hamtaa-texttools 1.1.8__py3-none-any.whl → 1.1.9__py3-none-any.whl

{hamtaa_texttools-1.1.8.dist-info → hamtaa_texttools-1.1.9.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: hamtaa-texttools
-Version: 1.1.8
+Version: 1.1.9
 Summary: A high-level NLP toolkit built on top of modern LLMs.
 Author-email: Tohidi <the.mohammad.tohidi@gmail.com>, Montazer <montazerh82@gmail.com>, Givechi <mohamad.m.givechi@gmail.com>, MoosaviNejad <erfanmoosavi84@gmail.com>
 License: MIT License
@@ -187,6 +187,26 @@ Use **TextTools** when you need to:
 
 ---
 
+## 🔍 Logging
+
+TextTools uses Python's standard `logging` module. The library's default logger level is `WARNING`, so if you want to modify it, follow instructions:
+
+
+```python
+import logging
+
+# Default: warnings and errors only
+logging.basicConfig(level=logging.WARNING)
+
+# Debug everything (verbose)
+logging.basicConfig(level=logging.DEBUG)
+
+# Complete silence
+logging.basicConfig(level=logging.CRITICAL)
+```
+
+---
+
 ## 📚 Batch Processing
 
 Process large datasets efficiently using OpenAI's batch API.

{hamtaa_texttools-1.1.8.dist-info → hamtaa_texttools-1.1.9.dist-info}/RECORD

@@ -1,4 +1,4 @@
-hamtaa_texttools-1.1.8.dist-info/licenses/LICENSE,sha256=Hb2YOBKy2MJQLnyLrX37B4ZVuac8eaIcE71SvVIMOLg,1082
+hamtaa_texttools-1.1.9.dist-info/licenses/LICENSE,sha256=Hb2YOBKy2MJQLnyLrX37B4ZVuac8eaIcE71SvVIMOLg,1082
 texttools/__init__.py,sha256=lFYe1jdssHC1h8qcPpV3whANxiDi8aiiFdY-7L0Ck10,164
 texttools/batch/__init__.py,sha256=DJGJTfR6F3Yv4_alsj9g1tesGzdcSV27Zw74DonhW_s,102
 texttools/batch/batch_manager.py,sha256=ZgLiO9maCHnx2cJbUjsYXFnlUsMLI2TP3Vc9uKU0BLg,8706
@@ -18,13 +18,13 @@ texttools/prompts/translate.yaml,sha256=mGT2uBCei6uucWqVbs4silk-UV060v3G0jnt0P6s
 texttools/tools/__init__.py,sha256=3fPoeB-E5wGxWgv7axztHkeolR7ZDUJudd0xmpPFjao,113
 texttools/tools/async_tools.py,sha256=2ZY7Lo6Jj9xoTF8bfdh_g8VOXZ7ljMMesd1_QHXyf4s,15395
 texttools/tools/sync_tools.py,sha256=XKgZuzriFnk8B-YihJfs6BKivxjGCgOFfe7hnCpEiXs,15161
-texttools/tools/internals/async_operator.py,sha256=fCi70LXasC_2G9iz8uVFptnZEvVeb9TXopMBLi-cFuE,9022
-texttools/tools/internals/base_operator.py,sha256=rV2WqGdiHK4ezYz1f1EWcdbKFSFJhBJpORnJzPICFvk,3471
+texttools/tools/internals/async_operator.py,sha256=egBsrcpGBmkDY5YzUvGHh1TjPmsH9IOVXDGmYMWjzMs,8960
+texttools/tools/internals/base_operator.py,sha256=qV9LlVo_DzSCzQnjYTFi-6mlHN4gE0edPE2y_9WwQFw,3292
 texttools/tools/internals/formatters.py,sha256=tACNLP6PeoqaRpNudVxBaHA25zyWqWYPZQuYysIu88g,941
-texttools/tools/internals/operator.py,sha256=UBDScStTUXf8CIhwXb-6e_YOWTLggoiBV71vXRzr0P0,8904
+texttools/tools/internals/operator.py,sha256=xgbt1Mm67SEC-KD9jwXjXGTCcaCsaVLhG6iCYOqLDcc,8709
 texttools/tools/internals/output_models.py,sha256=ekpbyocmXj_dee7ieOT1zOkMo9cPHT7xcUFCZoUaXA0,1886
-texttools/tools/internals/prompt_loader.py,sha256=1khayXcRC5w0Vf2SufpNaN1IUIhbKzS5ATiKheoBcGE,2082
-hamtaa_texttools-1.1.8.dist-info/METADATA,sha256=Cfb4VkcUELzRN6TrKdWK5jr4YsGbh_VlAtYVny86cb4,8690
-hamtaa_texttools-1.1.8.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-hamtaa_texttools-1.1.8.dist-info/top_level.txt,sha256=5Mh0jIxxZ5rOXHGJ6Mp-JPKviywwN0MYuH0xk5bEWqE,10
-hamtaa_texttools-1.1.8.dist-info/RECORD,,
+texttools/tools/internals/prompt_loader.py,sha256=8uD7JUatKXSLXhGwWs46iQpcjWdhF9p32SFDLMndy1o,1940
+hamtaa_texttools-1.1.9.dist-info/METADATA,sha256=nQFuGr_7aVHlO7nRsTbubEtO0QVUofcdUKwMATzHhUU,9129
+hamtaa_texttools-1.1.9.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+hamtaa_texttools-1.1.9.dist-info/top_level.txt,sha256=5Mh0jIxxZ5rOXHGJ6Mp-JPKviywwN0MYuH0xk5bEWqE,10
+hamtaa_texttools-1.1.9.dist-info/RECORD,,

texttools/tools/internals/async_operator.py

@@ -51,7 +51,7 @@ class AsyncOperator(BaseOperator):
         temperature: float,
         logprobs: bool = False,
         top_logprobs: int = 3,
-    ) -> tuple[Type[T], Any]:
+    ) -> tuple[T, Any]:
         """
         Parses a chat completion using OpenAI's structured output format.
         Returns both the parsed object and the raw completion for logging.
@@ -78,7 +78,7 @@ class AsyncOperator(BaseOperator):
         temperature: float,
         logprobs: bool = False,
         top_logprobs: int = 3,
-    ) -> tuple[Type[T], Any]:
+    ) -> tuple[T, Any]:
         """
         Generates a completion using vLLM with JSON schema guidance.
         Returns the parsed output model and raw completion.
@@ -138,7 +138,7 @@ class AsyncOperator(BaseOperator):
                 **extra_kwargs,
             )
 
-            messages: list[dict[str, str]] = []
+            messages = []
 
             if with_analysis:
                 analysis = await self._analyze(prompt_configs, temperature)
@@ -181,8 +181,7 @@ class AsyncOperator(BaseOperator):
 
             # Retry logic if validation fails
             if validator and not validator(output.result):
-                max_retries = 3
-                for attempt in range(max_retries):
+                for attempt in range(self.MAX_RETRIES):
                     logger.warning(
                         f"Validation failed, retrying for the {attempt + 1} time."
                     )

texttools/tools/internals/base_operator.py

@@ -1,4 +1,4 @@
-from typing import TypeVar, Type, Any
+from typing import TypeVar, Type, Any, Union
 import json
 import re
 import math
@@ -11,11 +11,16 @@ from openai import OpenAI, AsyncOpenAI
 # Base Model type for output models
 T = TypeVar("T", bound=BaseModel)
 
+ClientType = Union[OpenAI, AsyncOpenAI]
+
 logger = logging.getLogger("texttools.base_operator")
 
 
 class BaseOperator:
-    def __init__(self, client: OpenAI | AsyncOpenAI, model: str):
+    # Max retry in case of failed output validation
+    MAX_RETRIES = 3
+
+    def __init__(self, client: ClientType, model: str):
         self.client = client
         self.model = model
 
@@ -40,16 +45,10 @@ class BaseOperator:
         """
         Convert a JSON response string to output model.
         """
-        # Clean the response string
         cleaned_json = self._clean_json_response(response_string)
-
-        # Fix Python-style booleans
         cleaned_json = cleaned_json.replace("False", "false").replace("True", "true")
-
-        # Convert string to Python dictionary
         response_dict = json.loads(cleaned_json)
 
-        # Convert dictionary to output model
         return output_model(**response_dict)
 
     def _extract_logprobs(self, completion: dict) -> list[dict[str, Any]]:
@@ -63,7 +62,7 @@ class BaseOperator:
 
         for choice in completion.choices:
             if not getattr(choice, "logprobs", None):
-                logger.error("logprobs is not avalible in the chosen model.")
+                logger.error("logprobs is not available for the chosen model.")
                 return []
 
             for logprob_item in choice.logprobs.content:
@@ -93,8 +92,5 @@ class BaseOperator:
         """
         delta_temp = random.choice([-1, 1]) * random.uniform(0.1, 0.9)
         new_temp = base_temp + delta_temp
-        print(f"Base Temp: {base_temp}")
-        print(f"Delta Temp: {delta_temp}")
-        print(f"New Temp: {new_temp}")
 
         return max(0.0, min(new_temp, 1.5))

texttools/tools/internals/operator.py

@@ -6,7 +6,6 @@ from pydantic import BaseModel
 
 from texttools.tools.internals.output_models import ToolOutput
 from texttools.tools.internals.base_operator import BaseOperator
-from texttools.tools.internals.formatters import Formatter
 from texttools.tools.internals.prompt_loader import PromptLoader
 
 # Base Model type for output models
@@ -51,7 +50,7 @@ class Operator(BaseOperator):
         temperature: float,
         logprobs: bool = False,
         top_logprobs: int = 3,
-    ) -> tuple[Type[T], Any]:
+    ) -> tuple[T, Any]:
         """
         Parses a chat completion using OpenAI's structured output format.
         Returns both the parsed object and the raw completion for logging.
@@ -78,7 +77,7 @@ class Operator(BaseOperator):
         temperature: float,
         logprobs: bool = False,
         top_logprobs: int = 3,
-    ) -> tuple[Type[T], Any]:
+    ) -> tuple[T, Any]:
         """
         Generates a completion using vLLM with JSON schema guidance.
         Returns the parsed output model and raw completion.
@@ -126,7 +125,6 @@ class Operator(BaseOperator):
         Execute the LLM pipeline with the given input text.
         """
         prompt_loader = PromptLoader()
-        formatter = Formatter()
         output = ToolOutput()
 
         try:
@@ -138,7 +136,7 @@ class Operator(BaseOperator):
                 **extra_kwargs,
             )
 
-            messages: list[dict[str, str]] = []
+            messages = []
 
             if with_analysis:
                 analysis = self._analyze(prompt_configs, temperature)
@@ -159,7 +157,7 @@ class Operator(BaseOperator):
                 )
 
             messages.append(self._build_user_message(prompt_configs["main_template"]))
-            messages = formatter.user_merge_format(messages)
+            messages
 
             if resp_format == "vllm":
                 parsed, completion = self._vllm_completion(
@@ -181,8 +179,7 @@ class Operator(BaseOperator):
 
             # Retry logic if validation fails
             if validator and not validator(output.result):
-                max_retries = 3
-                for attempt in range(max_retries):
+                for attempt in range(self.MAX_RETRIES):
                     logger.warning(
                         f"Validation failed, retrying for the {attempt + 1} time."
                     )

texttools/tools/internals/prompt_loader.py

@@ -11,15 +11,10 @@ class PromptLoader:
     - Load and parse YAML prompt definitions.
     - Select the right template (by mode, if applicable).
     - Inject variables (`{input}`, plus any extra kwargs) into the templates.
-    - Return a dict with:
-        {
-            "main_template": "...",
-            "analyze_template": "..." | None
-        }
     """
 
-    MAIN_TEMPLATE: str = "main_template"
-    ANALYZE_TEMPLATE: str = "analyze_template"
+    MAIN_TEMPLATE = "main_template"
+    ANALYZE_TEMPLATE = "analyze_template"
 
     # Use lru_cache to load each file once
     @lru_cache(maxsize=32)