Add piecemeal import guide #1756
Conversation
|
Thanks! looking good to me so far. |
| ``` | ||
| The first import pulls the `Semigroup` instance for String into the scope, while the second import adds the `|+|` syntax. | ||
|
|
||
| You can also import all syntax or all instances by importing `cats.instances.all._` or `cats.syntax.all._` respectively. |
There was a problem hiding this comment.
Isn't this mixed up? I.e., it should be something like "(syntax or instances) ... (cats.syntax.all._ or cats.instances.all._) respectively". (Although english is not my first language.)
Codecov Report
@@ Coverage Diff @@
## master #1756 +/- ##
==========================================
+ Coverage 94.17% 94.22% +0.04%
==========================================
Files 256 256
Lines 4207 4208 +1
Branches 93 93
==========================================
+ Hits 3962 3965 +3
+ Misses 245 243 -2
Continue to review full report at Codecov.
|
|
🎉 so happy to see this finally land. It's something that I started on multiple times but never completed. Thanks @LukaJCB! |
| For example, if you'd like to import the `Monoid` instance for `String` and the corresponding syntax: | ||
| ```tut:book | ||
| import cats.instances.string._ | ||
| import cats.syntax.monoid._ |
There was a problem hiding this comment.
I was surprised this actually worked and we didn't need import cats.syntax.semigroup._, but it seems that MonoidSyntax extends SemigroupSyntax.
It seems that all the kernel syntax extends their "super syntax traits", which is another inconsistency between cats-kernel and cats-core (even though this syntax is in cats-core).
|
|
||
| //now we also need access to isEmpty from Monoid | ||
| import cats.syntax.monoid._ | ||
| (x |+| 1).isEmpty //error: value |+| is not a member of Int |
There was a problem hiding this comment.
If MonoidSyntax didn't extend SemigroupSyntax, we wouldn't have this issue.
You would only have this issue when you use cats.implicits._ or cats.xxx.all._ in combination with an a la carte import.
Still a WIP, but looking for some early feedback. Should resolve #1668