[Bug]: In-memory cache in data_loader.py is declared but never used — every request re-reads projects.json from disk

[Daksh7785]

Describe the bug

utils/data_loader.py declares a module-level cache variable and a clear_cache() function, but load_all_projects() never reads from or writes to the cache:

_projects_cache = None  # declared but never read

def load_all_projects():
    with open(DATA_FILE, "r", encoding="utf-8") as f:
        return json.load(f)  # ← no cache check, file read every call

This means every call to /api/recommend, every project detail page load (find_project_by_id), and every homepage render (get_project_stats) triggers a full disk read and JSON parse. clear_cache() resets a variable that is never read — it is a dead function.

Expected behavior

load_all_projects() should check the cache first and only read the file on a miss.

Suggested fix

def load_all_projects():
    global _projects_cache
    if _projects_cache is None:
        with open(DATA_FILE, "r", encoding="utf-8") as f:
            _projects_cache = json.load(f)
    return _projects_cache

Files affected

• utils/data_loader.py

[Daksh7785]

/assign

[Antra1705]

The fix is already identified — add global _projects_cache and a None check at the top of load_all_projects() in utils/data_loader.py so the file is read once and cached in memory on subsequent calls. The clear_cache() function already exists and correctly resets to None, so it will work as intended after this fix.
Single-function change, one file. I work with Python regularly and this is a straightforward cache-miss pattern fix.
Contributing under GSSoC '26. Could you please assign this to @Antra1705?

[samanvithashivapuram]

Hi @komalharshita,

I see the issue clearly—load_all_projects() is missing the None check and isn't utilizing the _projects_cache global variable, causing redundant disk reads on every API and page request.

If this isn't assigned yet, I have the fix ready to go and would love to open a PR for this under GSSoC '26. Could you please assign this to me?