Design an In-Memory Cloud Storage System (Incremental Levels)

Build an in-memory cloud storage system across four incremental levels. Model the system as a class that fulfills the CloudStorage interface described below, along with any internal data structures you need.

---

Level 1: Basic file operations

Implement these methods:

1. add_file(name: str, size: int) -> bool

   • If name is already present, return False. Otherwise store the file with its size and return True.

2. copy_file(name_from: str, name_to: str) -> bool

   • Return False when the source file is missing or the destination name is already taken. Otherwise duplicate the file (copy only its size) and return True.

3. get_file_size(name: str) -> int | None

   • Return the file's size when it exists; otherwise return None.

Constraints:

• Every size is strictly positive.
• Up to 100,000 files may be stored. Strive for roughly O(1) or O(log n) time per operation.

---

Level 2: Prefix/suffix search with ordering

Add:

4. find_file(prefix: str, suffix: str) -> list[str]
   • Return the names of every file that starts with prefix and ends with suffix.
   • Each entry in the returned list must be formatted as "{name}({size})".
   • Sort the results by (1) descending size, and then (2) ascending lexicographic order of the file name.

---

Level 3: Multi-user quotas and owner-charged copies

Extend the model so that every file belongs to a user. Each user has a capacity and a current usage. The built-in user "admin" has unlimited capacity.

New and changed methods:

5. add_user(user_id: str, capacity: int) -> bool

6. add_file_by(user_id: str, name: str, size: int) -> int | None

   • The operation fails (returns None) when the user does not exist, the file name is already taken, or adding the file would exceed the user's capacity.
   • On success, create the file owned by user_id, update that user's usage, and return their remaining capacity.

7. add_file(name: str, size: int) -> bool

   • Behaves as if "admin" is adding the file.

8. copy_file(name_from: str, name_to: str) -> bool

   • The destination file is owned by the same user who owns the source file.
   • The copy increases that owner's usage.
   • Fails when the source is missing, the destination name already exists, or the owner's quota would be exceeded.

9. update_capacity(user_id: str, capacity: int) -> int | None

   • Set the user's capacity to the new value. If the user's current usage now exceeds capacity, delete files until usage fits within the new limit.
   • Deletion order: remove the largest files first; break ties by lexicographic order of the file name.
   • Return the number of files deleted.

---

Level 4: Compress / decompress (quota-aware)

Only the file's owner may perform these operations.

10. compress_file(user_id: str, name: str) -> int | None

    • Replace the file named name with name + ".COMPRESSED".
    • The new size is floor(old_size / 2).
    • Adjust the owner's usage accordingly and return their remaining capacity.

11. decompress_file(user_id: str, name: str) -> int | None

    • name must exist and be owned by user_id.
    • Strip the ".COMPRESSED" suffix to obtain original_name.
    • Fail if original_name already exists.
    • The new size is current_size * 2; fail if the owner's quota would be exceeded.
    • Replace the compressed file with the decompressed version and return the owner's remaining capacity.

Example

Input:

add_file /a 10
add_file /a 10
copy_file /a /b
get_file_size /b

Output:

True
False
True
10