Package Manager

Overview

The Package Manager dynamically loads external Python packages into the executor at runtime. This allows you to extend the executor’s capabilities without modifying the source code.

When a package is added, all its public functions and classes are extracted and registered in the executor’s event dictionary with the naming convention {package}_{function}.

Usage via Action Executor

from je_web_runner import execute_action

actions = [
    # Load the 'time' package into executor
    ["WR_add_package_to_executor", {"package": "time"}],

    # Now you can use time.sleep as "time_sleep"
    ["time_sleep", [2]],
]

execute_action(actions)

Direct API Usage

from je_web_runner.utils.package_manager.package_manager_class import package_manager

# Check if a package exists and import it
module = package_manager.check_package("os")

# Add all functions from a package to the executor
package_manager.add_package_to_executor("math")

# Add to callback executor instead
package_manager.add_package_to_callback_executor("time")

Methods

Method	Parameters	Description
`check_package()`	`package: str`	Check and import a package, returns the module or None
`add_package_to_executor()`	`package: str`	Add package members to the Executor’s event_dict
`add_package_to_callback_executor()`	`package: str`	Add package members to the CallbackExecutor’s event_dict
`get_member()`	`package, predicate, target`	Extract and add specific members from a package
`add_package_to_target()`	`package: str, target`	Add functions/classes to a target executor

Naming Convention

When a package is added, functions are registered with the prefix {package_name}_.

For example, adding the time package registers:

time_sleep -> time.sleep
time_time -> time.time
time_monotonic -> time.monotonic
etc.

Package Caching

The installed_package_dict attribute caches imported packages to avoid re-importing them. If a package has already been loaded, subsequent calls to check_package() return the cached module.