[](https://github.com/Basdbruijne/autothread/actions/workflows/main.yml)
# Autothread
Parallelization made easy.
Autothread contains a collection of decorators that make it as easy as possible to add
threading or multiprocessing to your projects. Autothread has two types of decorators: blocking and non-blocking.
## Non-blocking
Autothreads non-blocking decorators are the easiest way to add threading/multiprocessing to your
project. You just need to add a single decorator, which changes your function to calculate in
the background instead of blocking the script.
```python
import autothread
import time
from time import sleep as heavyworkload
@autothread.async_threaded() # <-- This is all you need to add
def example(x, y) -> int:
heavyworkload(1)
return x*y
start = time.time()
results = []
for i in range(5):
results.append(example(i, 10)) # the thread is started
print(results) # autothread waits for the thread to end and gives you the result
print("Time expired: ", time.time()-start)
>>> [0, 10, 20, 30, 40]
Time expired: 1.002363681793213
```
`autothread.async_processed` works in the same way but uses multiprocessing instead of threading.
More info can be found in the [non-blocking README](https://github.com/Basdbruijne/autothread/blob/main/docs/README_non_blocking.md).
## Blocking
The blocking decorators of autothread change the function slightly, but give you more control
over when the function is executed:
```python
import autothread
import time
from time import sleep as heavyworkload
@autothread.multithreaded() # <-- This is all you need to add
def example(x: int, y: int):
heavyworkload(1)
return x*y
@autothread.multiprocessed() # <-- Or to use multiprocessing
def example2(x: int, y: int):
heavyworkload(1)
return x*y
```
Now, instead of integers, your function can take lists of integers. The function will
be repeated or each item in your list on a separate thread/process:
```python3
start = time.time()
result = example([1, 2, 3, 4, 5], 10)
print(result)
print("Time expired: ", time.time()-start)
>>> [10, 20, 30, 40, 50]
Time expired: 1.0041766166687012
```
More info can be found in the [blocking README](https://github.com/Basdbruijne/autothread/blob/main/docs/README_blocking.md).
## Installing
You can install autothread using:
```
pip install autothread
```
Or by cloning the source:
```
git clone https://github.com/Basdbruijne/autothread.git
cd autothread
pip install -e .
```
## Known issues
None at the moment, please open a bug if you run into an issue.
Raw data
{
"_id": null,
"home_page": "https://github.com/Basdbruijne/autothread",
"name": "autothread",
"maintainer": "",
"docs_url": null,
"requires_python": "",
"maintainer_email": "",
"keywords": "multithreading,multiprocessing,decorator",
"author": "Bas de Bruijne",
"author_email": "basdbruijne@gmail.com",
"download_url": "https://files.pythonhosted.org/packages/4b/ef/9927258f0224c0f20561ec15b3b86d2373cb28cd22f664ba712a393817b1/autothread-0.0.10.tar.gz",
"platform": null,
"description": "[](https://github.com/Basdbruijne/autothread/actions/workflows/main.yml)\n\n# Autothread\nParallelization made easy.\n\nAutothread contains a collection of decorators that make it as easy as possible to add\nthreading or multiprocessing to your projects. Autothread has two types of decorators: blocking and non-blocking.\n\n## Non-blocking\nAutothreads non-blocking decorators are the easiest way to add threading/multiprocessing to your\nproject. You just need to add a single decorator, which changes your function to calculate in\nthe background instead of blocking the script.\n\n```python\nimport autothread\nimport time\nfrom time import sleep as heavyworkload\n\n@autothread.async_threaded() # <-- This is all you need to add\ndef example(x, y) -> int:\n heavyworkload(1)\n return x*y\n\nstart = time.time()\nresults = []\nfor i in range(5):\n results.append(example(i, 10)) # the thread is started\n\nprint(results) # autothread waits for the thread to end and gives you the result\nprint(\"Time expired: \", time.time()-start)\n>>> [0, 10, 20, 30, 40]\n Time expired: 1.002363681793213\n```\n\n`autothread.async_processed` works in the same way but uses multiprocessing instead of threading.\nMore info can be found in the [non-blocking README](https://github.com/Basdbruijne/autothread/blob/main/docs/README_non_blocking.md).\n\n## Blocking\nThe blocking decorators of autothread change the function slightly, but give you more control\nover when the function is executed:\n\n```python\nimport autothread\nimport time\nfrom time import sleep as heavyworkload\n\n@autothread.multithreaded() # <-- This is all you need to add\ndef example(x: int, y: int):\n heavyworkload(1)\n return x*y\n\n@autothread.multiprocessed() # <-- Or to use multiprocessing\ndef example2(x: int, y: int):\n heavyworkload(1)\n return x*y\n```\n\nNow, instead of integers, your function can take lists of integers. The function will\nbe repeated or each item in your list on a separate thread/process:\n```python3\nstart = time.time()\nresult = example([1, 2, 3, 4, 5], 10)\nprint(result)\nprint(\"Time expired: \", time.time()-start)\n>>> [10, 20, 30, 40, 50]\n Time expired: 1.0041766166687012\n```\n\nMore info can be found in the [blocking README](https://github.com/Basdbruijne/autothread/blob/main/docs/README_blocking.md).\n\n## Installing\n\nYou can install autothread using:\n```\npip install autothread\n```\n\nOr by cloning the source:\n```\ngit clone https://github.com/Basdbruijne/autothread.git\ncd autothread\npip install -e .\n```\n\n## Known issues\nNone at the moment, please open a bug if you run into an issue.\n",
"bugtrack_url": null,
"license": "MIT",
"summary": "",
"version": "0.0.10",
"project_urls": {
"Homepage": "https://github.com/Basdbruijne/autothread"
},
"split_keywords": [
"multithreading",
"multiprocessing",
"decorator"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "4bef9927258f0224c0f20561ec15b3b86d2373cb28cd22f664ba712a393817b1",
"md5": "775072877009aeaa90f672dc021f80a9",
"sha256": "938aab973fdd3c85a58571d6cefc758f7f070a08a451244eedaa0f764a8807fd"
},
"downloads": -1,
"filename": "autothread-0.0.10.tar.gz",
"has_sig": false,
"md5_digest": "775072877009aeaa90f672dc021f80a9",
"packagetype": "sdist",
"python_version": "source",
"requires_python": null,
"size": 14463,
"upload_time": "2023-05-09T11:48:57",
"upload_time_iso_8601": "2023-05-09T11:48:57.137379Z",
"url": "https://files.pythonhosted.org/packages/4b/ef/9927258f0224c0f20561ec15b3b86d2373cb28cd22f664ba712a393817b1/autothread-0.0.10.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2023-05-09 11:48:57",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "Basdbruijne",
"github_project": "autothread",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"requirements": [
{
"name": "multiprocess",
"specs": []
},
{
"name": "psutil",
"specs": []
},
{
"name": "tqdm",
"specs": []
},
{
"name": "typeguard",
"specs": [
[
">=",
"3.0.2"
]
]
},
{
"name": "typing",
"specs": []
}
],
"tox": true,
"lcname": "autothread"
}
JSON
