Overly permissive egg fragment parsing in URL requirements

[woodruffw]

First of all, thanks for maintaining this parser! It's been invaluable within pip-audit.

Per pypa/pip#11567: #egg=packagename==packageversion isn't actually valid, but this parser (and some others) accept it.

I'll wait for a resolution on that pip issue first, but I'm happy to submit a PR here that brings this library into compatibility with whatever pip decides 🙂

[pombredanne]

@woodruffw re:

First of all, thanks for maintaining this parser! It's been invaluable within pip-audit.

That's great! and what you are doing with "pip-audit" is wonderful.

Per pypa/pip#11567: #egg=packagename==packageversion isn't actually valid, but this parser (and some others) accept it.

I'll wait for a resolution on that pip issue first, but I'm happy to submit a PR here that brings this library into compatibility with whatever pip decides Please do: this would be great

FWIW, this parser is also heavily used in all the scancode projects and related in this org scanning eventually billions of requirements so this is going to be there for as long as there are requirements files.