Rug Munch Media LLC
223c15b2f5
- Split scraping endpoints from api.py into routers/scraping.py - Split template endpoints into routers/templates.py with POST /v1/templates/batch - Introduce deps.py for shared singleton instances - Add e2e tests for scraping, templates, auth, and Apify schema (23 new tests) - Add reusable Apify actor input/output schema builder (apify_schema.py) - Add PRY_API_KEY startup warning and deploy key to Talos /srv/pry/.env - Register /v1/templates/batch as paid x402 operation
72 lines
2.7 KiB
Python
72 lines
2.7 KiB
Python
# SPDX-License-Identifier: MIT
|
|
# Copyright (c) 2026 Rug Munch Media LLC
|
|
#
|
|
# Part of Pry — https://git.rugmunch.io/RugMunchMedia/pryscraper
|
|
# Licensed under MIT.
|
|
"""Tests for the Apify schema builder."""
|
|
|
|
from apify_schema import ApifySchemaBuilder
|
|
|
|
|
|
def test_input_schema_basic() -> None:
|
|
schema = (
|
|
ApifySchemaBuilder("Product Scraper", "Scrape product details")
|
|
.add_string("url", "Start URL", description="The URL to scrape", required=True)
|
|
.add_integer("maxResults", "Max Results", default=10, minimum=1, maximum=100)
|
|
.add_enum("format", "Output Format", ["json", "csv", "markdown"], default="json")
|
|
.build_input_schema()
|
|
)
|
|
|
|
assert schema["title"] == "Product Scraper"
|
|
assert schema["description"] == "Scrape product details"
|
|
assert schema["type"] == "object"
|
|
assert schema["schemaVersion"] == 1
|
|
assert schema["required"] == ["url"]
|
|
assert schema["properties"]["url"]["type"] == "string"
|
|
assert schema["properties"]["maxResults"]["default"] == 10
|
|
assert schema["properties"]["format"]["enum"] == ["json", "csv", "markdown"]
|
|
|
|
|
|
def test_output_schema_includes_required() -> None:
|
|
schema = (
|
|
ApifySchemaBuilder("Product Output")
|
|
.add_string("title", "Title", required=True)
|
|
.add_number("price", "Price", minimum=0)
|
|
.add_boolean("inStock", "In Stock", default=True)
|
|
.build_output_schema()
|
|
)
|
|
|
|
assert schema["properties"]["title"]["type"] == "string"
|
|
assert schema["properties"]["price"]["minimum"] == 0
|
|
assert schema["properties"]["inStock"]["default"] is True
|
|
assert schema["required"] == ["title"]
|
|
|
|
|
|
def test_array_and_object_fields() -> None:
|
|
schema = (
|
|
ApifySchemaBuilder("Complex Actor")
|
|
.add_array("urls", "URLs", {"type": "string"}, required=True, min_items=1, max_items=50)
|
|
.add_object(
|
|
"proxy",
|
|
"Proxy Config",
|
|
{"host": {"type": "string"}, "port": {"type": "integer"}},
|
|
nested_required=["host"],
|
|
)
|
|
.build_input_schema()
|
|
)
|
|
|
|
assert schema["properties"]["urls"]["type"] == "array"
|
|
assert schema["properties"]["urls"]["minItems"] == 1
|
|
assert schema["properties"]["urls"]["items"]["type"] == "string"
|
|
assert schema["properties"]["proxy"]["type"] == "object"
|
|
assert schema["properties"]["proxy"]["required"] == ["host"]
|
|
|
|
|
|
def test_nullable_field() -> None:
|
|
schema = ApifySchemaBuilder().add_string("optional", "Optional", nullable=True).build_input_schema()
|
|
assert schema["properties"]["optional"]["type"] == ["string", "null"]
|
|
|
|
|
|
def test_prefab() -> None:
|
|
schema = ApifySchemaBuilder().set_prefab("START_URLS").add_string("url", "URL", required=True).build_input_schema()
|
|
assert schema["prefab"] == "START_URLS"
|