Ramiro Gómez

I upgraded the dependencies, but the code that is built after the upgrade doesn't work any more. I don't see a new release coming any time soon and you're probably...

I'd need a dataset with alternate country names. Also there is the possibility of ambiguities for example Korea and Kongo.

One thing that would certainly be interesting is comparing rendering performance using [sample graphs](http://snap.stanford.edu/data/) of different sizes in different browsers. Also which drawing technology is used by the library is...

A cookiecutter based approach could be interesting as well.

Thanks for the comprehensive report! The responses are converted to markdown using the [turndown](https://github.com/mixmark-io/turndown) package and ChatGPT uses [KaTeX](https://github.com/KaTeX/KaTeX) markup. If the conversion step would be skipped for responses containing...

I don't mean it is impossible to fix, but how would you retrieve the text of the response other than using the value of ELEMENT.textContent?

Given that source code is not converted correctly any more as well (see https://github.com/yaph/chatgpt-export/issues/8), maybe the current implementation should be replaced with triggering clicks on the copy button of each...

Sadly, I cannot access the content of the clipboard in the bookmarklet though, only in a browser extension that has the necessary permission.

I went through the click event in the dev console but couldn't figure out a way to get the text before it is copied to the clipboard.

Hi thanks for your proposal! Do you intend to add the type hints exclusively in .pyi files?