langroid 0.34.0__py3-none-any.whl → 0.34.1__py3-none-any.whl

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/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.dist-info → langroid-0.34.1.dist-info}/METADATA

@@ -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.dist-info → langroid-0.34.1.dist-info}/RECORD

@@ -14,7 +14,7 @@ langroid/agent/xml_tool_message.py,sha256=6SshYZJKIfi4mkE-gIoSwjkEYekQ8GwcSiCv7a
 langroid/agent/callbacks/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 langroid/agent/callbacks/chainlit.py,sha256=RH8qUXaZE5o2WQz3WJQ1SdFtASGlxWCA6_HYz_3meDQ,20822
 langroid/agent/special/__init__.py,sha256=gik_Xtm_zV7U9s30Mn8UX3Gyuy4jTjQe9zjiE3HWmEo,1273
-langroid/agent/special/doc_chat_agent.py,sha256=tI16jVavTSOen9OUoRTl5heDTeTBhWsxW17XU9ZcEko,63563
+langroid/agent/special/doc_chat_agent.py,sha256=EPRKhchk1nuExn4_Lbuu5hzGkBP06PvJAK1edoISQHc,63470
 langroid/agent/special/lance_doc_chat_agent.py,sha256=s8xoRs0gGaFtDYFUSIRchsgDVbS5Q3C2b2mr3V1Fd-Q,10419
 langroid/agent/special/lance_tools.py,sha256=qS8x4wi8mrqfbYV2ztFzrcxyhHQ0ZWOc-zkYiH7awj0,2105
 langroid/agent/special/relevance_extractor_agent.py,sha256=zIx8GUdVo1aGW6ASla0NPQjYYIpmriK_TYMijqAx3F8,4796
@@ -81,7 +81,7 @@ langroid/parsing/code_parser.py,sha256=AOxb3xbYpTBPP3goOm5dKfJdh5hS_2BhLVCEkifWZ
 langroid/parsing/document_parser.py,sha256=9xUOyrVNBAS9cpCvCptr2XK4Kq47W574i8zzGEoXc3c,24933
 langroid/parsing/para_sentence_split.py,sha256=AJBzZojP3zpB-_IMiiHismhqcvkrVBQ3ZINoQyx_bE4,2000
 langroid/parsing/parse_json.py,sha256=aADo38bAHQhC8on4aWZZzVzSDy-dK35vRLZsFI2ewh8,4756
-langroid/parsing/parser.py,sha256=bTG5TO2CEwGdLf9979j9_dFntKX5FloGF8vhts6ObU0,11978
+langroid/parsing/parser.py,sha256=N0jr1Zl_f_rx-8YMmSQftPHquqSQfec-3s7JAhhEe6I,12032
 langroid/parsing/repo_loader.py,sha256=3GjvPJS6Vf5L6gV2zOU8s-Tf1oq_fZm-IB_RL_7CTsY,29373
 langroid/parsing/routing.py,sha256=-FcnlqldzL4ZoxuDwXjQPNHgBe9F9-F4R6q7b_z9CvI,1232
 langroid/parsing/search.py,sha256=0i_r0ESb5HEQfagA2g7_uMQyxYPADWVbdcN9ixZhS4E,8992
@@ -121,7 +121,7 @@ langroid/vector_store/lancedb.py,sha256=b3_vWkTjG8mweZ7ZNlUD-NjmQP_rLBZfyKWcxt2v
 langroid/vector_store/meilisearch.py,sha256=6frB7GFWeWmeKzRfLZIvzRjllniZ1cYj3HmhHQICXLs,11663
 langroid/vector_store/momento.py,sha256=UNHGT6jXuQtqY9f6MdqGU14bVnS0zHgIJUa30ULpUJo,10474
 langroid/vector_store/qdrantdb.py,sha256=HRLCt-FG8y4718omwpFaQZnWeYxPj0XCwS4tjokI1sU,18116
-langroid-0.34.0.dist-info/METADATA,sha256=fo7ULfjnWFED6Cag8aUFjOaPqEatQKBXEz-Z_rFyHnk,59015
-langroid-0.34.0.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
-langroid-0.34.0.dist-info/licenses/LICENSE,sha256=EgVbvA6VSYgUlvC3RvPKehSg7MFaxWDsFuzLOsPPfJg,1065
-langroid-0.34.0.dist-info/RECORD,,
+langroid-0.34.1.dist-info/METADATA,sha256=_sKdjcKkVQBqsbSe4p79aSmHTjZ1ClwX9GXO1Fn15y0,59313
+langroid-0.34.1.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
+langroid-0.34.1.dist-info/licenses/LICENSE,sha256=EgVbvA6VSYgUlvC3RvPKehSg7MFaxWDsFuzLOsPPfJg,1065
+langroid-0.34.1.dist-info/RECORD,,