Adam Pash

@vhfmag This is an interesting thought. For some context, Mercury is focused _primarily_ on extracting metadata from articles on the web, and while we're not against supporting more content, that's...

Hmm, this is an interesting question, and you're right, it certain changes expectation. One other thing, just by looking at the test output, is that it's also complicated by the...

> Maybe a hybrid would be best? What if only the `content` key adopted the new behavior of `decodeEntities: false`, and all other keys retained the current behavior. I would...

@benubois Thanks again for looking into this. Just let me know when you've got something you'd like a review on.

Hmm, you're right, this is a tough one to resolve. Even in the failing tests for the images, there are query parameters in the URL that actually make a difference...

@benubois If this is only a problem for reading the HTML before it's actually rendered on a page, I agree that maybe it's not a huge concern. The one use...

Sorry, I know the maintainers are probably busy, but — does this seem possible? I need to decide if I will need to maintain my own branch or not on...

This is still an issue, right? I'm running into the same issue right now, and this seems like the canonical issue for this.

`yarn deploy` now deploys the the lambda and api gateway code, then also builds and deploys the static assets to S3. The API Gateway endpoint is configured to serve the...

I'm using Absinthe in production and have started running into this problem as well. Has anyone had any luck on workarounds, or is using 1.5 the best option for now?