simple-state-tracker 0.3.0

A minimal, type-safe, file-backed state tracker using Pydantic models.

simple-state-tracker

A minimal, file-backed, type-safe state tracker using Pydantic models — ideal for scripts, scrapers, ETL pipelines, and resumable workflows.

Features

• 🧠 Strong typing with pydantic models
• 💾 Transparent JSON persistence
• 🧰 Simple .get(), .set(), .edit() API
• ✅ Schema validation and strict key control
• 🪶 Lightweight — no databases, no DAGs, no dependencies beyond pydantic

Installation

pip install simple-state-tracker

---

Example

## Quick Start

```python
from simple_state_tracker import SimpleStateTracker, KeyModel, DataModel

# First, define a key model
class ScrapeKey(KeyModel):
    county: str
    municipality: str
    year: int

# Next, define a data model
class ScrapeState(DataModel):
    scraped: bool = False
    scrape_error: str | None = None
    processed: bool = False
    process_error: str | None = None

# Initialize the tracker
tracker = SimpleStateTracker(ScrapeKey, ScrapeState, path="tracker.json")

key = ScrapeKey(county="DAUPHIN", municipality="HARRISBURG", year=2022)

with tracker.edit(key) as state:
    state.scraped = True
    state.scrape_error = None

tracker.save()

---

Use Cases

• Track which URLs, files, or locations have been processed
• Resume scraping jobs or ETL pipelines
• Store structured state across CLI or batch job invocations
• Replace ad-hoc JSON or YAML logs with something type-safe and self-validating

Non-Use Cases

Do not use this if

• You are looking for optimization in storage or performance. This is meant to be easy to use and simple to understand, not the fastest or most efficient

API Overview

SimpleStateTracker(key_model, data_model, path)

Creates a new tracker instance.

• get(key) → returns the data model (or None)
• set(key, value) → manually sets a value
• edit(key) → yields a context-managed editable state
• save() → writes the cache to disk
• load() → reads from disk
• all() → returns a shallow copy of all state