<div dir="ltr"><div class="gmail_extra"><div class="gmail_quote">On Sat, Apr 23, 2016 at 1:05 AM, Andrew Berman <span dir="ltr"><<a href="mailto:rexxe98@gmail.com" target="_blank">rexxe98@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex"><div dir="ltr">Valid point. Then why test on is_uri? Every string should come back as true.</div></blockquote></div><br><div>I'd like to reiterate what Michał Muskała said. Why are you hung up on what a single, practically unused library is doing and judging an entire language on it?</div><div><br></div><div>It's strongly recommended to not create packages like is_url, we suggest to only publish packages that provide some minimum value and it's definitely against the spirit of what kind of packages we would like to be pushed to <a href="http://hex.pm">hex.pm</a>. It's not against Hex's code of conduct to create libraries like is_url because there is no practical way to enforce it because the limit you draw would be somewhat arbitrary - how do you determine what's a good or minimum viable package? There are nearly 2000 packages on <a href="http://hex.pm">hex.pm</a>, you are bound to found some that are tiny or of low quality if look hard enough.</div><div><br></div><div>To me it seems more like someone was trying out how to publish packages and published this one as a test or for fun.</div><div><br></div><div>It also seems like Kenneth Lakin has a very good explanation of why URI.parse works like it does. Let it crash is just as much a core philosophy in Elixir as it is in Erlang and if you'd look at the rest of the standard library that would be obvious.</div><div><br></div>-- <br><div class="gmail_signature">Eric Meadows-Jönsson</div>
</div></div>