checkpoint is disabled with .sql

[XiaoXiaoJiangYun]

App Version

3.25.21

API Provider

Google Gemini

Model Used

gemini 2.5 flash

Roo Code Task Links (Optional)

No response

🔁 Steps to Reproduce

checkpoint is disabled with .sql

1. can not diff or restore
2. change to .py, everything is OK

💥 Outcome Summary

.sql can use checkpoint

📄 Relevant Logs or Errors (Optional)

[roomote]

I'm exploring a potential fix for the checkpoint functionality with .sql files and will open a PR to share progress.

[roomote]

Opened a PR with my attempt at addressing this issue: #7311

The fix removes *.sql from the checkpoint exclusion patterns, allowing SQL files to use checkpoint functionality (diff/restore) just like other code files. All tests are passing and CI checks have completed successfully. Feedback welcome!

[daniel-lxs]

Hi @XiaoXiaoJiangYun, thank you for reporting this issue!

I've investigated the checkpoint behavior with SQL files and wanted to provide some context on why they're currently excluded.

SQL files are intentionally excluded from the checkpoint system because they can often contain very large amounts of data - such as database dumps, migration scripts with seed data, or extensive DDL statements. Including these files in checkpoints could significantly impact performance and memory usage, especially for users working with production-scale databases.

The checkpoint system is optimized for tracking changes in source code files that developers actively edit. While I understand that you may have SQL scripts you'd like to track, the current design prioritizes system performance and stability.

As a workaround, you could:

1. Keep your SQL scripts in a version control system (like Git) for tracking changes
2. For smaller SQL files that you actively develop, you could temporarily rename them to a different extension (as you discovered with .py) when you need checkpoint functionality
3. Consider splitting large SQL files into smaller, more manageable pieces

I appreciate you bringing this to our attention, and I hope this explanation helps clarify the design decision. If you have SQL files that are consistently small and would benefit from checkpoint tracking, please feel free to share more details about your use case - we're always looking to improve the developer experience!