Dev

[BAIKEMARK]

This pull request refactors the file hashing and database update logic to improve thread safety and maintainability. The main change is the extraction of database update operations into a dedicated, thread-safe function, and the adjustment of timeouts and thread pool configuration for better performance.

Thread safety and code organization:

• Added a new function update_hash_in_db to handle writing hash results to the database in a thread-safe manner, isolating database operations from the hashing worker logic. (utils.py)
• Refactored the hashing workflow to call update_hash_in_db from within the thread pool, removing direct database access from worker threads and simplifying error handling. (utils.py)

Performance and reliability improvements:

• Increased the single file hash timeout from 60 to 90 seconds to reduce the likelihood of hash operations timing out on large files. (utils.py)
• Adjusted the thread pool configuration to use all available CPUs instead of half, potentially improving hashing throughput. (utils.py)

[Copilot]

Pull Request Overview

This pull request refactors the file hashing and database update logic to improve thread safety and maintainability by extracting database operations into a dedicated function and optimizing performance configurations.

• Extracted database update operations into a thread-safe update_hash_in_db function
• Increased single file hash timeout from 60 to 90 seconds
• Changed thread pool to use all available CPUs instead of half

---

_{Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.}

[Copilot]

Using all available CPUs for hashing may starve other system processes. Consider using min(os.cpu_count() or 4, 8) to cap the maximum number of workers, especially since the previous implementation used half the CPUs for a reason.

Suggested change

	with ThreadPoolExecutor(max_workers=(os.cpu_count() or 4)) as executor:
	with ThreadPoolExecutor(max_workers=min(os.cpu_count() or 4, 8)) as executor:

[Copilot]

The function returns None on invalid input but True/False for success/failure cases. This inconsistent return type could confuse callers. Consider returning False for invalid input to maintain consistent boolean return values.

Suggested change

	return
	return False

[Copilot]

Mutating the res dictionary in-place could have side effects if the same dictionary is used elsewhere. Consider creating a copy or passing model_type as a separate parameter to update_hash_in_db.

Suggested change

	res['model_type'] = model_type
	if update_hash_in_db(res):
	new_res = {**res, 'model_type': model_type}
	if update_hash_in_db(new_res):