Python CSV imports: parse quoted fields before validating rows

Operation contract

The import reads a two-column header and one row whose note contains a comma. DictReader produces one amount and one note, then the application validates the amount as a positive integer. Splitting each line at commas would produce the wrong number of fields for this record.

Failure and ownership boundary

The program uses an in-memory text stream, not an uploaded file. A deployed import needs byte, row, field and retained-result bounds; accepting CSV does not provide them automatically. It also needs a duplicate-header and unexpected-column policy. Python input exercise: accept an explicit integer grammar and Python pathlib files: specify encoding and close the resource owner follow parsing.

Working program

import csv
import io

source = io.StringIO('amount_minor,note\n125,"review, accepted"\n', newline="")
reader = csv.DictReader(source)
if reader.fieldnames != ["amount_minor", "note"]:
    raise ValueError("expected receipt header")
for row in reader:
    if set(row) != {"amount_minor", "note"} or None in row.values():
        raise ValueError("complete row required")
    amount = int(row["amount_minor"])
    if amount <= 0:
        raise ValueError("positive amount required")
    print(amount)
    print(row["note"])

Output

125
review, accepted

Costs and limits

This one-row fixture retains the entire short text. Streaming rows can avoid retaining all parsed rows, but a very large individual field still needs a limit.

Common Mistakes

• Do not split CSV at commas manually.
• Check header and row shape before using fields.

Connected lessons

Python input exercise: accept an explicit integer grammar, Python pathlib files: specify encoding and close the resource owner, Python JSON validation: reject duplicate members and non-integer amounts.

Apply this boundary

Python CSV ingestion: cap bytes, rows and fields before publication, Pandas nullable integers: missing amounts are not zero.

Check the next state boundary

Python JSON Lines: cap line bytes and validate the whole batch before returning it.