Sweep: Auto Summarize Working Context (✓ Sandbox Passed) #24
Conversation
![sweep-ai[bot]](https://avatars.githubusercontent.com/in/307814?s=60&v=4){kind=small}
Sandbox Executions
Check backend/memory/memory_manager.py with contents:
Check backend/tests/test_memory_manager.py with contents:
Check backend/agent/coding_agent.py with contents: |
Rollback Files For Sweep
|
Apply Sweep Rules to your PR?
|
PR Analysis
PR Feedback💡 General suggestions: The PR is well-structured and includes relevant tests for the new functionalities. However, it would be beneficial to handle potential exceptions that might occur during the summarization process. Also, consider using a more descriptive variable name than 🤖 Code feedback:
✨ Usage tips:
|
|
Add gensim to requirements.txt |
🚀 Wrote ChangesDone. |
|
Lets use OpenAI for the summary. |
🚀 Wrote ChangesDone. |
|
/improve |
🚀 Wrote ChangesDone. |
![github-actions[bot]](https://avatars.githubusercontent.com/in/15368?s=60&v=4){kind=small}
| if self.memory_manager.working_context.turn_counter == 0: | ||
| print("Summarized Context: ", self.memory_manager.working_context.context) |
There was a problem hiding this comment.
Suggestion: Instead of directly printing the summarized context, consider returning it as part of the response from the query method. This would make the output more testable and flexible for different interfaces. [enhancement]
| if self.memory_manager.working_context.turn_counter == 0: | |
| print("Summarized Context: ", self.memory_manager.working_context.context) | |
| if self.memory_manager.working_context.turn_counter == 0: | |
| summarized_context = self.memory_manager.working_context.context | |
| return {"Summarized Context": summarized_context} |
backend/memory/memory_manager.py
Outdated
| def summarize_context(self): | ||
| # Call OpenAI API to get the summary | ||
| response = self.client.create_summary(self.context) | ||
| return response.choices[0].text.strip() |
There was a problem hiding this comment.
Suggestion: The summarize_context method currently returns the summary directly. It would be better to update the context attribute of the WorkingContext object within this method. This would ensure that the summarized context is used in subsequent operations. [enhancement]
| def summarize_context(self): | |
| # Call OpenAI API to get the summary | |
| response = self.client.create_summary(self.context) | |
| return response.choices[0].text.strip() | |
| def summarize_context(self): | |
| # Call OpenAI API to get the summary | |
| response = self.client.create_summary(self.context) | |
| self.context = response.choices[0].text.strip() |
backend/memory/memory_manager.py
Outdated
| self.turn_counter += 1 | ||
| if self.turn_counter == 5: | ||
| self.context = self.summarize_context() | ||
| self.turn_counter = 0 |
There was a problem hiding this comment.
Suggestion: The update_context method currently increments the turn_counter and resets it after 5 turns. However, it does not handle the case where the turn_counter exceeds 5. Consider adding a condition to reset the turn_counter when it exceeds 5. [bug]
| self.turn_counter += 1 | |
| if self.turn_counter == 5: | |
| self.context = self.summarize_context() | |
| self.turn_counter = 0 | |
| self.turn_counter = (self.turn_counter + 1) % 6 | |
| if self.turn_counter == 0: | |
| self.context = self.summarize_context() |
backend/tests/test_memory_manager.py
Outdated
| # Arrange: Reset the turn counter to 0 | ||
| self.memory_manager.working_context.turn_counter = 0 | ||
| # Act: Call the update_context method 5 times | ||
| for _ in range(5): | ||
| asyncio.run(self.memory_manager.update_context()) | ||
| # Assert: Check whether the turn_counter is reset to 0 | ||
| assert self.memory_manager.working_context.turn_counter == 0, 'Turn counter was not reset after 5th update' |
There was a problem hiding this comment.
Suggestion: The test test_turn_counter_reset_after_5th_update currently checks if the turn_counter is reset to 0 after 5 updates. However, it does not check if the context is summarized. Consider adding an assertion to check if the context is summarized after 5 updates. [enhancement]
| # Arrange: Reset the turn counter to 0 | |
| self.memory_manager.working_context.turn_counter = 0 | |
| # Act: Call the update_context method 5 times | |
| for _ in range(5): | |
| asyncio.run(self.memory_manager.update_context()) | |
| # Assert: Check whether the turn_counter is reset to 0 | |
| assert self.memory_manager.working_context.turn_counter == 0, 'Turn counter was not reset after 5th update' | |
| def test_turn_counter_reset_after_5th_update(self): | |
| # Arrange: Reset the turn counter to 0 | |
| self.memory_manager.working_context.turn_counter = 0 | |
| original_context = self.memory_manager.working_context.context | |
| # Act: Call the update_context method 5 times | |
| for _ in range(5): | |
| asyncio.run(self.memory_manager.update_context()) | |
| # Assert: Check whether the turn_counter is reset to 0 and context is summarized | |
| assert self.memory_manager.working_context.turn_counter == 0, 'Turn counter was not reset after 5th update' | |
| assert self.memory_manager.working_context.context != original_context, 'Context was not summarized after 5th update' |
|
Says it failed the pytest check |
🚀 Wrote ChangesDone. |
PR Feedback (click)
Description
This pull request introduces changes to the codebase that implement auto summarization of the working context in the coding agent. It also includes tests to verify the functionality of the new features.
Summary
querymethod ofcoding_agent.pyto print the summarized context if the turn counter is 0.summarizefunction fromgensim.summarizeinmemory_manager.py.summarize_contextinmemory_manager.pyto summarize the context using the imported function.update_contextmethod inmemory_manager.pyto update the context and reset the turn counter to 0 after every 5th update.test_memory_manager.pyto verify the functionality ofsummarize_contextand the turn counter reset.Fixes #23.
🎉 Latest improvements to Sweep:
💡 To get Sweep to edit this pull request, you can: