scala-hamsters / hamsters Goto Github PK
View Code? Open in Web Editor NEWA mini Scala utility library
License: Apache License 2.0
A mini Scala utility library
License: Apache License 2.0
The first one returns a list of Right types and the second one returns directly a List of errors (Left side type).
Hello all,
Thanks all for this lib, keep it up!
Retry.withWait
could wrap the Thread.sleep
inside a blocking
block, which warns the global execution context (so that it might add more threads).
Also I'd appreciate make it clearer in the docs that the method blocks a Thread.
To be able to make monad transformers work with more types
To be more consistent
Example :
val sum = (optionalInt1,optionalInt2).mapN((l, r) => l+r)
Maybe we could do something for Either, Validation, UnionType...
Validation.result vs Validation.failures. Use plural form for both
Edit : use rename Validation.result to Validation.run (see comments)
sbt +test
should run tests in scala 2.11 and scala 2.12.
Since last merge (scala.js branch) only 2.11 seems to be run.
Same issue occurs with sbt +publishSigned
.
@Gen
case class User(firstName : String, lastName : String)
We could generate a companion object with methods :
implicit def toHlist(user) : String :: String :: HNil=???
implicit def fromHlist(v : String :: String :: HNil) : User=???
It's a problem if you need to type check the content of the resulting HList, for example if it's returned by a method.
So we may enhance the map/filter methods but only if it is possible to keep a (quite) simple HList model.
For more advanced Hlist features, people should use Shapeless instead.
In the same time, +++
private method in HCons could be removed in favor of ++
Hello,
Is there a reason why we can not build a Validation from list through apply() method as
Validation(List(Ok, KO, KO))
Currently, I have to use Traditional approach with the new operator.
We could add the apply(l:List) in companion object of Validation class, what do you think about it ?
Many thanks for you help,
Fabrice
to enable usage in maven-projects, it would be really useful to add this library to maven central
Add another method like this :
def apply[T](maxRetries: Int, waitInMilliSeconds: Int, errorFn: (String) => Unit = _=> Unit)(fn: => T): Future[T]
Future will be used to avoid blocking in the main thread
error : publish repository is undefined
See branch sbt_1_0
It would be nice to be able to use hamsters with Scala.js.
There is a typo in the dependency, it should be:
libraryDependencies ++= Seq(
"io.github.scala-hamsters" %% "hamsters" % "1.0.0-SNAPSHOT",
)
For now, you need to explicitly and manually write the following method
implicit val seasonEnumerable = new Enumerable[Season] {
override def list: List[Season] = List(Winter, Spring, Summer, Fall)
}
What's happen when i add another value.
If we generate this method with a macro (annotation vs def macros...), that could help.
Like shapeless select on HList, it would be great to be able to select a value by type like
def select[U]
Current message : could not find implicit value for parameter evidence: io.github.hamsters.Monad[scala.concurrent.Future]
We should use @implicitNotFound to define a better one
In a Validation, there only 1 error type, but valid data can have different types.
It would be great to have a way to retrieve all valid data, for example something like :
val e1 = OK("1")
val e2 = OK(2)
val e3 = OK(3)
val validation = Validation(e1,e2, e3)
validation.map(_ + _ + _) // Validation("123")
validation.getOrElse("failure")//123
map
should be triggered only if all eithers are valid.
Current state of FutureOption
and FutureEither
does not allow pattern matching in for-comprehensions.
For example, the following wouldn't compile :
def foa: Future[Option[String]] = Future(Some(("a", 42)))
val composedAB: Future[Option[Int]] = (for {
(a, i) <- FutureOption(foa) if i > 5
} yield a).future
This would be solved by implementing withFilter
in both classes
Documentation should explain how to use it with option and list and how to extend to other types with the right implicit definitions.
First of all nice project. Could you please publish a first release eg. on bintray?
val stringOpt: Option[String] = None
stringOpt.orEmpty // ""
val intOpt: Option[Int] = None
intOpt.orEmpty // 0
val listOpt: Option[List[Int]] = None
listOpt.orEmpty // List()
We should keep a link to the doc in the Readme
The goal is to automatically provide a better toString for case classes with field names.
Example :
case class Person(name: String, age: Int)
implicit personShow = ... // calling macro here
val p = Person("bob", 40)
Show.show(p) // returns "Person(name = bob, age = 40)"
//or
// import implicit class
p.show // returns "Person(name = bob, age = 40)"
If i create an empty project and make
sbt clean && sbt clean && sbt assembly && ls -lah target/scala-2.12/*.jar
I get a weight :
libraryDependencies += "io.github.scala-hamsters" %% "hamsters" % "1.5.1"
=> 14Mo
libraryDependencies += "com.chuusai" %% "shapeless" % "2.3.2"
=> 7.7 Mo
libraryDependencies += "org.scalaz" %% "scalaz-core" % "7.2.16"
=> 13.1Mo
Our dependencies depends of scala.meta, it should be a provided depency
Warning : Either is already right biased in scala 2.12
sbt doc generates html files for hamsters + scala meta
Create a REPL with hamsters dependencies
scala> implicitly[DefaultValue[BigDecimal]].get
java.lang.ClassCastException: java.lang.Integer cannot be cast to scala.math.BigDecimal
... 38 elided
You need to use the .zero
member of Numeric[T]
instead of trying to cast 0
.
Validation.result(t1,t2, ...)
should accept Try[T] too. It could convert to Either[Throwable, T] internally.
Missing doc :
Now show only supports String
and Int
.
We should support all the numeric primitives, java.util.Date
. java.time.*Date
...
The dependency version is not correct in the readme, it states:
libraryDependencies ++= Seq(
"io.github.scala-hamsters" %% "hamsters" % "1.0-BETA1"
)
but it should be:
libraryDependencies ++= Seq(
"io.github.scala-hamsters" %% "hamsters" % "1.0.0-BETA1"
)
Please change 1.0-BETA1 to 1.0**.0**-BETA1
This is working :
val sum = (2.0 :: "hi" :: HNil) ++ (1 :: HNil)
sum shouldBe 2.0 :: "hi" :: 1 :: HNil
sum shouldBe a[HCons[_, HCons[_, HCons[_, HNil]]]]
Folding/Iterating on sum content works as expected :
sum.foreach(println)
Result :
// 2.0
// hi
// 1
But this fails :
sum.tail.tail.head
--> compilation error : value head is not a member of io.github.hamsters.HNil
We can see with the first test that sum.tail.tail is a HCons but the compiler returns a HNIL because ++ keeps the type of the first HList, HCons[_, HCons[_, HNil]]]
in this example instead of [HCons[_, HCons[_, HCons[_, HNil]]]]
.
Command does not fail but no html is generated.
It would be nice to be able to automatically :
With list, toString, parse methods working with any case object
We should use macros to generate union classes from 1 to n, n to be defined.
n should not be to high as t would not be convenient to pattern match.
trait Enumerable[A] {
def name(a: A): String = a.toString.toLowerCase
def parse(s: String): Option[A] = list.find(a => name(a) == s)
def list: List[A]
}
should become
trait Enumerable[A] {
def name(a: A): String = a.toString.toLowerCase
def parse(s: String): Option[A] = list.find(a => name(a).equalsIgnoreCase(s))
def list: List[A]
}
This method could be named Retry.fromTry
Then refactor retry to avoid code duplication.
val hList : String :: Boolean :: HNil
should compile.
Currently only val hList : Hcons[String :: HCons[Boolean :: HNil]]
is working.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.