url-strip 0.0.1

Library to strip http urls of tracking elements, and use shorthand variants of urls

Url Strip

A library for stripping urls of tracking and bloat

Non Pythonic

Usage was designed around typesafety and zero runtime surprises, so instead of raising exceptions, url_strip will return exceptions instead of results, this library works best with a typechecker because of this

Usage

Some basic usage. These examples are all runnable and meet typing standards.

Using is_instance and unwrap

"""
Uses is_instance and unwrap to parse a url
"""

from url_strip import Ok, strip_url

# Ok and Err provide the same 3 methods with fully typed outcomes for a Result[T, E]:
#  is_instance will TypeGuard the result to the variant
#  get will return an Optional[T] of the variant
#    (and is not suitable for cases where T or E are None)
#  unwrap with return the requested variant or raise an exception
if Ok.is_instance(
    v := strip_url(
        "https://youtube.com/watch?v=dQw4w9WgXcQ"
        "&trackerinfo=youraddresshere&mldata=whattimeyouwokeupthismorning"
        )
    ):
    # using unwrap is runtime safe here as we just typeguarded it is an Ok variant
    url = Ok.unwrap(v)
    # strip_url returns a Result[HttpUrl, UrlError], and HttpUrl provides a into_str method
    #  to get what most people expect as a final output
    value = url.into_str()

    print(value)  # -> https://youtu.be/dQw4w9WgXcQ
else:
    ...  # Insert error handling here

Using get

"""
Uses get and a None check to parse an invalid url
"""

from url_strip import strip_url, Ok, Err

if (v := Ok.get(result := strip_url("foo"))) is not None:
    v.into_str()
else:
    raise Err.unwrap(result)

Writing extra domain rules

"""
Uses register to add foo.com as a checked domain
"""
from url_strip import UrlError, Ok, Err, StripFuncResult, HttpUrl, register

@register(domain="foo.com")
def foo_com_strip( # pyright: ignore[reportUnusedFunction]
    url: HttpUrl
    ) -> StripFuncResult:  # ('ok', HttpUrl) | ('err', UrlError)

    if url.path == "/failure":
        return Err(UrlError("This is an awful failure state"))

    return Ok(url)  # dont change url at all, its perfect as is

Popular Websites Supported

Some popular websites have pre-register'd rules, while any non registered websites will go to a default query stripper
The pre registered domains are as follows:

Site	Domains
amazon	www.amazon.com, www.amazon.co.uk
ebay	ebay.com, www.ebay.com, www.ebay.co.uk
reddit	www.reddit.com
tiktok	www.tiktok.com, vm.tiktok.com
twitter	twitter.com
youtube	youtube.com, www.youtube.com