python 42 lines · 8 steps

Parsing fixed-width records in Python

A generator turns fixed-width text lines into typed Transaction records, slicing by column offsets and normalizing each field.

Explained by highlit

Intermediate parsing generators dataclasses fixed-width data-cleaning

1import datetime
2from dataclasses import dataclass
3 
4 
5@dataclass
6class Transaction:
7    account_id: str
8    posted_on: datetime.date
9    description: str
10    amount: float
11    status: str
12 
13 
14_FIELDS = [
15    ("account_id", 0, 12),
16    ("posted_on", 12, 20),
17    ("description", 20, 60),
18    ("amount", 60, 74),
19    ("status", 74, 76),
20]
21 
22_STATUS_CODES = {"PO": "posted", "PN": "pending", "RV": "reversed"}
23 
24 
25def parse_report(lines):
26    for lineno, raw in enumerate(lines, start=1):
27        line = raw.rstrip("\n")
28        if not line.strip() or line.startswith("#"):
29            continue
30 
31        fields = {name: line[start:end].strip() for name, start, end in _FIELDS}
32 
33        try:
34            yield Transaction(
35                account_id=fields["account_id"],
36                posted_on=datetime.datetime.strptime(fields["posted_on"], "%Y%m%d").date(),
37                description=fields["description"],
38                amount=int(fields["amount"]) / 100.0,
39                status=_STATUS_CODES.get(fields["status"], "unknown"),
40            )
41        except ValueError as exc:
42            raise ValueError(f"malformed record on line {lineno}: {exc}") from exc

01 / 01

STEP 01

Walkthrough

Space play ←→ step click any line

Three takeaways

1A table of (name, start, end) offsets keeps fixed-width parsing declarative and easy to adjust.
2Yielding records lazily lets callers stream through large reports without loading everything into memory.
3Wrapping conversions in a try/except that adds the line number turns cryptic errors into actionable ones.

Related explainers

rust

use std::collections::HashMap;
 
#[derive(Debug)]
pub struct RequestHead {

Parsing an HTTP request head in Rust

parsing error-handling iterators

Intermediate 9 steps

php

<?php
 
final class MarkdownParser
{

Building a small Markdown-to-HTML parser in PHP

state-machine parsing closures

Intermediate 10 steps

python

from itertools import product
from dataclasses import dataclass
from decimal import Decimal

Generating product variants with itertools.product

cartesian-product dataclass decimal

Intermediate 7 steps

python

from typing import Callable, Dict, Type
 
 
class PluginRegistry:

A decorator-based plugin registry in Python

decorators registry pattern factory

Intermediate 9 steps

package humanize
 
import (
	"fmt"

Parsing human-readable byte sizes in Go

parsing regex lookup-table

Intermediate 8 steps

python

from fastapi import APIRouter, BackgroundTasks, Depends, HTTPException, status
from pydantic import BaseModel, EmailStr
from sqlalchemy.orm import Session