Update Only That Item

The preview confirms that EL-300 is the projector and its current period is 2.5. The next call should change that one value to 8 while leaving every other item alone:

change_loan_days(connection, "EL-300", 8)

The arguments after the open connection are the identifying asset tag and the new period. Open catalog_changes.py and find the new function:

def change_loan_days(connection, asset_tag, loan_days):
    statement = """
        UPDATE items
        SET loan_days = ?
        WHERE asset_tag = ?
    """
    connection.execute(statement, (asset_tag, loan_days))

UPDATE items names the table to change. SET loan_days = ? gives the first placeholder the new value. WHERE asset_tag = ? makes the second placeholder identify the intended row. The changing statement carries its own precise scope, even though the caller previewed the row first.

The starter tuple has the two Python values reversed. Compare positions before editing:

Notice that the earlier preview and this update are two separate statements. Seeing the projector in the preview helped you decide what to do, but it does not limit a later update. The UPDATE must carry its own WHERE asset_tag = ?. When the call uses "EL-300", the database receives the statement unchanged and receives (8, "EL-300") as its values: 8 answers “what should change?” and "EL-300" answers “which row?”

That distinction becomes especially clear for a missing tag. (8, "NO-404") is still a valid pair of bound values, but the WHERE condition finds nothing, so no row changes. The neighboring speaker does not become a fallback target. A precise condition means “this row if it exists,” not “some similar row.”

Change the final tuple to (loan_days, asset_tag), then press Run.

before EL-300
(5, 'EL-300', 'Projector', 'electronics', 2.5)
change_loan_days(connection, "EL-300", 8)
None
after EL-300
(5, 'EL-300', 'Projector', 'electronics', 8)

The function still returns None for now. Read the before-and-after rows as the evidence: only the last value changed. Run also tries NO-404, which matches no row, and displays EL-400 afterward as an unchanged neighbor.

The table’s existing positive-period rule still applies. This function does not replace that rule or catch its error. It safely binds the proposed value; the database still decides whether the value satisfies the schema.

Submit uses unseen tags, a missing tag, apostrophes, SQL-looking text, and different positive whole and fractional periods. Keep both changing values outside the statement, in placeholder order. Do not filter by a non-unique name or category, and do not open another connection.

A safe change needs both halves you can now see: the new value in SET, and the unique identifying value in the same statement’s WHERE.