assistant/ConfigureAssistant.py

@@ -2,6 +2,7 @@ import os
 import io
 from openai import OpenAI
 
+
 class ConfigureAssistant:
     """
     A class to configure an OpenAI assistant for aiding designers using the ArchiMajor project.
@@ -9,9 +10,32 @@ class ConfigureAssistant:
     """
 
     SUPPORTED_FORMATS = {
-        "c", "cpp", "css", "docx", "gif", "html", "java", "jpeg", "jpg", "js",
+        "c",
-        "json", "md", "pdf", "php", "png", "pptx", "py", "rb", "tar", "tex", "ts", "txt",
+        "cpp",
-        "webp", "xlsx", "xml", "zip",
+        "css",
+        "docx",
+        "gif",
+        "html",
+        "java",
+        "jpeg",
+        "jpg",
+        "js",
+        "json",
+        "md",
+        "pdf",
+        "php",
+        "png",
+        "pptx",
+        "py",
+        "rb",
+        "tar",
+        "tex",
+        "ts",
+        "txt",
+        "webp",
+        "xlsx",
+        "xml",
+        "zip",
         # "csv", # CSV is supported but not actually parsed so we're going to treat it as text
     }
 
@@ -76,7 +100,7 @@ class ConfigureAssistant:
 
         # TO DO: Preprocess Outjob and PcbDoc files into something OpenAI (or future vector DB) can understand
         excluded_extensions = ["schdoc", "exe", "so", "dll", "outjob", "pcbdoc", "png"]
-        
+
         if extension in self.SUPPORTED_FORMATS and extension not in excluded_extensions:
             return os.path.basename(file_path), True
         else:
@@ -108,8 +132,10 @@ class ConfigureAssistant:
                     spoofed_file = io.BytesIO(file_content)
                     spoofed_file.name = new_filename  # Spoof the filename
                     # Upload the file to the vector store
-                    file_batch = self.client.beta.vector_stores.file_batches.upload_and_poll(
+                    file_batch = (
-                        vector_store_id=self.vector_store.id, files=[spoofed_file]
+                        self.client.beta.vector_stores.file_batches.upload_and_poll(
+                            vector_store_id=self.vector_store.id, files=[spoofed_file]
+                        )
                     )
                     print(f"Successfully uploaded: {new_filename}")
             except Exception as e:
@@ -144,16 +170,16 @@ if __name__ == "__main__":
     root_path = os.path.dirname(os.path.dirname(__file__))
     # Create an instance of ConfigureAssistant with the root path
     configurator = ConfigureAssistant(root_path=root_path)
-    
+
     # Retrieve file paths
     file_paths = configurator.get_file_paths(excluded_folders=[".git"])
-    
+
     # Preprocess and observe the files that will be uploaded
     print("Files to be uploaded:")
     for path in file_paths:
         new_filename, should_upload = configurator.preprocess_file(path)
         if should_upload:
             print(f"Original: {path}, New: {new_filename}")
-    
+
     # Configure the assistant
     configurator.configure()

assistant/OpenAIResourceManager.py

@@ -3,6 +3,7 @@ import logging
 import pandas as pd
 from openai import OpenAI
 
+
 class OpenAIResourceManager:
     """
     A class to manage OpenAI resources such as assistants, vector stores, and files.
@@ -112,14 +113,20 @@ class OpenAIResourceManager:
         :param max_length: The maximum length of the string.
         :return: The truncated string.
         """
-        return (s[:max_length] + '...') if len(s) > max_length else s
+        return (s[:max_length] + "...") if len(s) > max_length else s
 
     def show_all_assistants(self):
         """
         Display all assistants in a table.
         """
         assistants = self.get_all_assistants()
-        assistant_data = [{k: self.truncate_string(str(v), max_length=25) for k, v in assistant.dict().items()} for assistant in assistants]
+        assistant_data = [
+            {
+                k: self.truncate_string(str(v), max_length=25)
+                for k, v in assistant.dict().items()
+            }
+            for assistant in assistants
+        ]
         df = pd.DataFrame(assistant_data)
         print("Assistants:")
         print(df.to_markdown(index=False))
@@ -129,7 +136,10 @@ class OpenAIResourceManager:
         Display all vector stores in a table.
         """
         vector_stores = self.get_all_vector_stores()
-        vector_store_data = [{k: self.truncate_string(str(v)) for k, v in vector_store.dict().items()} for vector_store in vector_stores]
+        vector_store_data = [
+            {k: self.truncate_string(str(v)) for k, v in vector_store.dict().items()}
+            for vector_store in vector_stores
+        ]
         df = pd.DataFrame(vector_store_data)
         print("Vector Stores:")
         print(df.to_markdown(index=False))
@@ -139,7 +149,10 @@ class OpenAIResourceManager:
         Display all files in a table.
         """
         files = self.get_all_files()
-        file_data = [{k: self.truncate_string(str(v)) for k, v in file.dict().items()} for file in files]
+        file_data = [
+            {k: self.truncate_string(str(v)) for k, v in file.dict().items()}
+            for file in files
+        ]
         df = pd.DataFrame(file_data)
         print("Files:")
         print(df.to_markdown(index=False))

assistant/QueryAssistant.py

@@ -8,6 +8,7 @@ from typing_extensions import override
 # Configure logging
 logging.basicConfig(level=logging.INFO)
 
+
 class QueryAssistant:
     """
     A class to manage querying an OpenAI assistant.
@@ -72,22 +73,23 @@ class QueryAssistant:
         """
         logging.info(f"Fetching response for thread {thread_id}...")
         run = self.client.beta.threads.runs.create_and_poll(
-            thread_id=thread_id,
+            thread_id=thread_id, assistant_id=self.assistant_id
-            assistant_id=self.assistant_id
         )
 
         # Poll the run status with a delay to reduce the number of GET requests
-        while run.status != 'completed' and run.status != 'failed':
+        while run.status != "completed" and run.status != "failed":
             time.sleep(2)  # Add a 2-second delay between checks
-            run = self.client.beta.threads.runs.retrieve(thread_id=thread_id, run_id=run.id)
+            run = self.client.beta.threads.runs.retrieve(
+                thread_id=thread_id, run_id=run.id
+            )
             logging.info(f"Run status: {run.status}")
 
-        if run.status == 'completed':
+        if run.status == "completed":
             messages = self.client.beta.threads.messages.list(thread_id=thread_id).data
             for message in messages:
-                if message.role == 'assistant':
+                if message.role == "assistant":
                     for content in message.content:
-                        if content.type == 'text':
+                        if content.type == "text":
                             print(content.text.value)
         else:
             logging.error(f"Run failed with status: {run.status}")
@@ -144,6 +146,7 @@ class QueryAssistant:
                         if output.type == "logs":
                             print(f"\n{output.logs}", flush=True)
 
+
 def main(query: str, assistant_id: str, context: str, use_streaming: bool):
     """
     The main function to run the assistant query.
@@ -169,6 +172,7 @@ def main(query: str, assistant_id: str, context: str, use_streaming: bool):
         assistant.fetch_response(thread_id=thread.id)
     print("\n")
 
+
 if __name__ == "__main__":
     # Default query and context
     DEFAULT_QUERY = "What are you capable of as an assistant?"