00:02 <FromGitter> <Blacksmoke16> nice was able to do some refactoring, brought compile time for specs down from 27s to 5s

00:02 <FromGitter> <Blacksmoke16> big unions are bad ok

00:32 <FromGitter> <Blacksmoke16> say i have a `Regex` constraint, and you want to validate a non string value what would you think is most ideal:

00:33 <FromGitter> <Blacksmoke16> 1) Cast the value to a string ⏎ 2) Throw an exception (because it has to be a string)

00:33 <FromGitter> <Blacksmoke16> 1) Noop

01:27 <FromGitter> <Blacksmoke16> i think im leaning towards 1, prob would pair well with custom `to_s` methods and such

04:01 <FromGitter> <mattrberry> Anyone have any feedback/criticism for my bitfield macro? It's my first macro that's more than a couple lines. At the moment it accomplishes everything that I wanted it to, but I'd love to know if there's anything that I could be doing better! https://github.com/mattrberry/bitfield

04:03 <FromGitter> <mattrberry> One thing I'd like to add is the ability to override the setters on the inheriting classes, but I'm not sure that that's possible in my current implementation since I add the methods in the `finished` macro..

04:04 <FromGitter> <mattrberry> I also plan on adding a to_s method sometime soon :)

07:14 <FromGitter> <grkek> How do I get the name of the class?

07:34 <FromGitter> <grkek> in a macro

08:52 <oprypin> @grkek `{{@type}}`

09:35 <FromGitter> <Nicolab> Hello, ⏎ ⏎ I want define a logger from a parent class but `self` is the parent class. ⏎ How to get the concret class from a parent class ? ⏎ ... [https://gitter.im/crystal-lang/crystal?at=5f6f0b7ab8a99f4519aa8b59]

10:03 <FromGitter> <Nicolab> Ok found: `{{ @type.subclasses.last }}`

10:17 <FromGitter> <Nicolab> But works on 2 levels

11:24 <FromGitter> <Nicolab> Solution: https://forum.crystal-lang.org/t/how-to-get-concrete-class-name-from-the-parent/2562/2?u=nico

12:00 <FromGitter> <naqvis> > i think im leaning towards 1, prob would pair well with custom `to_s` methods and such ⏎ ⏎ I would have gone with 2, as library shouldn't make assumptions about user. If user is not utilizing api in a way its designed, it should let the user know that. just my two cents :P

12:00 <FromGitter> <naqvis> @Blacksmoke16 👆

12:19 <raz> i think it depends on whether the object to be validated will always be passed in explicitly or not.

12:19 <raz> if i pass in `foo` then i don't want to be surprised by the library casting it around (if i want it to be treated as string, i'll pass in `foo.to_s`)

12:21 <raz> otoh if String is the only value-type that can make sense, and i get a solid compile-time error when i pass something wrong, hard-wiring it can also make sense

13:30 <FromGitter> <Blacksmoke16> its a tricky thing, i can do an error but it would be a runtime error

13:31 <FromGitter> <Blacksmoke16> i imagine most of the time people would be interacting with the annotation API, so having this happen automatically would allow you to do like

13:32 <FromGitter> <Blacksmoke16> ```@[Assert::URL] ⏎ @uri : URI``` [https://gitter.im/crystal-lang/crystal?at=5f6f42d01c5b0d210ad1a576]

13:32 <FromGitter> <Blacksmoke16> as the `uri` object would be passed in, turned into the string representation of it, and validated

13:33 <FromGitter> <Blacksmoke16> plus i dont really think there's some common sense involved, like why would you be trying to assert `foo` is an `Email`?

13:33 <FromGitter> <Blacksmoke16> i think there is some common sense involved*

14:09 <FromGitter> <maxbertinetti> @jgillich @faustinoaq @kofno can you please remove your plugins from VS Code Marketplace? ⏎ As said also by @asterite on the Forum actually the Marketplace is a mess of obsolete Crystal plugins. ⏎ Removing them will make it simple for beginner find the right one. ⏎ Also, anyone can contribute to the one on crystal-language-tool organisation (https://crystal-lang-tools.github.io/)

14:09 <FromGitter> ... [https://gitter.im/crystal-lang/crystal?at=5f6f4b7caaff36059b5a6cd0]

18:40 <FromGitter> <Nicolab> @maxbertinetti Just for info, the @faustinoaq plugin works a bit for me. Yours doesn't. I will test with Crystalline but it is possible that others will have the same result.

18:41 <FromGitter> <Daniel-Worrall> +1

19:30 <FromGitter> <Blacksmoke16> well, am i missing any constraints? https://github.com/athena-framework/validator/tree/develop-v2/src/constraints

19:31 <FromGitter> <Blacksmoke16> prob some more i can add but this is a good start

19:40 <FromGitter> <Blacksmoke16> now to start on documenting everything 😬

23:01 <FromGitter> <maxbertinetti> @Nicolab the official is the new version of the @faustinoaq one.