Data Layer

The data layer covers everything needed to chain requests, drive tests with external data, and switch environments without rewriting your action JSON.

VariableStore

A thread-safe key/value singleton (variable_store).

from je_api_testka import variable_store

variable_store.set("token", "abc")
variable_store.get("token")        # 'abc'
variable_store.snapshot()           # {'token': 'abc'}
variable_store.clear()

Templating

render_template walks dicts, lists, and strings recursively, replacing {{name}} placeholders with values from the store.

from je_api_testka import render_template, variable_store

variable_store.set("base", "https://api.invalid")
render_template({"url": "{{base}}/v1"})
# -> {'url': 'https://api.invalid/v1'}

Pass strict=True to raise when a placeholder has no matching variable.

Chaining requests

extract_and_store pulls a value from a response payload via a dotted path and persists it for the next request.

from je_api_testka import extract_and_store, variable_store

response_payload = {"data": {"user": {"id": 7}}}
extract_and_store(response_payload, "data.user.id", "user_id")
variable_store.get("user_id")  # 7

Data-driven tests

from je_api_testka import iter_csv_rows, iter_json_rows

for row in iter_csv_rows("data/users.csv"):
    ...
for row in iter_json_rows("data/users.json"):
    ...

Fake data helpers

fake_uuid, fake_email, and fake_word are stdlib-only helpers good enough for boundary cases. For richer data sets, install Faker separately.

Environment profiles

load_env_profile reads a JSON file and pushes every key/value pair into the variable store, so the same actions can target dev / staging / prod by swapping profiles.

from je_api_testka import load_env_profile

load_env_profile("envs/dev.json")

Executor commands

AT_set_variable
AT_get_variable
AT_clear_variables
AT_extract_and_store
AT_render_template
AT_fake_uuid
AT_fake_email
AT_fake_word
AT_load_env_profile