Shawn Presser

Results 41 comments of Shawn Presser

You're right! `ustring.new("a") == {"a"}` evaluates to `true`. I've made another attempt. Can you think of any clever ways of breaking this?

Actually... It has nothing to do with `document.cookie`. When I set `cookie: ...` in the `headers: ` field, I also set `'user-agent': 'User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_14_5)...

No, I don't think so. If you grep the codebase for `fetch(...`, you'll notice `src/DOM.js` doesn't actually pass in exokit's useragent as a fetch parameter. It's using the default value...

I can confirm that changing every `fetch(foo)` to ```fetch(foo, {headers: {'user-agent': `User-Agent: ${this.ownerDocument.implementation._window.navigator.userAgent}`}})``` in `src/DOM.js` fixes this issue. Is there a better way to access `window.navigator.userAgent` for the current HTML...

https://stackoverflow.com/questions/16010204/get-reference-of-window-object-from-a-dom-element seems to work: ``` i=document.createElement('img'); document.body.appendChild(i); doc=i.ownerDocument; doc.defaultView || doc.parentWindow; ```

backtrace ![image](https://user-images.githubusercontent.com/59632/61353913-8a176480-a836-11e9-9a19-e9ebd9696251.png)

+1. This would be a step toward a ClojureScript webpack loader.

You can now download the original epub files for bookcorpus: https://battle.shawwn.com/bookcorpus-epub.tar It's 14.2GB with 17,876 epub files. The tarball also contains `bookcorpus/2020-08-27-epub_urls.txt` which is a file containing the original URLs...

> Does it include [books3.tar.gz](https://the-eye.eu/public/AI/pile_preliminary_components/books3.tar.gz) the LibGen db (in *.txt)? @ofou No, but I do have a copy of those epub files. Someday I'll get around to packaging them up.

Oh yes, MIT is perfect!