Initial commit: add compliance_checks table, per-check metadata on assets, and compliance audit trail

app/services/csv_service.py

@@ -0,0 +1,60 @@
+import csv
+import io
+
+
+# Maps our field names to possible CSV column header aliases (case-insensitive)
+FIELD_ALIASES = {
+    'windows_id': ['windows_id', 'employeeid', 'employee_id', 'wid', 'id', 'user_id', 'samaccountname'],
+    'first_name': ['first_name', 'firstname', 'givenname', 'given_name', 'prenom'],
+    'last_name':  ['last_name', 'lastname', 'surname', 'sn', 'family_name', 'nom'],
+    'email':      ['email', 'mail', 'email_address', 'emailaddress'],
+    'department': ['department', 'dept', 'division'],
+    'job_title':  ['job_title', 'title', 'position', 'jobtitle', 'job_position'],
+    'phone':      ['phone', 'telephone', 'mobile', 'phonenumber', 'telephonenumber'],
+    'location':   ['location', 'office', 'site', 'physicaldeliveryofficename'],
+}
+
+
+def parse_users_csv(file_stream):
+    """
+    Parse a CSV file stream and return (users_list, errors_list).
+
+    Accepts BOM-prefixed UTF-8 or plain UTF-8. Column headers are
+    matched case-insensitively against FIELD_ALIASES.
+    """
+    try:
+        content = file_stream.read().decode('utf-8-sig')
+    except (UnicodeDecodeError, AttributeError):
+        return [], ['Could not decode file. Please use UTF-8 encoding.']
+
+    reader = csv.DictReader(io.StringIO(content))
+
+    if not reader.fieldnames:
+        return [], ['CSV file is empty or has no header row.']
+
+    # Build a lookup: normalised header -> our field name
+    norm_headers = {h.lower().strip().replace(' ', '_'): h for h in reader.fieldnames}
+    col_map = {}  # our_field -> actual_csv_header
+    for field, aliases in FIELD_ALIASES.items():
+        for alias in aliases:
+            if alias in norm_headers:
+                col_map[field] = norm_headers[alias]
+                break
+
+    errors = []
+    users = []
+
+    for row_num, row in enumerate(reader, start=2):
+        user = {}
+        for field in FIELD_ALIASES:
+            csv_col = col_map.get(field)
+            user[field] = (row.get(csv_col, '') or '').strip() if csv_col else ''
+
+        wid = user.get('windows_id', '').strip()
+        if not wid:
+            errors.append(f'Row {row_num}: missing windows_id — skipped.')
+            continue
+
+        users.append(user)
+
+    return users, errors