Server IP : 85.214.239.14 / Your IP : 3.145.186.28 Web Server : Apache/2.4.62 (Debian) System : Linux h2886529.stratoserver.net 4.9.0 #1 SMP Tue Jan 9 19:45:01 MSK 2024 x86_64 User : www-data ( 33) PHP Version : 7.4.18 Disable Function : pcntl_alarm,pcntl_fork,pcntl_waitpid,pcntl_wait,pcntl_wifexited,pcntl_wifstopped,pcntl_wifsignaled,pcntl_wifcontinued,pcntl_wexitstatus,pcntl_wtermsig,pcntl_wstopsig,pcntl_signal,pcntl_signal_get_handler,pcntl_signal_dispatch,pcntl_get_last_error,pcntl_strerror,pcntl_sigprocmask,pcntl_sigwaitinfo,pcntl_sigtimedwait,pcntl_exec,pcntl_getpriority,pcntl_setpriority,pcntl_async_signals,pcntl_unshare, MySQL : OFF | cURL : OFF | WGET : ON | Perl : ON | Python : ON | Sudo : ON | Pkexec : OFF Directory : /usr/lib/python3/dist-packages/requests_toolbelt/threaded/ |
Upload File : |
""" This module provides the API for ``requests_toolbelt.threaded``. The module provides a clean and simple API for making requests via a thread pool. The thread pool will use sessions for increased performance. A simple use-case is: .. code-block:: python from requests_toolbelt import threaded urls_to_get = [{ 'url': 'https://api.github.com/users/sigmavirus24', 'method': 'GET', }, { 'url': 'https://api.github.com/repos/requests/toolbelt', 'method': 'GET', }, { 'url': 'https://google.com', 'method': 'GET', }] responses, errors = threaded.map(urls_to_get) By default, the threaded submodule will detect the number of CPUs your computer has and use that if no other number of processes is selected. To change this, always use the keyword argument ``num_processes``. Using the above example, we would expand it like so: .. code-block:: python responses, errors = threaded.map(urls_to_get, num_processes=10) You can also customize how a :class:`requests.Session` is initialized by creating a callback function: .. code-block:: python from requests_toolbelt import user_agent def initialize_session(session): session.headers['User-Agent'] = user_agent('my-scraper', '0.1') session.headers['Accept'] = 'application/json' responses, errors = threaded.map(urls_to_get, initializer=initialize_session) .. autofunction:: requests_toolbelt.threaded.map Inspiration is blatantly drawn from the standard library's multiprocessing library. See the following references: - multiprocessing's `pool source`_ - map and map_async `inspiration`_ .. _pool source: https://hg.python.org/cpython/file/8ef4f75a8018/Lib/multiprocessing/pool.py .. _inspiration: https://hg.python.org/cpython/file/8ef4f75a8018/Lib/multiprocessing/pool.py#l340 """ from . import pool from .._compat import queue def map(requests, **kwargs): r"""Simple interface to the threaded Pool object. This function takes a list of dictionaries representing requests to make using Sessions in threads and returns a tuple where the first item is a generator of successful responses and the second is a generator of exceptions. :param list requests: Collection of dictionaries representing requests to make with the Pool object. :param \*\*kwargs: Keyword arguments that are passed to the :class:`~requests_toolbelt.threaded.pool.Pool` object. :returns: Tuple of responses and exceptions from the pool :rtype: (:class:`~requests_toolbelt.threaded.pool.ThreadResponse`, :class:`~requests_toolbelt.threaded.pool.ThreadException`) """ if not (requests and all(isinstance(r, dict) for r in requests)): raise ValueError('map expects a list of dictionaries.') # Build our queue of requests job_queue = queue.Queue() for request in requests: job_queue.put(request) # Ensure the user doesn't try to pass their own job_queue kwargs['job_queue'] = job_queue threadpool = pool.Pool(**kwargs) threadpool.join_all() return threadpool.responses(), threadpool.exceptions()