Merge branch 'aw-dotenv' of https://github.com/AaronWard/autogen into…

… aw-dotenv

README.md

@@ -15,7 +15,7 @@ This project is a spinoff from [FLAML](https://github.com/microsoft/FLAML).
 
 :fire: autogen has graduated from [FLAML](https://github.com/microsoft/FLAML) into a new project.
 
-<!-- :fire: Heads-up: We're preparing to migrate [autogen](https://microsoft.github.io/FLAML/docs/Use-Cases/Autogen) into a dedicated github repository. Alongside this move, we'll also launch a dedicated Discord server and a website for comprehensive documentation.
+<!-- :fire: Heads-up: We're preparing to migrate [autogen](https://microsoft.github.io/FLAML/docs/Use-Cases/Autogen) into a dedicated Github repository. Alongside this move, we'll also launch a dedicated Discord server and a website for comprehensive documentation.
 
 :fire: FLAML is highlighted in OpenAI's [cookbook](https://github.com/openai/openai-cookbook#related-resources-from-around-the-web).
 
@@ -26,17 +26,17 @@ This project is a spinoff from [FLAML](https://github.com/microsoft/FLAML).
 
 ## What is AutoGen
 
-AutoGen is a framework that enables development of LLM applications using multiple agents that can converse with each other to solve tasks. AutoGen agents are customizable, conversable, and seamlessly allow human participation. They can operate in various modes that employ combinations of LLMs, human inputs, and tools.
+AutoGen is a framework that enables the development of LLM applications using multiple agents that can converse with each other to solve tasks. AutoGen agents are customizable, conversable, and seamlessly allow human participation. They can operate in various modes that employ combinations of LLMs, human inputs, and tools.
 
 ![AutoGen Overview](https://github.com/microsoft/autogen/blob/main/website/static/img/autogen_agentchat.png)
 
-* AutoGen enables building next-gen LLM applications based on **multi-agent conversations** with minimal effort. It simplifies the orchestration, automation and optimization of a complex LLM workflow. It maximizes the performance of LLM models and overcomes their weaknesses.
+* AutoGen enables building next-gen LLM applications based on **multi-agent conversations** with minimal effort. It simplifies the orchestration, automation, and optimization of a complex LLM workflow. It maximizes the performance of LLM models and overcomes their weaknesses.
 * It supports **diverse conversation patterns** for complex workflows. With customizable and conversable agents, developers can use AutoGen to build a wide range of conversation patterns concerning conversation autonomy,
 the number of agents, and agent conversation topology.
 * It provides a collection of working systems with different complexities. These systems span a **wide range of applications** from various domains and complexities. This demonstrates how AutoGen can easily support diverse conversation patterns.
 * AutoGen provides a drop-in replacement of `openai.Completion` or `openai.ChatCompletion` as an **enhanced inference API**. It allows easy performance tuning, utilities like API unification and caching, and advanced usage patterns, such as error handling, multi-config inference, context programming, etc.
 
-AutoGen is powered by collaborative [research studies](https://microsoft.github.io/autogen/docs/Research) from Microsoft, Penn State University, and University of Washington.
+AutoGen is powered by collaborative [research studies](https://microsoft.github.io/autogen/docs/Research) from Microsoft, Penn State University, and the University of Washington.
 
 ## Installation
 
@@ -61,7 +61,7 @@ For LLM inference configurations, check the [FAQ](https://microsoft.github.io/au
 
 ## Multi-Agent Conversation Framework
 
-Autogen enables the next-gen LLM applications with a generic multi-agent conversation framework. It offers customizable and conversable agents which integrate LLMs, tools, and humans.
+Autogen enables the next-gen LLM applications with a generic multi-agent conversation framework. It offers customizable and conversable agents that integrate LLMs, tools, and humans.
 By automating chat among multiple capable agents, one can easily make them collectively perform tasks autonomously or with human feedback, including tasks that require using tools via code.
 
 Features of this use case include:
@@ -94,7 +94,7 @@ The figure below shows an example conversation flow with AutoGen.
 Please find more [code examples](https://microsoft.github.io/autogen/docs/Examples/AutoGen-AgentChat) for this feature.
 ## Enhanced LLM Inferences
 
-Autogen also helps maximize the utility out of the expensive LLMs such as ChatGPT and GPT-4. It offers a drop-in replacement of `openai.Completion` or `openai.ChatCompletion` adding powerful functionalities like tuning, caching, error handling, and templating. For example, you can optimize generations by LLM with your own tuning data, success metrics and budgets.
+Autogen also helps maximize the utility out of the expensive LLMs such as ChatGPT and GPT-4. It offers a drop-in replacement of `openai.Completion` or `openai.ChatCompletion` adding powerful functionalities like tuning, caching, error handling, and templating. For example, you can optimize generations by LLM with your own tuning data, success metrics, and budgets.
 ```python
 # perform tuning
 config, analysis = autogen.Completion.tune(
@@ -114,7 +114,7 @@ Please find more [code examples](https://microsoft.github.io/autogen/docs/Exampl
 
 ## Documentation
 
-You can find a detailed documentation about AutoGen [here](https://microsoft.github.io/autogen/).
+You can find detailed documentation about AutoGen [here](https://microsoft.github.io/autogen/).
 
 In addition, you can find:
 
@@ -147,15 +147,15 @@ in this repository under the [Creative Commons Attribution 4.0 International Pub
 see the [LICENSE](LICENSE) file, and grant you a license to any code in the repository under the [MIT License](https://opensource.org/licenses/MIT), see the
 [LICENSE-CODE](LICENSE-CODE) file.
 
-Microsoft, Windows, Microsoft Azure and/or other Microsoft products and services referenced in the documentation
+Microsoft, Windows, Microsoft Azure, and/or other Microsoft products and services referenced in the documentation
 may be either trademarks or registered trademarks of Microsoft in the United States and/or other countries.
 The licenses for this project do not grant you rights to use any Microsoft names, logos, or trademarks.
 Microsoft's general trademark guidelines can be found at http://go.microsoft.com/fwlink/?LinkID=254653.
 
 Privacy information can be found at https://privacy.microsoft.com/en-us/
 
 Microsoft and any contributors reserve all other rights, whether under their respective copyrights, patents,
-or trademarks, whether by implication, estoppel or otherwise.
+or trademarks, whether by implication, estoppel, or otherwise.
 
 
 ## Citation

autogen/agentchat/conversable_agent.py

@@ -2,6 +2,7 @@
 from collections import defaultdict
 import copy
 import json
+import logging
 from typing import Any, Callable, Dict, List, Optional, Tuple, Type, Union
 from autogen import oai
 from .agent import Agent
@@ -21,6 +22,9 @@ def colored(x, *args, **kwargs):
         return x
 
 
+logger = logging.getLogger(__name__)
+
+
 class ConversableAgent(Agent):
     """(In preview) A class for generic conversable agents which can be configured as assistant or user proxy.
 
@@ -757,7 +761,11 @@ def generate_reply(
         Returns:
             str or dict or None: reply. None if no reply is generated.
         """
-        assert messages is not None or sender is not None, "Either messages or sender must be provided."
+        if all((messages is None, sender is None)):
+            error_msg = f"Either {messages=} or {sender=} must be provided."
+            logger.error(error_msg)
+            raise AssertionError(error_msg)
+
         if messages is None:
             messages = self._oai_messages[sender]
 
@@ -804,7 +812,11 @@ async def a_generate_reply(
         Returns:
             str or dict or None: reply. None if no reply is generated.
         """
-        assert messages is not None or sender is not None, "Either messages or sender must be provided."
+        if all((messages is None, sender is None)):
+            error_msg = f"Either {messages=} or {sender=} must be provided."
+            logger.error(error_msg)
+            raise AssertionError(error_msg)
+
         if messages is None:
             messages = self._oai_messages[sender]
 

autogen/code_utils.py

@@ -26,6 +26,8 @@
 WIN32 = sys.platform == "win32"
 PATH_SEPARATOR = WIN32 and "\\" or "/"
 
+logger = logging.getLogger(__name__)
+
 
 def infer_lang(code):
     """infer the language for the code.
@@ -250,7 +252,11 @@ def execute_code(
         str: The error message if the code fails to execute; the stdout otherwise.
         image: The docker image name after container run when docker is used.
     """
-    assert code is not None or filename is not None, "Either code or filename must be provided."
+    if all((code is None, filename is None)):
+        error_msg = f"Either {code=} or {filename=} must be provided."
+        logger.error(error_msg)
+        raise AssertionError(error_msg)
+
     timeout = timeout or DEFAULT_TIMEOUT
     original_filename = filename
     if WIN32 and lang in ["sh", "shell"]:
@@ -276,7 +282,7 @@ def execute_code(
             f".\\{filename}" if WIN32 else filename,
         ]
         if WIN32:
-            logging.warning("SIGALRM is not supported on Windows. No timeout will be enforced.")
+            logger.warning("SIGALRM is not supported on Windows. No timeout will be enforced.")
             result = subprocess.run(
                 cmd,
                 cwd=work_dir,

autogen/math_utils.py

@@ -35,8 +35,9 @@ def remove_boxed(string: str) -> Optional[str]:
     """
     left = "\\boxed{"
     try:
-        assert string[: len(left)] == left
-        assert string[-1] == "}"
+        if not all((string[: len(left)] == left, string[-1] == "}")):
+            raise AssertionError
+
         return string[len(left) : -1]
     except Exception:
         return None
@@ -94,7 +95,8 @@ def _fix_fracs(string: str) -> str:
                 new_str += substr
             else:
                 try:
-                    assert len(substr) >= 2
+                    if not len(substr) >= 2:
+                        raise AssertionError
                 except Exception:
                     return string
                 a = substr[0]
@@ -129,7 +131,8 @@ def _fix_a_slash_b(string: str) -> str:
     try:
         a = int(a_str)
         b = int(b_str)
-        assert string == "{}/{}".format(a, b)
+        if not string == "{}/{}".format(a, b):
+            raise AssertionError
         new_string = "\\frac{" + str(a) + "}{" + str(b) + "}"
         return new_string
     except Exception:
@@ -143,7 +146,8 @@ def _remove_right_units(string: str) -> str:
     """
     if "\\text{ " in string:
         splits = string.split("\\text{ ")
-        assert len(splits) == 2
+        if not len(splits) == 2:
+            raise AssertionError
         return splits[0]
     else:
         return string

autogen/oai/completion.py

@@ -582,23 +582,31 @@ def eval_func(responses, **data):
         cls._prompts = space.get("prompt")
         if cls._prompts is None:
             cls._messages = space.get("messages")
-            assert isinstance(cls._messages, list) and isinstance(
-                cls._messages[0], (dict, list)
-            ), "messages must be a list of dicts or a list of lists."
+            if not all((isinstance(cls._messages, list), isinstance(cls._messages[0], (dict, list)))):
+                error_msg = "messages must be a list of dicts or a list of lists."
+                logger.error(error_msg)
+                raise AssertionError(error_msg)
             if isinstance(cls._messages[0], dict):
                 cls._messages = [cls._messages]
             space["messages"] = tune.choice(list(range(len(cls._messages))))
         else:
-            assert space.get("messages") is None, "messages and prompt cannot be provided at the same time."
-            assert isinstance(cls._prompts, (str, list)), "prompt must be a string or a list of strings."
+            if space.get("messages") is not None:
+                error_msg = "messages and prompt cannot be provided at the same time."
+                logger.error(error_msg)
+                raise AssertionError(error_msg)
+            if not isinstance(cls._prompts, (str, list)):
+                error_msg = "prompt must be a string or a list of strings."
+                logger.error(error_msg)
+                raise AssertionError(error_msg)
             if isinstance(cls._prompts, str):
                 cls._prompts = [cls._prompts]
             space["prompt"] = tune.choice(list(range(len(cls._prompts))))
         cls._stops = space.get("stop")
         if cls._stops:
-            assert isinstance(
-                cls._stops, (str, list)
-            ), "stop must be a string, a list of strings, or a list of lists of strings."
+            if not isinstance(cls._stops, (str, list)):
+                error_msg = "stop must be a string, a list of strings, or a list of lists of strings."
+                logger.error(error_msg)
+                raise AssertionError(error_msg)
             if not (isinstance(cls._stops, list) and isinstance(cls._stops[0], list)):
                 cls._stops = [cls._stops]
             space["stop"] = tune.choice(list(range(len(cls._stops))))
@@ -969,7 +977,10 @@ def eval_func(responses, **data):
         elif isinstance(agg_method, dict):
             for key in metric_keys:
                 metric_agg_method = agg_method[key]
-                assert callable(metric_agg_method), "please provide a callable for each metric"
+                if not callable(metric_agg_method):
+                    error_msg = "please provide a callable for each metric"
+                    logger.error(error_msg)
+                    raise AssertionError(error_msg)
                 result_agg[key] = metric_agg_method([r[key] for r in result_list])
         else:
             raise ValueError(

autogen/retrieve_utils.py

@@ -29,6 +29,7 @@
     "yml",
     "pdf",
 ]
+VALID_CHUNK_MODES = frozenset({"one_line", "multi_lines"})
 
 
 def num_tokens_from_text(
@@ -96,7 +97,8 @@ def split_text_to_chunks(
     overlap: int = 10,
 ):
     """Split a long text into chunks of max_tokens."""
-    assert chunk_mode in {"one_line", "multi_lines"}
+    if chunk_mode not in VALID_CHUNK_MODES:
+        raise AssertionError
     if chunk_mode == "one_line":
         must_break_at_empty_line = False
     chunks = []

setup.py

@@ -39,15 +39,18 @@
     install_requires=install_requires,
     extras_require={
         "test": [
-            "pytest>=6.1.1",
+            "chromadb",
             "coverage>=5.3",
-            "pre-commit",
             "datasets",
+            "ipykernel",
             "nbconvert",
             "nbformat",
-            "ipykernel",
+            "pre-commit",
             "pydantic==1.10.9",
+            "pytest-asyncio",
+            "pytest>=6.1.1",
             "sympy",
+            "tiktoken",
             "wolframalpha",
         ],
         "blendsearch": ["flaml[blendsearch]"],

test/agentchat/test_conversable_agent.py

@@ -2,6 +2,17 @@
 from autogen.agentchat import ConversableAgent
 
 
+@pytest.fixture
+def conversable_agent():
+    return ConversableAgent(
+        "conversable_agent_0",
+        max_consecutive_auto_reply=10,
+        code_execution_config=False,
+        llm_config=False,
+        human_input_mode="NEVER",
+    )
+
+
 def test_trigger():
     agent = ConversableAgent("a0", max_consecutive_auto_reply=0, llm_config=False, human_input_mode="NEVER")
     agent1 = ConversableAgent("a1", max_consecutive_auto_reply=0, human_input_mode="NEVER")
@@ -217,6 +228,17 @@ def add_num(num_to_be_added):
     ), "generate_reply not working when messages is None"
 
 
+def test_generate_reply_raises_on_messages_and_sender_none(conversable_agent):
+    with pytest.raises(AssertionError):
+        conversable_agent.generate_reply(messages=None, sender=None)
+
+
+@pytest.mark.asyncio
+async def test_a_generate_reply_raises_on_messages_and_sender_none(conversable_agent):
+    with pytest.raises(AssertionError):
+        await conversable_agent.a_generate_reply(messages=None, sender=None)
+
+
 if __name__ == "__main__":
     test_trigger()
     # test_context()

test/test_code.py

@@ -264,6 +264,11 @@ def test_execute_code(use_docker=None):
     assert isinstance(image, str) or docker is None or os.path.exists("/.dockerenv") or use_docker is False
 
 
+def test_execute_code_raises_when_code_and_filename_are_both_none():
+    with pytest.raises(AssertionError):
+        execute_code(code=None, filename=None)
+
+
 @pytest.mark.skipif(
     sys.platform in ["darwin"],
     reason="do not run on MacOS",

test/test_retrieve_utils.py

@@ -48,6 +48,10 @@ def test_split_text_to_chunks(self):
         chunks = split_text_to_chunks(long_text, max_tokens=1000)
         assert all(num_tokens_from_text(chunk) <= 1000 for chunk in chunks)
 
+    def test_split_text_to_chunks_raises_on_invalid_chunk_mode(self):
+        with pytest.raises(AssertionError):
+            split_text_to_chunks("A" * 10000, chunk_mode="bogus_chunk_mode")
+
     def test_extract_text_from_pdf(self):
         pdf_file_path = os.path.join(test_dir, "example.pdf")
         assert "".join(expected_text.split()) == "".join(extract_text_from_pdf(pdf_file_path).strip().split())

website/docs/Getting-Started.md

@@ -75,4 +75,4 @@ response = autogen.Completion.create(context=test_instance, **config)
 
 If you like our project, please give it a [star](https://github.com/microsoft/autogen/stargazers) on GitHub. If you are interested in contributing, please read [Contributor's Guide](/docs/Contribute).
 
-<!-- <iframe src="https://ghbtns.com/github-btn.html?user=microsoft&amp;repo=autogen&amp;type=star&amp;count=true&amp;size=large" frameborder="0" scrolling="0" width="170" height="30" title="GitHub"></iframe> -->
+<iframe src="https://ghbtns.com/github-btn.html?user=microsoft&amp;repo=autogen&amp;type=star&amp;count=true&amp;size=large" frameborder="0" scrolling="0" width="170" height="30" title="GitHub"></iframe>

website/yarn.lock

@@ -5475,10 +5475,10 @@ multicast-dns@^7.2.5:
     dns-packet "^5.2.2"
     thunky "^1.0.2"
 
-nanoid@^3.3.4:
-  version "3.3.4"
-  resolved "https://registry.npmmirror.com/nanoid/-/nanoid-3.3.4.tgz#730b67e3cd09e2deacf03c027c81c9d9dbc5e8ab"
-  integrity sha512-MqBkQh/OHTS2egovRtLk45wEyNXwF+cokD+1YPf9u5VfJiRdAiRwB2froX5Co9Rh20xs4siNPm8naNotSD6RBw==
+nanoid@^3.3.6:
+  version "3.3.6"
+  resolved "https://registry.yarnpkg.com/nanoid/-/nanoid-3.3.6.tgz#443380c856d6e9f9824267d960b4236ad583ea4c"
+  integrity sha512-BGcqMMJuToF7i1rt+2PWSNVnWIkGCU78jBG3RxO/bZlnZPK2Cmi2QaffxGO/2RvWi9sL+FAiRiXMgsyxQ1DIDA==
 
 negotiator@0.6.3:
   version "0.6.3"
@@ -6166,11 +6166,11 @@ postcss-zindex@^5.1.0:
   integrity sha512-fgFMf0OtVSBR1va1JNHYgMxYk73yhn/qb4uQDq1DLGYolz8gHCyr/sesEuGUaYs58E3ZJRcpoGuPVoB7Meiq9A==
 
 postcss@^8.2.15, postcss@^8.3.11, postcss@^8.3.5, postcss@^8.3.7:
-  version "8.4.18"
-  resolved "https://registry.npmmirror.com/postcss/-/postcss-8.4.18.tgz#6d50046ea7d3d66a85e0e782074e7203bc7fbca2"
-  integrity sha512-Wi8mWhncLJm11GATDaQKobXSNEYGUHeQLiQqDFG1qQ5UTDPTEvKw0Xt5NsTpktGTwLps3ByrWsBrG0rB8YQ9oA==
+  version "8.4.31"
+  resolved "https://registry.yarnpkg.com/postcss/-/postcss-8.4.31.tgz#92b451050a9f914da6755af352bdc0192508656d"
+  integrity sha512-PS08Iboia9mts/2ygV3eLpY5ghnUcfLV/EXTOW1E2qYxJKGGBUtNjN76FYHnMs36RmARn41bC0AZmn+rR0OVpQ==
   dependencies:
-    nanoid "^3.3.4"
+    nanoid "^3.3.6"
     picocolors "^1.0.0"
     source-map-js "^1.0.2"