Yesterday you read the CSV file. You got lines. But look at one:

'Alice Chen,1250.50,West,confirmed'  # readlines() includes the newline

You have 7 lines of text. But each line is still just one long string. 'Alice Chen,1250.50,West,confirmed' isn't four values — it's one. You need to split it.

The Problem: Strings vs. Data

You have text that looks like data. But in your code, it's just a string. You can't ask: "What's the amount?" You'd have to do messy string slicing. We need structure.

# This doesn't work:
line = 'Alice Chen,1250.50,West,confirmed'
amount = line[11:19]  # ❌ fragile, error-prone

# This is what we want:
record = {'name': 'Alice Chen', 'amount': '1250.50', 'region': 'West', 'status': 'confirmed'}
amount = record['amount']  # ✓ clear, safe

Splitting with .split(',')

The solution is the .split() method. It breaks a string on a delimiter.

line = 'Alice Chen,1250.50,West,confirmed'
fields = line.split(',')
print(fields)
# ['Alice Chen', '1250.50', 'West', 'confirmed']

Now we have a list. But there's a problem: the header row.

Maya(thinking): The first line is different. It's the header. 'name,amount,region,status'. So I read that separately?

Exactly. The first line tells you what each field means. The rest are data.

Header-Driven Parsing

Here's the pattern:

1. Read all lines from the file
2. The first line becomes your headers (column names)
3. Each remaining line becomes a record (a dict mapping header → value)

def parse_csv(filepath):
    with open(filepath) as f:
        lines = f.readlines()
    
    if not lines:
        return []
    
    # First line is headers
    header_line = lines[0].strip()  # strip() removes the trailing newline
    headers = header_line.split(',')
    
    records = []
    for line in lines[1:]:  # Skip the header, process the rest
        line = line.strip()  # Clean up whitespace
        if not line:  # Skip empty lines
            continue
        values = line.split(',')
        record = dict(zip(headers, values))
        records.append(record)
    
    return records

zip() pairs up two lists element-by-element:

headers = ['name', 'amount', 'region', 'status']
values = ['Alice Chen', '1250.50', 'West', 'confirmed']

for h, v in zip(headers, values):
    print(f"{h}: {v}")
# name: Alice Chen
# amount: 1250.50
# region: West
# status: confirmed

# zip() creates tuples, dict() converts them to key-value pairs
record = dict(zip(headers, values))
# {'name': 'Alice Chen', 'amount': '1250.50', 'region': 'West', 'status': 'confirmed'}

The Whitespace Problem

Real CSV files are messy. Look at the test data:

name,amount,region,status
Alice Chen,1250.50,West,confirmed
Bob Kumar,340.50,East,pending
...Eve Williams,520.00,North,confirmed

Eve's name has a leading space. If you don't strip it, you'll get ' Eve Williams' (with the space). That breaks lookups.

Maya(excited): So I strip each field?

Yes! After you split, strip each value:

values = [v.strip() for v in line.split(',')]
# ['Eve Williams']  ✓ space is gone

Now the dict has clean data.

Your Challenge

Write parse_csv(filepath) that:

1. Reads the file (you can reuse read_csv_lines() from yesterday, or inline it)
2. Uses the first line as headers
3. For each remaining line:
   • Strip the line
   • Skip empty lines
   • Split by comma
   • Strip each field
   • Zip with headers and create a dict
4. Return a list of dicts

Test it with sales.csv. You should get 6 records (header + empty line skipped). Each record is a dict. You can access record['name'], record['amount'], etc.

Next time: String cleaning. Stripping spaces is just the start. Next we handle case normalization, quoted fields, and other CSV gotchas that make the real world messy.