hamtaa-texttools 1.1.12__py3-none-any.whl → 1.1.13__py3-none-any.whl

{hamtaa_texttools-1.1.12.dist-info → hamtaa_texttools-1.1.13.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: hamtaa-texttools
-Version: 1.1.12
+Version: 1.1.13
 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

{hamtaa_texttools-1.1.12.dist-info → hamtaa_texttools-1.1.13.dist-info}/RECORD

@@ -1,4 +1,4 @@
-hamtaa_texttools-1.1.12.dist-info/licenses/LICENSE,sha256=Hb2YOBKy2MJQLnyLrX37B4ZVuac8eaIcE71SvVIMOLg,1082
+hamtaa_texttools-1.1.13.dist-info/licenses/LICENSE,sha256=Hb2YOBKy2MJQLnyLrX37B4ZVuac8eaIcE71SvVIMOLg,1082
 texttools/__init__.py,sha256=EZPPNPafVGvBaxjG9anP0piqH3gAC0DdjdAckQeAgNU,251
 texttools/batch/batch_config.py,sha256=FCDXy9TfH7xjd1PHvn_CtdwEQSq-YO5sktiaMZEId58,740
 texttools/batch/batch_runner.py,sha256=zzzVIXedmaq-8fqsFtGRR64F7CtYRLlhQeBu8uMwJQg,9385
@@ -18,12 +18,13 @@ texttools/prompts/text_to_question.yaml,sha256=UheKYpDn6iyKI8NxunHZtFpNyfCLZZe5c
 texttools/prompts/translate.yaml,sha256=mGT2uBCei6uucWqVbs4silk-UV060v3G0jnt0P6sr50,634
 texttools/tools/async_tools.py,sha256=60VAAZyVRxI2rKVFFiCnbY--F4kNtVxYQticE0RyhOs,24677
 texttools/tools/sync_tools.py,sha256=F5TN3KQ_vlF7AC9J0vm2NzjIZC19Ox11tpc9K1SMRwQ,24448
-texttools/tools/internals/async_operator.py,sha256=aHXhCweEhyl8uP4qTeccAQOmAxVrThFOmdtgtlChQWs,6804
+texttools/tools/internals/async_operator.py,sha256=9OzF5FFXYrXX1C6ZDbad1zw9A6BZsDQ65jQVrpqTlPw,6961
+texttools/tools/internals/formatters.py,sha256=tACNLP6PeoqaRpNudVxBaHA25zyWqWYPZQuYysIu88g,941
 texttools/tools/internals/operator_utils.py,sha256=w1k0RJ_W_CRbVc_J2w337VuL-opHpHiCxfhEOwtyuOo,1856
 texttools/tools/internals/output_models.py,sha256=ekpbyocmXj_dee7ieOT1zOkMo9cPHT7xcUFCZoUaXA0,1886
 texttools/tools/internals/prompt_loader.py,sha256=4g6-U8kqrGN7VpNaRcrBcnF-h03PXjUDBP0lL0_4EZY,1953
-texttools/tools/internals/sync_operator.py,sha256=ar1o5apJgDtQEPLCiGBghS_Y3IXfJMEbXOG9EeS7a9Q,6710
-hamtaa_texttools-1.1.12.dist-info/METADATA,sha256=iPj8bakwOVrpkdHe0n2sn36ZjphP1gREP7Ud5Z0uNxU,9179
-hamtaa_texttools-1.1.12.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-hamtaa_texttools-1.1.12.dist-info/top_level.txt,sha256=5Mh0jIxxZ5rOXHGJ6Mp-JPKviywwN0MYuH0xk5bEWqE,10
-hamtaa_texttools-1.1.12.dist-info/RECORD,,
+texttools/tools/internals/sync_operator.py,sha256=zbFLbFvaT9hAdIgpbDv17ljuqqu6ZeIOwCquM4gHTI8,6867
+hamtaa_texttools-1.1.13.dist-info/METADATA,sha256=4qkKZKb9DEb1vx2FMD-xRBDj_LCnDeytl-ea3zMVKdc,9179
+hamtaa_texttools-1.1.13.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+hamtaa_texttools-1.1.13.dist-info/top_level.txt,sha256=5Mh0jIxxZ5rOXHGJ6Mp-JPKviywwN0MYuH0xk5bEWqE,10
+hamtaa_texttools-1.1.13.dist-info/RECORD,,

texttools/tools/internals/async_operator.py

@@ -6,6 +6,7 @@ from pydantic import BaseModel
 
 from texttools.tools.internals.output_models import ToolOutput
 from texttools.tools.internals.operator_utils import OperatorUtils
+from texttools.tools.internals.formatters import Formatter
 from texttools.tools.internals.prompt_loader import PromptLoader
 
 # Base Model type for output models
@@ -92,6 +93,7 @@ class AsyncOperator:
         Execute the async LLM pipeline with the given input text. (Async)
         """
         prompt_loader = PromptLoader()
+        formatter = Formatter()
         output = ToolOutput()
 
         try:
@@ -131,6 +133,8 @@ class AsyncOperator:
                 OperatorUtils.build_user_message(prompt_configs["main_template"])
             )
 
+            messages = formatter.user_merge_format(messages)
+
             parsed, completion = await self._parse_completion(
                 messages, output_model, temperature, logprobs, top_logprobs
             )

texttools/tools/internals/formatters.py

@@ -0,0 +1,24 @@
+class Formatter:
+    @staticmethod
+    def user_merge_format(messages: list[dict[str, str]]) -> list[dict[str, str]]:
+        """
+        Merges consecutive user messages into a single message, separated by newlines.
+
+        This is useful for condensing a multi-turn user input into a single
+        message for the LLM. Assistant and system messages are left unchanged and
+        act as separators between user message groups.
+        """
+        merged: list[dict[str, str]] = []
+
+        for message in messages:
+            role, content = message["role"], message["content"].strip()
+
+            # Merge with previous user turn
+            if merged and role == "user" and merged[-1]["role"] == "user":
+                merged[-1]["content"] += "\n" + content
+
+            # Otherwise, start a new turn
+            else:
+                merged.append({"role": role, "content": content})
+
+        return merged

texttools/tools/internals/sync_operator.py

@@ -6,6 +6,7 @@ from pydantic import BaseModel
 
 from texttools.tools.internals.output_models import ToolOutput
 from texttools.tools.internals.operator_utils import OperatorUtils
+from texttools.tools.internals.formatters import Formatter
 from texttools.tools.internals.prompt_loader import PromptLoader
 
 # Base Model type for output models
@@ -92,6 +93,7 @@ class Operator:
         Execute the LLM pipeline with the given input text.
         """
         prompt_loader = PromptLoader()
+        formatter = Formatter()
         output = ToolOutput()
 
         try:
@@ -131,6 +133,8 @@ class Operator:
                 OperatorUtils.build_user_message(prompt_configs["main_template"])
             )
 
+            messages = formatter.user_merge_format(messages)
+
             parsed, completion = self._parse_completion(
                 messages, output_model, temperature, logprobs, top_logprobs
             )