Change organization

.gitignore

@@ -16,4 +16,4 @@ lib_managed/
 src_managed/
 project/boot/
 project/plugins/project/
-
+.idea/

build.sbt

@@ -1,6 +1,6 @@
 name := "sjson"
 organization := "io.github.lock-free"
-version := "0.2.0"
+version := "0.2.5"
 scalaVersion := "2.12.4"
 
 useGpg := true 
@@ -11,5 +11,7 @@ publishTo := sonatypePublishTo.value
 libraryDependencies ++= Seq(
   "org.scala-lang" % "scala-reflect" % scalaVersion.value,
   // test suite
-  "org.scalatest" %% "scalatest" % "3.0.1" % Test
+  "org.scalatest" %% "scalatest" % "3.0.1" % Test,
+  //performance test suite
+  "com.storm-enroute" %% "scalameter" % "0.18"
 )

project/plugins.sbt

@@ -2,4 +2,4 @@ addSbtPlugin("com.typesafe.sbt" % "sbt-native-packager" % "1.3.3")
 addSbtPlugin("org.scoverage" % "sbt-scoverage" % "1.5.1")
 
 addSbtPlugin("org.xerial.sbt" % "sbt-sonatype" % "2.3")
-addSbtPlugin("com.jsuereth" % "sbt-pgp" % "1.1.0")
+addSbtPlugin("com.github.sbt" % "sbt-pgp" % "2.1.2")

src/main/scala/io/github/shopee/idata/sjson/JSONUtil.scala

@@ -110,8 +110,31 @@ object JSONUtil {
     s""""${txtBuilder.toString()}""""
   }
 
+  //returns a boolean array indicating where the start of a unicode substring is
+  //e.g. "\"\\uD835\" returns [False, True, False, False, False, False, False]
+  def uniCodeArrayBuilder(txt: String): Array[Boolean] = {
+    val unicodeArray = new Array[Boolean](txt.length)
+    val (allStringHexaDecimal, startPos) = (4, 3)
+    var (hexadecimalCount, startPtr) = (0, 0)
+    for (endPtr ← startPos to unicodeArray.length -  1){
+      startPtr = endPtr - 4
+      if(txt.charAt(endPtr).isDigit || isHexAlphabet(txt.charAt(endPtr))) hexadecimalCount = hexadecimalCount + 1
+      //startPtr starts from first possible hexadecimal character (e.g. Ds in "\uD835) that needs to be removed
+      //in a sliding window
+      if(startPtr >= 3 && (txt.charAt(startPtr).isDigit || isHexAlphabet(txt.charAt(startPtr))))  hexadecimalCount = hexadecimalCount - 1
+      if(hexadecimalCount == allStringHexaDecimal) unicodeArray.update(startPtr - 1, true)
+    }
+    unicodeArray
+  }
+
+  def isHexAlphabet(ch: Character): Boolean = {
+    if(ch >= 'A' && ch <= 'F') return true
+    false
+  }
+
   def unescapeString(txt: String): String = {
     val txtBuilder = new StringBuilder // use txt builder to collect text
+    val uniCodeArray = uniCodeArrayBuilder(txt)
 
     var i   = 1
     var len = txt.length - 1
@@ -126,10 +149,23 @@ object JSONUtil {
           case 'n' => '\n'
           case 'f' => '\f'
           case 'r' => '\r'
+          case 'u' => 'u'
           case _   => next
         }
-        txtBuilder.append(newChar)
-        i += 2
+        if(newChar == 'u' && uniCodeArray(i)) {
+          val unicodeString = s"\\u${txt.substring(i + 2, i + 6)}"
+          val unicodeChar = Integer.parseInt(unicodeString.drop(2), 16).toChar
+          txtBuilder.append(unicodeChar)
+          i += 6
+        }
+        else if (newChar == 'u') {
+          txtBuilder.append("\\u")
+          i += 2
+        }
+        else {
+          txtBuilder.append(newChar)
+          i += 2
+        }
       } else {
         txtBuilder.append(ch)
         i += 1

src/test/scala/io/github/shopee/idata/sjson/JSONUtil.scala

@@ -1,5 +1,7 @@
 package io.github.free.lock.sjson
 
+import org.scalatest.Matchers.{convertToAnyShouldWrapper, equal}
+
 class JSONUtilTest extends org.scalatest.FunSuite {
   test("unescapeString") {
     assert(JSONUtil.unescapeString(JSONUtil.escapeString("1234")) == "1234")
@@ -10,4 +12,34 @@ class JSONUtilTest extends org.scalatest.FunSuite {
     assert(JSONUtil.unescapeString(JSONUtil.escapeString("12\f34")) == "12\f34")
     assert(JSONUtil.unescapeString(JSONUtil.escapeString("12\\34")) == "12\\34")
   }
+
+  test("unicodeArrayBuilderSimple"){
+    val currString = "\"\\uD835\\uD83\""
+    val unicodeArray = new Array[Boolean](currString.length)
+    unicodeArray.update(1, true)
+    JSONUtil.uniCodeArrayBuilder(currString) should equal (unicodeArray)
+  }
+
+  test("unicodeArrayBuilderWithInvalidUniCodeInBetween2"){
+    val currString = "\"\\uD835\\uD835\\uD83\\uDC07\""
+    val unicodeArray = new Array[Boolean](currString.length)
+    unicodeArray.update(1, true)
+    unicodeArray.update(7, true)
+    unicodeArray.update(18, true)
+    JSONUtil.uniCodeArrayBuilder(currString) should equal (unicodeArray)
+  }
+
+  test("unicodeArrayBuilderWithInvalidUniCodeInBetween"){
+    val currString = "\"\\uD835\\uD83\\uD835\""
+    val unicodeArray = new Array[Boolean](currString.length)
+    unicodeArray.update(1, true)
+    unicodeArray.update(12, true)
+    JSONUtil.uniCodeArrayBuilder(currString) should equal (unicodeArray)
+  }
+
+  test("unicodeArrayNoUnicode"){
+    val currString = "\"\\uD83za\\uD83zxd\""
+    val unicodeArray = new Array[Boolean](currString.length)
+    JSONUtil.uniCodeArrayBuilder(currString) should equal (unicodeArray)
+  }
 }

src/test/scala/io/github/shopee/idata/sjson/ParseTest.scala

@@ -12,6 +12,38 @@ class ParseTest extends org.scalatest.FunSuite {
     assert(JSON.parse(JSON.stringify(v)) == v)
   }
 
+  test("parse: map unicode") {
+    val input = "[{\"\\u\":\"\\u\"}]"
+    val output = JSON.parse(input).asInstanceOf[List[Map[String, String]]](0)("\\u")
+    assert(output ==  "\\u")
+  }
+
+  test("parse: map unicode2") {
+    val input = "[{\"\\uD835\":\"\\uD835\"}]"
+    val output = JSON.parse(input).asInstanceOf[List[Map[String, String]]](0)("\uD835")
+    assert(output ==  "\uD835")
+  }
+
+  test("parse: map unicode3") {
+    //actual string is 𝐇𝐨𝐧𝐝𝐚
+    val input = "[{\"\\uD835\\uDC07\\uD835\\uDC28\\uD835\\uDC27\\uD835\\uDC1D\\uD835\\uDC1A\":\"\\uD835\\uDC07\\uD835\\uDC28\\uD835\\uDC27\\uD835\\uDC1D\\uD835\\uDC1A\"}]"
+    val output = JSON.parse(input).asInstanceOf[List[Map[String, String]]](0)("\uD835\uDC07\uD835\uDC28\uD835\uDC27\uD835\uDC1D\uD835\uDC1A")
+    assert(output ==  "\uD835\uDC07\uD835\uDC28\uD835\uDC27\uD835\uDC1D\uD835\uDC1A")
+  }
+
+  test("parse: map unicode4") {
+    val input = "[{\"シリアライゼーション\":\"シリアライゼーション\"}]"
+    val output = JSON.parse(input).asInstanceOf[List[Map[String, String]]](0)("シリアライゼーション")
+    assert(output ==  "シリアライゼーション")
+  }
+
+  test("parse: map unicode5") {
+    val input = "[{\"\\uDC\\uD835\\uDC07\\uDC\":\"\\uD835\\uDC07\"}]"
+    //𝐇
+    val output = JSON.parse(input).asInstanceOf[List[Map[String, String]]](0)("\\uDC\uD835\uDC07\\uDC")
+    assert(output ==  "\uD835\uDC07")
+  }
+
   test("parse: true|false|null") {
     List[Any](true, false, null).map(testParseSym)
     assert(JSON.parse(JSON.stringify(None)) == null)

src/test/scala/io/github/shopee/idata/sjson/TokenParserTest.scala

@@ -29,7 +29,9 @@ class TokenParseTest extends org.scalatest.FunSuite {
   }
 
   test("toTokens: single string") {
-    List(s"""""""", s""""hello, world"""", s""""123"""", s""""\\""""", s""""\n"""", s""""\t"""", s""""\\\\"""").map((txt) => {
+    List(s"""""""", s""""hello, world"""", s""""123"""", s""""\\""""",
+      s""""\\n"""", s""""\t"""", s""""\\\\"""", s""""\\r"""", s""""\\b"""",
+      s""""\\f"""", s""""\\/"""").map((txt) => {
       testToToken(txt, List(JSONToken(JSONToken.STRING, txt)))
     })
   }