langroid 0.34.0__tar.gz → 0.34.1__tar.gz

{langroid-0.34.0 → langroid-0.34.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: langroid
-Version: 0.34.0
+Version: 0.34.1
 Summary: Harness LLMs with Multi-Agent Programming
 Author-email: Prasad Chalasani <pchalasani@gmail.com>
 License: MIT
@@ -287,7 +287,11 @@ teacher_task.run()
 <details>
 <summary> <b>Click to expand</b></summary>
 
+- **Jan 2025:**
+  - [0.33.0](https://github.com/langroid/langroid/releases/tag/0.33.3) Move from Poetry to uv!
+  - [0.32.0](https://github.com/langroid/langroid/releases/tag/0.32.0) DeepSeek v3 support.
 - **Dec 2024:**
+  - [0.31.0](https://github.com/langroid/langroid/releases/tag/0.31.0) Azure OpenAI Embeddings
   - [0.30.0](https://github.com/langroid/langroid/releases/tag/0.30.0) Llama-cpp embeddings.
   - [0.29.0](https://github.com/langroid/langroid/releases/tag/0.29.0) Custom Azure OpenAI Client 
   - [0.28.0](https://github.com/langroid/langroid/releases/tag/0.28.0) `ToolMessage`: `_handler` field to override 

{langroid-0.34.0 → langroid-0.34.1}/README.md

@@ -135,7 +135,11 @@ teacher_task.run()
 <details>
 <summary> <b>Click to expand</b></summary>
 
+- **Jan 2025:**
+  - [0.33.0](https://github.com/langroid/langroid/releases/tag/0.33.3) Move from Poetry to uv!
+  - [0.32.0](https://github.com/langroid/langroid/releases/tag/0.32.0) DeepSeek v3 support.
 - **Dec 2024:**
+  - [0.31.0](https://github.com/langroid/langroid/releases/tag/0.31.0) Azure OpenAI Embeddings
   - [0.30.0](https://github.com/langroid/langroid/releases/tag/0.30.0) Llama-cpp embeddings.
   - [0.29.0](https://github.com/langroid/langroid/releases/tag/0.29.0) Custom Azure OpenAI Client 
   - [0.28.0](https://github.com/langroid/langroid/releases/tag/0.28.0) `ToolMessage`: `_handler` field to override 

{langroid-0.34.0 → langroid-0.34.1}/langroid/agent/special/doc_chat_agent.py

@@ -72,14 +72,13 @@ def apply_nest_asyncio() -> None:
 
 logger = logging.getLogger(__name__)
 
-DEFAULT_DOC_CHAT_INSTRUCTIONS = """
-Your task is to answer questions about various documents.
-You will be given various passages from these documents, and asked to answer questions
-about them, or summarize them into coherent answers.
-"""
 
 DEFAULT_DOC_CHAT_SYSTEM_MESSAGE = """
 You are a helpful assistant, helping me understand a collection of documents.
+
+Your TASK is to answer questions about various documents.
+You will be given various passages from these documents, and asked to answer questions
+about them, or summarize them into coherent answers.
 """
 
 CHUNK_ENRICHMENT_DELIMITER = "<##-##-##>"
@@ -113,7 +112,6 @@ class ChunkEnrichmentAgentConfig(ChatAgentConfig):
 
 class DocChatAgentConfig(ChatAgentConfig):
     system_message: str = DEFAULT_DOC_CHAT_SYSTEM_MESSAGE
-    user_message: str = DEFAULT_DOC_CHAT_INSTRUCTIONS
     summarize_prompt: str = SUMMARY_ANSWER_PROMPT_GPT4
     # extra fields to include in content as key=value pairs
     # (helps retrieval for table-like data)

{langroid-0.34.0 → langroid-0.34.1}/langroid/parsing/parser.py

@@ -1,4 +1,5 @@
 import logging
+import re
 from enum import Enum
 from typing import Dict, List, Literal
 
@@ -250,12 +251,12 @@ class Parser:
                 continue
 
             # Find the last period or punctuation mark in the chunk
-            last_punctuation = max(
-                chunk_text.rfind("."),
-                chunk_text.rfind("?"),
-                chunk_text.rfind("!"),
-                chunk_text.rfind("\n"),
-            )
+            punctuation_matches = [
+                (m.start(), m.group())
+                for m in re.finditer(r"(?:[.!?][\s\n]|\n)", chunk_text)
+            ]
+
+            last_punctuation = max([pos for pos, _ in punctuation_matches] + [-1])
 
             # If there is a punctuation mark, and the last punctuation index is
             # after MIN_CHUNK_SIZE_CHARS
@@ -268,7 +269,7 @@ class Parser:
 
             # Remove any newline characters and strip any leading or
             # trailing whitespace
-            chunk_text_to_append = chunk_text.replace("\n", " ").strip()
+            chunk_text_to_append = re.sub(r"\n{2,}", "\n", chunk_text).strip()
 
             if len(chunk_text_to_append) > self.config.discard_chunk_chars:
                 # Append the chunk text to the list of chunks

{langroid-0.34.0 → langroid-0.34.1}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "langroid"
-version = "0.34.0"
+version = "0.34.1"
 authors = [
     {name = "Prasad Chalasani", email = "pchalasani@gmail.com"},
 ]