Merge pull request #46 from GhanaNLP/feat-cleanup

Feat cleanup

.flake8

@@ -0,0 +1,26 @@
+[flake8]
+max-line-length = 120
+max-complexity = 10
+select = C,E,F,W,B,B950
+count = True
+statistics = True
+ignore =
+    E203,
+    E302,
+    E501,
+    W503,
+    D100,
+    D101,
+    D102,
+    D103,
+    D104,
+    D105,
+    D106,
+    D107
+exclude =
+    .git,
+    __pycache__,
+    *.egg-info,
+    .pytest_cache,
+    .mypy_cache,
+    notebooks/*

Makefile

@@ -0,0 +1,46 @@
+.DEFAULT_GOAL := help
+.PHONY: help lint test
+
+SRD_DIR = khaya
+TEST_DIR = tests
+
+ci: lint typecheck test ## Run all CI checks 
+.PHONY: ci
+
+lint:  ## Run linting
+	poetry run isort $(SRC_DIR) $(TEST_DIR)
+	poetry run flake8 --max-line-length 120 $(SRC_DIR) $(TEST_DIR)
+.PHONY: lint
+
+
+typecheck: ## Run type checking
+	poetry run mypy $(SRD_DIR)
+.PHONY: typecheck
+
+test: ## Run tests
+	poetry run coverage run --source=$(SRD_DIR) -m pytest -v $(TEST_DIR) && poetry run coverage report -m
+.PHONY: test
+
+clean-py: ## Remove python cache files
+	find . -name '__pycache__' -type d -exec rm -r {} +
+	find . -name '*.pyc' -type f -exec rm {} +
+	rm -rf ./*.egg-info
+.PHONY: clean-py
+
+.PHONY: help
+help: ## Show help message
+	@IFS=$$'\n' ; \
+	help_lines=(`fgrep -h "##" $(MAKEFILE_LIST) | fgrep -v fgrep | sed -e 's/\\$$//' | sed -e 's/##/:/'`); \
+	printf "%s\n\n" "Usage: make [task]"; \
+	printf "%-20s %s\n" "task" "help" ; \
+	printf "%-20s %s\n" "------" "----" ; \
+	for help_line in $${help_lines[@]}; do \
+		IFS=$$':' ; \
+		help_split=($$help_line) ; \
+		help_command=`echo $${help_split[0]} | sed -e 's/^ *//' -e 's/ *$$//'` ; \
+		help_info=`echo $${help_split[2]} | sed -e 's/^ *//' -e 's/ *$$//'` ; \
+		printf '\033[36m'; \
+		printf "%-20s %s" $$help_command ; \
+		printf '\033[0m'; \
+		printf "%s\n" $$help_info; \
+	done

khaya/__init__.py

@@ -1 +1,3 @@
 from .khaya_interface import KhayaInterface as khayaAPI
+
+__all__ = ["khayaAPI"]

khaya/asr_api.py

@@ -4,7 +4,7 @@
 
 
 class AsrApi(BaseApi):
-    def transcribe(self, audio_file_path: str, language="tw") -> Response:
+    def transcribe(self, audio_file_path: str, language="tw") -> Response | dict[str, str]:
         """
         Convert speech to text from audio binary data in an African language using the GhanaNLP STT API.
 

khaya/base_api.py

@@ -1,5 +1,6 @@
 from abc import ABC
 from typing import Optional
+
 import requests
 
 
@@ -13,7 +14,7 @@ def __init__(self, api_key: str, base_url: Optional[str] = None):
             "Cache-Control": "no-cache",
         }
 
-    def _make_request(self, method: str, url: str, **kwargs) -> requests.Response:
+    def _make_request(self, method: str, url: str, **kwargs) -> requests.Response | dict[str, str]:
         """
         Make an HTTP request.
 

khaya/khaya_interface.py

@@ -6,6 +6,8 @@
 from khaya.translation_api import TranslationApi
 from khaya.tts_api import TtsApi
 
+# custom type hint for Response or dict[str, str]
+ResponseOrDict = Response | dict[str, str]
 
 class KhayaInterface:
     """
@@ -58,7 +60,7 @@ def __init__(self, api_key: str, base_url: Optional[str] = "https://translation-
         self.asr_api = AsrApi(api_key, base_url)
         self.tts_api = TtsApi(api_key, base_url)
 
-    def translate(self, text: str, language_pair: str = "en-tw") -> Response:
+    def translate(self, text: str, language_pair: str = "en-tw") -> ResponseOrDict:
         """
         Translate text from one language to another.
 
@@ -72,7 +74,7 @@ def translate(self, text: str, language_pair: str = "en-tw") -> Response:
 
         return self.translation_api.translate(text, language_pair)
 
-    def asr(self, audio_file_path: str, language: str = "tw") -> Response:
+    def asr(self, audio_file_path: str, language: str = "tw") -> ResponseOrDict:
         """
         Get the transcription of an audio file from a given language.
 
@@ -85,7 +87,7 @@ def asr(self, audio_file_path: str, language: str = "tw") -> Response:
         """
         return self.asr_api.transcribe(audio_file_path, language)
 
-    def tts(self, text: str, lang: str) -> Response:
+    def tts(self, text: str, lang: str) -> ResponseOrDict:
         """
         Synthesize speech from text.
 

khaya/translation_api.py

@@ -4,7 +4,7 @@
 
 
 class TranslationApi(BaseApi):
-    def translate(self, text: str, language_pair: str = "en-tw") -> Response:
+    def translate(self, text: str, language_pair: str = "en-tw") -> Response | dict[str, str]:
         """
         Translate text from one language to another using the GhanaNLP translation API.
 

khaya/tts_api.py

@@ -1,11 +1,12 @@
 import json
+
 from requests.models import Response
 
 from khaya.base_api import BaseApi
 
 
 class TtsApi(BaseApi):
-    def synthesize(self, text: str, lang: str) -> Response:
+    def synthesize(self, text: str, lang: str) -> Response | dict[str, str]:
         """
         Convert text to speech in a specified African language using the GhanaNLP TTS API.