hamtaa-texttools 0.1.43__py3-none-any.whl → 0.1.44__py3-none-any.whl

{hamtaa_texttools-0.1.43.dist-info → hamtaa_texttools-0.1.44.dist-info}/METADATA

@@ -1,12 +1,12 @@
 Metadata-Version: 2.4
 Name: hamtaa-texttools
-Version: 0.1.43
+Version: 0.1.44
 Summary: A set of high-level NLP tools
 Author: Tohidi, Montazer, Givechi, Mousavinezhad
 Requires-Python: >=3.8
 Description-Content-Type: text/markdown
-Requires-Dist: openai==1.97.1
-Requires-Dist: numpy==1.26.4
+Requires-Dist: openai>=1.97.0
+Requires-Dist: numpy>=1.26.4
 
 # Text Tools
 
@@ -20,7 +20,7 @@ Requires-Dist: numpy==1.26.4
 </p>
 
 
-### How to Install
+## How to Install
 
 Install the package using:
 

{hamtaa_texttools-0.1.43.dist-info → hamtaa_texttools-0.1.44.dist-info}/RECORD

@@ -16,7 +16,7 @@ texttools/batch_manager/batch_manager.py,sha256=jAmKskL3OTYwwsO1mWsWAB3VxMlOF07c
 texttools/batch_manager/batch_runner.py,sha256=kW0IPauI11xpssApMA7b4XI19FePImywym3V7tBaa-o,7404
 texttools/formatter/__init__.py,sha256=KHz2tFZctbit_HVbQNCTMi46JzmKlg-uB6Ost63IpVU,46
 texttools/formatter/base.py,sha256=0fiM6E7NdJevAVpL6yyPaUZVJGKWxE3fr-Ay1oqgJqQ,879
-texttools/formatter/gemma3_formatter.py,sha256=c7YRj6fIPqhs_nvnSbWRTuguRoNQJvuIvk_bcaVDioM,1634
+texttools/formatter/gemma3_formatter.py,sha256=AmdKBYLj6HMsI2DDX4KHNEEVYJmz_VVNUBOv8ScGjsY,1865
 texttools/handlers/__init__.py,sha256=sv0JloipQ57AI0xo-3w9k6cK5rYjZP3ltR2EbBhkHTA,121
 texttools/handlers/handlers.py,sha256=LtC4FBuzRUDy3Jw-Fp21WR-QS1jOcDhsGaMPFQGjfTw,2381
 texttools/handlers/categorizer/__init__.py,sha256=mE05vt_ma6vcP8pQ37BZ85WVQ8jhcjDS0iZV81_LFCY,127
@@ -35,7 +35,7 @@ texttools/tools/merger/gemma_question_merger.py,sha256=JAC-52kBbabIzEWp0MFi9viiu
 texttools/tools/ner/__init__.py,sha256=BW84BcItel6Mc2JlaDL6qvAktVMkti67VXceeCnOB1g,70
 texttools/tools/ner/gemma_ner_extractor.py,sha256=YhyIwX_8bdwkFb4gY8g9mZdYHW_r1jCvbmjjNCK9Wfo,5384
 texttools/tools/question_detector/__init__.py,sha256=ulArGttooSoxEe0vUDQSxUQrnsxr7gH9l-LjSER2dVI,162
-texttools/tools/question_detector/gemma_detector.py,sha256=dHWHcthjMArW42CNPGmk3Xbj1AxjM33A34dOmLUA64U,4141
+texttools/tools/question_detector/gemma_detector.py,sha256=DhlCAA6Hws_OTuYil6UY4sYlbjdQQU6EqHdoTl3a--w,3772
 texttools/tools/question_detector/llm_detector.py,sha256=zo89eh359hqQGGf83-6M22AaiH7q-m0m91SjTyxZaYs,3862
 texttools/tools/question_generator/__init__.py,sha256=EAElpB_YeyMoBqvFNjbW2a_j18SLtiKQ7sRmdS58Fww,61
 texttools/tools/question_generator/gemma_question_generator.py,sha256=V5QcXmHZ5shTvrThOxUrKJ4FqP0P58NIJbsPdyyy5IM,6744
@@ -54,7 +54,7 @@ texttools/tools/summarizer/gemma_summarizer.py,sha256=ikhsBv7AiZD1dT_d12AyjXxojz
 texttools/tools/summarizer/llm_summerizer.py,sha256=-0rUKbSnl1aDeBfJ5DCSbIlwd2k-9qIaCKgoQJa0hWc,3412
 texttools/tools/translator/__init__.py,sha256=KO1m08J2BZwRqBGO9ICB4l4cnH1jfHLHL5HbgYFUWM8,72
 texttools/tools/translator/gemma_translator.py,sha256=57NMfJAZHQjZSr_eCBePE_Pnag8pu3O00Jicxhzn6Jc,7572
-hamtaa_texttools-0.1.43.dist-info/METADATA,sha256=GjVLyZZclY4hp29Yd1DpRtqvFDmTAGOoYEOI-FFvbA0,1482
-hamtaa_texttools-0.1.43.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-hamtaa_texttools-0.1.43.dist-info/top_level.txt,sha256=5Mh0jIxxZ5rOXHGJ6Mp-JPKviywwN0MYuH0xk5bEWqE,10
-hamtaa_texttools-0.1.43.dist-info/RECORD,,
+hamtaa_texttools-0.1.44.dist-info/METADATA,sha256=OImC1zmuJh7p8SY3s3mhm8poOzYOuuqx6vjOeDy5O3k,1481
+hamtaa_texttools-0.1.44.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+hamtaa_texttools-0.1.44.dist-info/top_level.txt,sha256=5Mh0jIxxZ5rOXHGJ6Mp-JPKviywwN0MYuH0xk5bEWqE,10
+hamtaa_texttools-0.1.44.dist-info/RECORD,,

texttools/formatter/gemma3_formatter.py

@@ -1,5 +1,3 @@
-from typing import Literal
-
 from texttools.formatter.base import ChatFormatter
 
 
@@ -10,14 +8,13 @@ class Gemma3Formatter(ChatFormatter):
     """
 
     ROLE = "role"
+    CONTENT = "content"
     USER_ROLE = "user"
     ASSISTANT_ROLE = "assistant"
-    CONTENT = "content"
     VALID_ROLES = {USER_ROLE, ASSISTANT_ROLE}
+    VALID_KEYS = {ROLE, CONTENT}
 
-    def format(
-        self, messages: list[dict[Literal["role", "content"], str]]
-    ) -> list[dict[str, str]]:
+    def format(self, messages: list[dict[str, str]]) -> list[dict[str, str]]:
         """
         :param messages: list of {"role": ..., "content": ...}, where role is "user", "assistant", or "system"
         :return: a new list where consecutive "user" messages are merged into single entries
@@ -25,14 +22,20 @@ class Gemma3Formatter(ChatFormatter):
 
         merged: list[dict[str, str]] = []
 
-        for msg in messages:
-            role, content = msg[self.ROLE], msg[self.CONTENT].strip()
+        for message in messages:
+            # Validate keys strictly
+            if set(message.keys()) != self.VALID_KEYS:
+                raise ValueError(
+                    f"Message dict keys must be exactly {self.VALID_KEYS}, got {set(message.keys())}"
+                )
+
+            role, content = message[self.ROLE], message[self.CONTENT].strip()
 
             # Replace "system" role with "user" role
             if role == "system":
                 role = self.USER_ROLE
 
-            # Raise value error if msg["role"] wan't a valid role
+            # Raise value error if message["role"] wan't a valid role
             if role not in self.VALID_ROLES:
                 raise ValueError(f"Unexpected role: {role}")
 

texttools/tools/question_detector/gemma_detector.py

@@ -36,64 +36,48 @@ class GemmaQuestionDetector(BaseQuestionDetector):
         self.model = model
         self.temperature = temperature
         self.client_kwargs = client_kwargs
-
         self.chat_formatter = chat_formatter or Gemma3Formatter()
-
         self.use_reason = use_reason
         self.prompt_template = prompt_template
 
-        self.json_schema = {"is_question": bool}
-
     def _build_messages(self, text: str, reason: str = None) -> list[dict[str, str]]:
-        clean = self.preprocess(text)
-        schema_instr = f"respond only in JSON format: {self.json_schema}"
+        clean_text = self.preprocess(text)
         messages: list[dict[str, str]] = []
 
         if reason:
             messages.append({"role": "user", "content": reason})
 
-        messages.append({"role": "user", "content": schema_instr})
         if self.prompt_template:
             messages.append({"role": "user", "content": self.prompt_template})
-        messages.append({"role": "user", "content": clean})
+        messages.append({"role": "user", "content": clean_text})
 
-        # this line will restructure the messages
-        # based on the formatter that we provided
-        # some models will require custom settings
+        # Restructure the messages based on the formatter; some models will require custom settings
         restructured = self.chat_formatter.format(messages=messages)
 
         return restructured
 
     def _reason(self, text: str) -> list:
-        messages = [
-            {
-                "role": "user",
-                "content": """
-                    we want to analyze this text snippet to see if it contains any question
-                    or request of some kind or not
-                    read the text, and reason about it being a request or not
-                    summerized
-                    short answer
-                    """,
-            },
-            {
-                "role": "user",
-                "content": f"""
+        reason_prompt = f"""
+                    We want to analyze this text snippet to see if it contains any question
+                    or request of some kind or not.
+                    Read the text, and reason about it being a request or not.
+                    Summerized, Short answer
                     {text}
-                    """,
-            },
+                    """
+        messages = [
+            {"role": "user", "content": reason_prompt},
         ]
 
         restructured = self.chat_formatter.format(messages=messages)
 
-        resp = self.client.chat.completions.create(
+        response = self.client.chat.completions.create(
             model=self.model,
             messages=restructured,
             temperature=self.temperature,
             **self.client_kwargs,
         )
 
-        reason = resp.choices[0].message.content.strip()
+        reason = response.choices[0].message.content.strip()
         return reason
 
     def detect(self, text: str) -> bool:
@@ -125,6 +109,6 @@ class GemmaQuestionDetector(BaseQuestionDetector):
                 f"Failed to parse the response. Raw content: {message.content}"
             )
 
-        # dispatch and return
+        # Dispatch and return
         self._dispatch({"question": text, "result": result})
         return result