JarCastingsjson
JSON library for Scala
Converting JSON and Scala automaticly
JSON specification
Features
-
Type converting automaticly
-
Support customization converting
-
Streaming support
Quick example
package io.github.free.lock.json.example
import io.github.free.lock.sjson.JSON
case class User(name: String, age: Int)
object Main {
def main(args: Array[String]) {
// [1,2,3]
JSON.stringify(List(1,2,3))
// {"items": [1,2,3], "type": "testl"}
JSON.stringify(Map("items" -> List(1,2,3), "type" -> "test"))
// {"name": "NoName", age: 7}
JSON.stringify(User("NoName", 7))
/* parse to plain scala object */
// List(1,2,3)
JSON.parse("[1,2,3]")
// Map("a" -> 1, "b" -> 2)
JSON.parse(s"""{"a":1,"b":2}""")
// parse to target type
// User("ddchen", 10)
JSON.parseTo[User](s"""{"name":"ddchen","age":10}""")
}
}
Install
- install by sbt
libraryDependencies ++= Seq(
"io.github.lock-free" %% "sjson" % "0.1.2"
)
Main Apis
JSON.stringify
JSON.stringify(value: Any[, replacer]): String
Convert a scala object to json text.
- Default converting rules
At most times, you just need to follow the basic rules and do not need to write any converters by yourself. SJSON would use following basic rules, stringify value recursively.
| scala type | json type |
|---|---|
| String | string |
| Char | string |
| java.sql.Timestamp | string |
| java.util.Date | string |
| Number | number |
| Boolean | bool |
| None | null |
| Null | null |
| Unit | null |
| NaN | null |
| Infinity | null |
| AbstractSeq[_] | array |
| scala.collection.mutable.AbstractSeq[_] | array |
| Array[_] | array |
| scala.collection.mutable.Map[_, _] | object |
| scala.collection.immutable.Map[_, _] | object |
| other class | object (field name -> field value) |
import io.github.free.lock.sjson.JSON
// case class
case class User(name: String, age: Int)
val user = User("ddchen", 10)
JSON.stringify(user) // {"name":"ddchen","age":10}
JSON.stringify(List(1, 2, 3)) // [1,2,3]
JSON.stringify(Map(
"user" -> user,
"data" -> List(1,2,3)
)) // {"user":{"name":"ddchen","age":10},"data":[1,2,3]}
- Customize replacer
If you want to change the process of stringify, you can use optional paramater replacer.
replacer: (data, path) -> Option[newData], data is the value you are stringifying currently and path is the reversed json path.
case class User(name: String, age: Int, login: java.util.Date)
val user = User("ddchen", 10, new java.util.Date(1990 - 1900, 2, 12))
// {"name":"ddchen","age":10,"login":"1990,3,12"}
JSON.stringify(user, (data, path) => {
// path stored json paths in a reversed way. (path is a stack)
if (path.reserse.mkString(".") == "login") {
val date = data.asInstanceOf[java.util.Date]
Some(s""""${date.getYear() + 1900},${date.getMonth() + 1},${date.getDate()}"""") // new stringify result for data
} else {
// None means do not change the origin converting rule
None
}
})
JSON.parse
JSON.parse(jsonTxt: String[, replacer]): Any
Convert a json text to a scala object
| json type | scala type |
|---|---|
| string | String |
| number | Number |
| bool | Boolean |
| null | Null |
| array | List[Any] |
| object | Map[String, Any] |
After parsing, you will get a plain scala object. Then you can use JSON.convert[T](obj) to convert this object to target scala type.
JSON.convert
JSON.convert[T](value: Any): T
- Default converting rules
| plain scala type | scala type |
|---|---|
| same type | same type |
| List | Array |
| List | Vector |
| Map | Map |
| Map | Case class |
| Nothing | Null |
JSON.parseTo
JSON.parseTo[T](txt) = JSON.convert[T](JSON.parse(txt))
Streaming example
import io.github.free.lock.sjson.{JSON, AsyncIterator}
val textIter = new AsyncIterator[Char]()
// stream handle part
val parseIter = JSON.parseAsyncIterator(textIter, (data, pathStack, _) => {
if (JSON.toJsonPath(pathStack) == "data.[0]") { // because we wiped this data, so the json path should always be "data.[0]"
// handle data here
print(data) // can caputure all data like {"value":"datak"} in this example
JSON.WIPE_VALUE // do not store data in memory
} else data
}
)
// another place to accept streaming data
textIter.pushList(s"""{"type": "normal", """) // 1st chunk
// data part
textIter.pushList(s""""data": [""") // 2th chunk
textIter.pushList(s"""{"value":"data1"}, {"value":"data2"}, """)
textIter.pushList(s"""{"value":"data3"}, {"value":"data4"}, """)
textIter.pushList(s"""{"value":"data5"}, {"value":"data6"}, """)
// ...
textIter.pushList(s"""{"value":"datan"}""")
textIter.pushList("]}") // last chunk
