One gap I’ve personally felt, is PDF parsing. I know that PDFs suck and all, but still!

A complete authentication and authorisation library like .NET’s Identity. Users, roles, claims, tokens, everything is covered.

We often complain that the ecosystem is not very big, not appealing to companies and therefore it’s hard to find jobs for developers. At the same time we are quick to dismiss a noob-proof solution as too generic. We prefer to roll our own, supposedly complete, authn and authz solution, because we’re not noobs.

Yeah, I don’t prefer to roll my own solutions to problems that should be solved by de facto standard libraries. Also don’t want to switch to Ash, just to use their solution. A generic solution can be specialised afterwards but a missing or already specialised can’t be generalised.

Sorry for the rant. I’m mostly happy with the ecosystem but let’s not act like the grapes are sour because they’re out of reach.

Not a library but I love the thousands of “build it from scratch” guides, books, and videos exist in the Javascript and PHP community

You might be interested in this thread:

With regard to missing libraries, I think we’ve got most bases covered now (particularly when you combine decades of Erlang libraries - most of which you can use with Elixir) however what I would like to see, is perhaps some consolidated libraries, where maybe the best of the bunch get together and combine their libraries to make a super library

Perhaps even going beyond that into something more official or semi-official, like what the Elixir Core Team recently did with the official language server team and where the idea was somewhat expanded as per this post