in enumerate(urls):
+        file = pathlib.Path(f"{output_base_name}_{index}.png")
+        utils.download_file(file, url)