⚡ Fast, async, stream-based link checker written in Rust. Finds broken URLs and mail addresses inside Markdown, HTML, reStructuredText, websites and more!
Previously, we didn't pass (auth) headers to the initial set of inputs -- only to the links which were discovered in the initial input set. That's inconsistent and led to some confusion.
This PR fixes that. There is probably a nicer design for input parsing; the constructor has quite a few "dummy" params now, which tend to be set to the same values. I guess we can refactor that to, say, introduce the notion of an InputBuilder if it becomes to tedious to set these values. I guess we should decouple input parsing for the different input types in general because header handling is only needed for web requests and not for file inputs.
For now, I just introduced a helper method Input::from_value.
Previously, we didn't pass (auth) headers to the initial set of inputs -- only to the links which were discovered in the initial input set. That's inconsistent and led to some confusion.
This PR fixes that. There is probably a nicer design for input parsing; the constructor has quite a few "dummy" params now, which tend to be set to the same values. I guess we can refactor that to, say, introduce the notion of an
InputBuilderif it becomes to tedious to set these values. I guess we should decouple input parsing for the different input types in general because header handling is only needed for web requests and not for file inputs.For now, I just introduced a helper method
Input::from_value.This fixes #1441.