Randomly generate a fake useragent.

This project's idea is inspired by fake-useragent. I rewrote the whole codes in order to boost performance by:

• using asyncio and aiohttp to improve fetching speed
• taking advantage of Xpath to improve parsing speed
• changing random choice algorithm to improve random choice speed

Supported browsers are: chrome, edge, firefox, safari, and opera. Browser name is case insensitive. Some other possible spellings of each browser are mapped to the right one (e.g. "ie" -> "edge", "google" -> "chrome").

It takes less than 2s for the first run, including fetching, parsing, and writing cache.

It will take less than 0.01s starting from the second time by using cache by default.

Usage

As a binary

On your terminal, just simply enter fakeua

Other usages on terminal:

# Get a useragent in debug mode
fakeua <browser or omit> --debug

# Get a useragent without local caching
fakeua <browser or omit> --nocache

# Remove cache from $HOME/.cache/fakeua folder
fakeua --remove  

# Print the current version of the program
fakeua --version

As a library

In your python script, import the function. Every time you run the script, the useragent value will be different.

from fake_user_agent import user_agent

# Not to specify a browser
ua = user_agent()

# Specify a browser to randomly choose from
ua = user_agent("chrome")

# Specify not using cache, it will take < 1s to run, including fetching and parsing.
ua = user_agent(use_cache=False)

# If there is an async function needing a useragent in your script,
# don't put `user_agent()` in your async function, put it above instead.

# Remove cache in a python script. 
# May need `sudo python yourscript.py` for Linux.
from fake_user_agent import rm_cache
rm_cache()  

Installation

pip install fake_user_agent