File Deletion Optimization and UUID Implementation

[pulgamecanica]

💡 Description

The current implementation deletes files twice:

1. Before using unzip.
2. After files are uploaded to S3.

This approach may introduce inefficiencies, especially if the same cleanup operation is repeated. Additionally, using temporary file names without unique identifiers could lead to file name conflicts if multiple users upload files with the same name concurrently.

---

📝 Proposal

• Implement a UUID-based naming system for all temporary files and directories. This ensures each user's file is uniquely identified, preventing conflicts.
• Perform file cleanup only once, after all operations (e.g., unzip, S3 upload) are completed successfully or after any failure.

---

🔍 Discussion Points

• UUID Naming: Should we use a UUID library, or is there a simpler approach?
• Centralized Cleanup: Should cleanup be part of a finally block or managed by a dedicated function?

---

✍ Tasks (If Accepted)

• Replace direct use of file.filename with a UUID-based file name.
• Consolidate cleanup logic to ensure files and directories are deleted only once, at the end of processing.
• Test the solution with concurrent file uploads to verify no conflicts occur.

---

🎯 Acceptance Criteria

• Files are uniquely named using UUIDs, preventing conflicts.
• Cleanup occurs only once, ensuring no leftover files or directories.
• File handling logic is optimized for clarity and maintainability.

[kgoedecke]

I think you can just use uuid.uuid4() from the default python package import uuid.

[pulgamecanica]

Yeah I agree, I'll implement this.