code refactor

README.md

@@ -55,6 +55,7 @@ print(translated_text)  # Output: Hello everyone
 ---
 
 ### Using `AsyncTranslator` (Asynchronous)
+
 ```python
 import asyncio
 from llmtranslate import AsyncTranslator
@@ -63,41 +64,43 @@ from langchain_openai import ChatOpenAI
 # Initialize the LLM and AsyncTranslator
 llm = ChatOpenAI(model_name="gpt-4o", openai_api_key="your_openai_api_key")
 
+
 async def translate_text():
     translator = AsyncTranslator(
-        llm=llm, 
-        max_length_text_chunk_to_translate=100, 
-        max_length_text_chunk_to_translate_multiple_languages=50, 
+        llm=llm,
+        max_translation_chunk_length=100,
+        max_translation_chunk_length_multilang=50,
         max_concurrent_llm_calls=10
     )
     tasks = [
-        translator.get_text_language("Hi how are you?"), 
+        translator.get_text_language("Hi how are you?"),
         translator.translate("Hi how are you?", "Spanish")
     ]
     results = await asyncio.gather(*tasks)
     # Output the detected language information
     text_language = results[0]
     if results:
-        print(text_language.ISO_639_1_code) # Output: en
-        print(text_language.ISO_639_2_code) # Output: eng
-        print(text_language.ISO_639_3_code) # Output: eng
-        print(text_language.language_name) # Output: English
-    
+        print(text_language.ISO_639_1_code)  # Output: en
+        print(text_language.ISO_639_2_code)  # Output: eng
+        print(text_language.ISO_639_3_code)  # Output: eng
+        print(text_language.language_name)  # Output: English
+
     # Output the translated text
     print(results[1])  # Output: Hola, ¿cómo estás?
 
+
 # Run the asynchronous translation
 asyncio.run(translate_text())
 ```
 
 ---
 ## Key Parameters
 
-### `max_length_text_chunk_to_translate`
+### `max_translation_chunk_length`
 - **Description**: Defines the maximum length (in characters) of a text chunk to be translated in a single call when the text is in one language.
 - **Recommendation**: If translations are not accurate, try reducing this value as weaker LLMs struggle with large chunks of text.
 
-### `max_length_text_chunk_to_translate_multiple_languages`
+### `max_translation_chunk_length_multilang`
 - **Description**: Defines the maximum length (in characters) of a text chunk when the text contains multiple languages.
 - **Recommendation**: Reduce this value for better accuracy with multi-language inputs.
 

llmtranslate/translator.py

@@ -94,19 +94,19 @@ class HowManyLanguages(BaseModel):
 
 
 
-    def __init__(self, llm: BaseChatModel, max_length_text_chunk_to_translate: int = 200, max_length_text_chunk_to_translate_multiple_languages: int = 50):
+    def __init__(self, llm: BaseChatModel, max_translation_chunk_length: int = 200, max_translation_chunk_length_multilang: int = 50):
         """
         Initializes the Translator with the given parameters.
 
         Args:
             llm (BaseChatModel):
                 The large language model used for performing language detection
                 and translation tasks.
-            max_length_text_chunk_to_translate (int, optional):
+            max_translation_chunk_length (int, optional):
                 The maximum length of text to be translated in a single chunk
                 when dealing with a single language. If not provided,
                 a default value is used.
-            max_length_text_chunk_to_translate_multiple_languages (int, optional):
+            max_translation_chunk_length_multilang (int, optional):
                 The maximum length of text to be translated in a single chunk
                 when multiple languages are present. If not provided,
                 a different default value is used.
@@ -115,8 +115,8 @@ def __init__(self, llm: BaseChatModel, max_length_text_chunk_to_translate: int =
             raise MissingLangchainChatModelError()
 
         self.llm = llm
-        self.max_length_text_chunk_to_translate = max_length_text_chunk_to_translate  if max_length_text_chunk_to_translate else MAX_LENGTH
-        self.max_length_text_chunk_to_translate_multiple_languages = max_length_text_chunk_to_translate_multiple_languages if max_length_text_chunk_to_translate_multiple_languages else MAX_LENGTH_MINI_TEXT_CHUNK
+        self.max_translation_chunk_length = max_translation_chunk_length  if max_translation_chunk_length else MAX_LENGTH
+        self.max_translation_chunk_length_multilang = max_translation_chunk_length_multilang if max_translation_chunk_length_multilang else MAX_LENGTH_MINI_TEXT_CHUNK
         ######## detect language ##########
         structured_llm_detect_language = llm.with_structured_output(BaseTranslator.TextLanguageFormat)
         self.few_shot_structured_llm_detect_language = BaseTranslator.prompt_detect_language | structured_llm_detect_language
@@ -156,7 +156,7 @@ def get_text_language(self, text: str) -> TextLanguage:
         >>> print(language.language_name)
         French
         """
-        text = get_first_n_words(text, self.max_length_text_chunk_to_translate)
+        text = get_first_n_words(text, self.max_translation_chunk_length)
         response =  self.few_shot_structured_llm_detect_language.invoke(text)
         response_message = response.language_ISO_639_1_code
         try:
@@ -189,13 +189,13 @@ def translate(self, text: str, to_language ="Spanish") -> str:
             >>> print(result)
             "Hola mundo. Hola el mundo."
         """
-        text_chunks = split_text_to_chunks(text, self.max_length_text_chunk_to_translate)
+        text_chunks = split_text_to_chunks(text, self.max_translation_chunk_length)
         counted_number_of_languages =  [self.how_many_languages_are_in_text(text_chunk) for text_chunk in text_chunks]
 
         translated_list = []
         for index, text_chunk in enumerate(text_chunks):
             if counted_number_of_languages[index] > 1:
-                mini_text_chunks = split_text_to_chunks(text_chunk, self.max_length_text_chunk_to_translate_multiple_languages)
+                mini_text_chunks = split_text_to_chunks(text_chunk, self.max_translation_chunk_length_multilang)
                 for mini_text_chunk in mini_text_chunks:
                     translated_list.append(self.translate_chunk_of_text(mini_text_chunk, to_language))
             else:
@@ -221,11 +221,11 @@ class AsyncTranslator(BaseTranslator):
     def __init__(
             self,
             llm: BaseChatModel,
-            max_length_text_chunk_to_translate: int = 200,
-            max_length_text_chunk_to_translate_multiple_languages: int = 50,
+            max_translation_chunk_length: int = 200,
+            max_translation_chunk_length_multilang: int = 50,
             max_concurrent_llm_calls: int = 100
     ):
-        super().__init__(llm, max_length_text_chunk_to_translate, max_length_text_chunk_to_translate_multiple_languages)
+        super().__init__(llm, max_translation_chunk_length, max_translation_chunk_length_multilang)
         self.semaphore = asyncio.Semaphore(max_concurrent_llm_calls)
 
 
@@ -256,9 +256,9 @@ async def get_text_language(self, text) -> TextLanguage:
         >>> print(language.language_name)
         French
         """
-        text = get_first_n_words(text, self.max_length_text_chunk_to_translate)
+        text = get_first_n_words(text, self.max_translation_chunk_length)
         async with self.semaphore:
-            text = get_first_n_words(text, self.max_length_text_chunk_to_translate)
+            text = get_first_n_words(text, self.max_translation_chunk_length)
             response = await self.few_shot_structured_llm_detect_language.ainvoke(text)
         response_message = response.language_ISO_639_1_code
         try:
@@ -297,7 +297,7 @@ async def translate(self, text: str, to_language ="Spanish") -> str:
             >>> print(result)
             "Hola mundo. Hola el mundo."
         """
-        text_chunks = split_text_to_chunks(text, self.max_length_text_chunk_to_translate)
+        text_chunks = split_text_to_chunks(text, self.max_translation_chunk_length)
 
         # Run how_many_languages_are_in_text concurrently
         # Chunks that contain more than one language will be split (this will simplify translation for the LLM)
@@ -306,7 +306,7 @@ async def translate(self, text: str, to_language ="Spanish") -> str:
         tasks = []
         for index, text_chunk in enumerate(text_chunks):
             if counted_number_of_languages[index] > 1:
-                mini_text_chunks = split_text_to_chunks(text_chunk, self.max_length_text_chunk_to_translate_multiple_languages)
+                mini_text_chunks = split_text_to_chunks(text_chunk, self.max_translation_chunk_length_multilang)
                 for mini_text_chunk in mini_text_chunks:
                     tasks.append(self.translate_chunk_of_text(mini_text_chunk, to_language))
             else:

my_notes/my_tests/new_langchain_test.py

@@ -11,7 +11,7 @@
 
 llm = ChatOpenAI(model="gpt-4o-mini")
 
-translator = AsyncTranslator(llm=llm, max_length_text_chunk_to_translate=100, max_length_text_chunk_to_translate_multiple_languages=100)
+translator = AsyncTranslator(llm=llm, max_translation_chunk_length=100, max_translation_chunk_length_multilang=100)
 
 
 async def run_test():

my_notes/my_tests/testt_new_translator.py

@@ -3,7 +3,7 @@
 from langchain_openai import ChatOpenAI
 
 llm = ChatOpenAI(model_name="gpt-4o", openai_api_key=os.getenv("OPENAI_API_KEY"))
-translator = Translator(llm=llm, max_length_text_chunk_to_translate=100, max_length_text_chunk_to_translate_multiple_languages=100)
+translator = Translator(llm=llm, max_translation_chunk_length=100, max_translation_chunk_length_multilang=100)
 text_language = translator.get_text_language("Hi how are you?")
 if text_language:
     print(text_language.ISO_639_1_code)

my_notes/my_tests/testt_new_translator_async.py

@@ -6,7 +6,7 @@
 
 llm = ChatOpenAI(model_name="gpt-4o", openai_api_key=os.getenv("OPENAI_API_KEY"))
 async def translate_text():
-    translator = AsyncTranslator(llm=llm, max_length_text_chunk_to_translate=100, max_length_text_chunk_to_translate_multiple_languages=50, max_concurrent_llm_calls=10)
+    translator = AsyncTranslator(llm=llm, max_translation_chunk_length=100, max_translation_chunk_length_multilang=50, max_concurrent_llm_calls=10)
     tasks = [translator.get_text_language("Hi how are you?"), translator.translate("Hi how are you?", "Spanish")]
     results = await asyncio.gather(*tasks)
     print(results[0])

tests/test_translator.py

@@ -16,7 +16,7 @@ class TestTranslator:
     @pytest.fixture
     def translator(self):
         llm = ChatOpenAI(model_name="gpt-4o-mini", openai_api_key=os.getenv("OPENAI_API_KEY"))
-        return Translator(llm=llm, max_length_text_chunk_to_translate=100, max_length_text_chunk_to_translate_multiple_languages=100)
+        return Translator(llm=llm, max_translation_chunk_length=100, max_translation_chunk_length_multilang=100)
 
     def test_set_api_key_success(self, translator):
         assert translator.llm is not None
@@ -116,7 +116,7 @@ class TestAsyncTranslator:
     @pytest.fixture
     def translator(self):
         llm = ChatOpenAI(model_name="gpt-4o-mini", openai_api_key=os.getenv("OPENAI_API_KEY"))
-        return AsyncTranslator(llm=llm, max_length_text_chunk_to_translate=100, max_length_text_chunk_to_translate_multiple_languages=100, max_concurrent_llm_calls=10)
+        return AsyncTranslator(llm=llm, max_translation_chunk_length=100, max_translation_chunk_length_multilang=100, max_concurrent_llm_calls=10)
 
     def test_set_api_key_success(self, translator):
         assert translator.llm is not None