expectfs

Declarative validation for things that are not data models.

expectfs lets you define clear, readable expectations about files, directories, and system state — without schemas, test frameworks, or boilerplate.

This is for validating the world, not validating Python objects.

---

Installation

pip install expectfs

---

Quick Example

from expect import expect, validate

expect.file("metrics.json") \
    .is_json() \
    .size_gt(100)

result = validate()

result.print()

If any expectation fails, a clear, actionable error message will be displayed after running validation.

---

Why expect?

Most validation tools focus on:

• JSON schemas
• Data models
• API payloads
• Function inputs

But engineers constantly need to validate things like:

• “Does this file exist?”
• “Is this directory populated?”
• “Is this JSON valid?”
• “Did my pipeline actually produce output?”

expectfs exists for that gap.

---

Key Features

• Declarative, chainable expectations
• Automatic dependency resolution (rules run in the correct order)
• Rule caching (each rule runs at most once)
• Clear failure messages

---

How it Works (Conceptually)

expect.file("output.json").is_json().size_gt(1024)

• is_json automatically ensures the file exists
• size_gt reuses prior results instead of re-running checks
• Dependencies are resolved recursively

Users don’t need to think about ordering.

---

Supported Targets

Currently supported:

• Files
• Directories

Planned:

• Globs
• Environment variables
• Command outputs

---

Status

This project is in alpha.

• APIs may evolve
• Backwards compatibility is not yet guaranteed
• Feedback and contributions are welcome

---

License

MIT