While I certainly don't want to discourage anyone from hacking on anything they enjoy, if your primary purpose is to <i>use</i> ordered collections, FSet [0] has them, and much, much, <i>much</i> more. It's Quicklisp-loadable.<p>[0] <a href="https://github.com/slburson/fset" rel="nofollow">https://github.com/slburson/fset</a>
A real set implementation is 250x faster than using a list as a set (doesn't mention which operations or include the numbers though..)? One would hope so for large sets. That is what the author is referring to by built-in set functionality, right?
I somehow missed previous submissions from this blog, but I've now favorited some of them and subscribed to its feed.<p>In particular, <a href="http://vicsydev.blogspot.com/2017/01/on-tests.html" rel="nofollow">http://vicsydev.blogspot.com/2017/01/on-tests.html</a> resonates very much with my approach to organizing programs:<p>1. <a href="https://github.com/akkartik/wart/blob/91356d9385/organization" rel="nofollow">https://github.com/akkartik/wart/blob/91356d9385/organizatio...</a><p>2. <a href="http://akkartik.github.io/mu/html/000organization.cc.html" rel="nofollow">http://akkartik.github.io/mu/html/000organization.cc.html</a> (<a href="https://github.com/akkartik/mu/blob/61fb1da0b6/000organization.cc" rel="nofollow">https://github.com/akkartik/mu/blob/61fb1da0b6/000organizati...</a>)<p>The ability to 'tag' tests into multiple suites is particularly nice. I might steal that at some point, but I'm first going to think about how it might synergize with my Literate Programming approach (<a href="http://akkartik.name/post/wart-layers" rel="nofollow">http://akkartik.name/post/wart-layers</a>)
Are you the author of the blog? Scrolling the code snippet keeps causing prev/next post page changes. It's really frustrating are you able to fix it?