fix CharReaderFromSyncStream read characters less than chunkSize (#2108)

korge-core/src/korlibs/io/stream/CharReader.kt

@@ -13,33 +13,50 @@ fun ByteArray.toCharReader(charset: Charset, chunkSize: Int = 1024): CharReader
 fun SyncStream.toCharReader(charset: Charset, chunkSize: Int = 1024): CharReader =
     CharReaderFromSyncStream(this, charset, chunkSize)
 
-class CharReaderFromSyncStream(val stream: SyncStream, val charset: Charset, val chunkSize: Int = 1024) : CharReader {
+class CharReaderFromSyncStream(val stream: SyncStream, val charset: Charset, val chunkSize: Int = DEFAULT_CHUNK_SIZE) : CharReader {
     private val temp = ByteArray(chunkSize)
     private val buffer = ByteArrayDeque()
     private var tempStringBuilder = StringBuilder()
 
+    init {
+        require(chunkSize >= MIN_CHUNK_SIZE) { "chunkSize must be greater than $MIN_CHUNK_SIZE, was $chunkSize" }
+    }
+
     override fun clone(): CharReader = CharReaderFromSyncStream(stream.clone(), charset, chunkSize)
 
     override fun read(out: StringBuilder, count: Int): Int {
-        while (buffer.availableRead < temp.size) {
-            val readCount = stream.read(temp)
-            if (readCount <= 0) break
-            buffer.write(temp, 0, readCount)
-        }
-
+        bufferUp()
         while (tempStringBuilder.length < count) {
             val readCount = buffer.peek(temp)
             val consumed = charset.decode(tempStringBuilder, temp, 0, readCount)
-            if (consumed <= 0) break
-            buffer.skip(consumed)
+            if (consumed <= 0) {
+                if (bufferUp() <= 0) break
+            } else {
+                buffer.skip(consumed)
+            }
         }
 
-        //println("tempStringBuilder=$tempStringBuilder")
-
         val slice = tempStringBuilder.substring(0, kotlin.math.min(count, tempStringBuilder.length))
         tempStringBuilder = StringBuilder(slice.length).append(tempStringBuilder.substring(slice.length))
 
         out.append(slice)
         return slice.length
     }
+
+    private fun bufferUp(): Int {
+        var totalReadCount = 0
+        while (buffer.availableRead < temp.size) {
+            val readCount = stream.read(temp)
+            if (readCount <= 0) break
+            totalReadCount += readCount
+            buffer.write(temp, 0, readCount)
+        }
+
+        return totalReadCount
+    }
+
+    companion object {
+        const val DEFAULT_CHUNK_SIZE = 1024
+        const val MIN_CHUNK_SIZE = 8
+    }
 }

korge-core/test/korlibs/io/stream/CharReaderTest.kt

@@ -9,4 +9,43 @@ class CharReaderTest {
         val reader = "áéíóúñ".toByteArray(UTF8).toCharReader(UTF8)
         assertEquals("á,éí,óúñ", listOf(reader.read(1), reader.read(2), reader.read(10)).joinToString(","))
     }
+
+
+    @Test
+    fun test2() {
+        val reader = "áéíóúñ".repeat(10).toByteArray(UTF8).toCharReader(charset = UTF8, chunkSize = 8)
+        assertEquals("á,éí,óúñ", listOf(reader.read(1), reader.read(2), reader.read(3)).joinToString(","))
+
+        assertFailsWith<IllegalArgumentException> { "áéíóúñ".toByteArray(UTF8).toCharReader(charset = UTF8, chunkSize = CharReaderFromSyncStream.MIN_CHUNK_SIZE -1) }
+
+    }
+
+    @Test
+    fun testCharReaderWithRandomStrings() {
+        val randomStrings = listOf(
+            "abcdefghijklm",
+            "ä<a>ä</a>",
+            "©頷ӨҤもタ編倏病Ҿ0沑âチ麕üӨҀ🙌とガӃ🙄Ҥzせø觧ҥ",
+            "Ђヹ肯みцë匓ンê😺り磬バëӇØ゚琫ら儂脸😨D亢JZEÕキ燗😨🙉ュӼ`ぺ",
+            "😳捇😗Ҧヿィ😲😵SӚåぐルҩS😯yѹ=ӪӠÀrキえÄ🙎へ¶Mたじ😞冃😃a\\xa0ÙҒ樗ボ😨",
+            "らーçウごネ粦😓õ姗ӏ(",
+            "Sm糛҆Òう楢ょ😽ê.",
+            "X5O!P%@AP[4\\PZX54(P^)7CC)7}\$EICAR-STANDARD-ANTIVIRUS-TEST-FILE!\$H+H*"
+        )
+        (1..30).forEach { readCount ->
+            randomStrings.forEach { inputData ->
+                val dataSegments: List<String> = inputData.splitInChunks(readCount)
+
+                for (chunkSize in 8 until 2000) {
+                    val charReader = inputData.openSync().toCharReader(charset = Charsets.UTF8, chunkSize = chunkSize)
+                    dataSegments.forEach { data ->
+                        val strBuilder = StringBuilder()
+                        assertEquals(data.length, charReader.read(strBuilder, readCount))
+                        assertEquals(data, strBuilder.toString())
+                    }
+                    assertEquals("", charReader.read(1)) // stream is empty all chars already read
+                }
+            }
+        }
+    }
 }