#792 Implement lazy evaluated encoders for all single character code pages.

Author: yruslan

cobol-parser/src/main/scala/za/co/absa/cobrix/cobol/parser/encoding/codepage/CodePage1144.scala

@@ -22,7 +22,7 @@ package za.co.absa.cobrix.cobol.parser.encoding.codepage
   * It corresponds to code page 280 and only differs from it in position 9F, where the euro sign € is located instead
   * of the international currency symbol ¤.
   */
-class CodePage1144 extends SingleByteCodePage(CodePage1144.ebcdicToAsciiMapping, Some(CodePage1144.asciiToEbcdicMapping)) {
+class CodePage1144 extends SingleByteCodePage(CodePage1144.ebcdicToAsciiMapping) {
   override def codePageShortName: String = "cp1144"
 }
 

cobol-parser/src/main/scala/za/co/absa/cobrix/cobol/parser/encoding/codepage/CodePageCommon.scala

@@ -21,7 +21,7 @@ package za.co.absa.cobrix.cobol.parser.encoding.codepage
   *
   * It is an "invariant" subset of EBCDIC. Each converted symbol should be present in all EBCDIC pages.
   */
-class CodePageCommon extends SingleByteCodePage(CodePageCommon.ebcdicToAsciiMapping, Some(CodePageCommon.asciiToEbcdicMapping)) {
+class CodePageCommon extends SingleByteCodePage(CodePageCommon.ebcdicToAsciiMapping) {
   override def codePageShortName: String = "common"
 }
 

cobol-parser/src/main/scala/za/co/absa/cobrix/cobol/parser/encoding/codepage/SingleByteCodePage.scala

@@ -21,7 +21,7 @@ import java.util
 /**
   * The base class for all single-byte EBCDIC decoders.
   */
-abstract class SingleByteCodePage(ebcdicToAsciiMapping: Array[Char], asciiToEbcdicMapping: Option[Int => Byte]=None)
+abstract class SingleByteCodePage(ebcdicToAsciiMapping: Array[Char])
   extends CodePage {
   private val ConversionTableElements = 256
   private val conversionTable = ebcdicToAsciiMapping
@@ -34,7 +34,7 @@ abstract class SingleByteCodePage(ebcdicToAsciiMapping: Array[Char], asciiToEbcd
   /**
     * Decodes bytes encoded as single byte EBCDIC code page to string.
     */
-  final def convert(bytes: Array[Byte]): String = {
+  final override def convert(bytes: Array[Byte]): String = {
     var i = 0
     val buf = new StringBuffer(bytes.length)
     while (i < bytes.length) {
@@ -51,10 +51,8 @@ abstract class SingleByteCodePage(ebcdicToAsciiMapping: Array[Char], asciiToEbcd
    * @param length          The length of the output (in bytes)
    * @return A string representation of the binary data
    */
-  def convert(string: String, length: Int): Array[Byte] = {
+  final override def convert(string: String, length: Int): Array[Byte] = {
     require(length >= 0, s"Field length cannot be negative, got $length")
-    require(asciiToEbcdicMapping.isDefined, s"Cannot encode strings for Code Page without ASCII to EBCDIC " +
-      s"mapping ${this.getClass.getSimpleName}")
 
     var i = 0
     val buf = new Array[Byte](length)
@@ -63,12 +61,22 @@ abstract class SingleByteCodePage(ebcdicToAsciiMapping: Array[Char], asciiToEbcd
     util.Arrays.fill(buf, 0x40.toByte)
 
     while (i < string.length && i < length) {
-      val unicodeCodePoint: Int = string.codePointAt(i)
-      buf(i) = asciiToEbcdicMapping.get(unicodeCodePoint)
+      buf(i) = reverseEbcdicToAsciiMapping.getOrElse(string.charAt(i), 0x40.toByte)
       i = i + 1
     }
     buf
   }
 
-  override def supportsEncoding: Boolean = asciiToEbcdicMapping.isDefined
+  override def supportsEncoding: Boolean = true
+
+  lazy val reverseEbcdicToAsciiMapping: Map[Char, Byte] = {
+    val map = scala.collection.mutable.Map[Char, Byte]()
+    for (i <- ebcdicToAsciiMapping.indices) {
+      val asciiChar = ebcdicToAsciiMapping(i)
+      if (!map.contains(asciiChar)) {
+        map(asciiChar) = i.toByte
+      }
+    }
+    map.toMap
+  }
 }

cobol-parser/src/test/scala/za/co/absa/cobrix/cobol/parser/decoders/StringDecodersSpec.scala

@@ -251,9 +251,29 @@ class StringDecodersSpec extends AnyWordSpec {
         assert(actual == expected)
       }
 
+      "decode a CP1025 string special characters" in {
+        val expectedUnicode = "АБВГДЕЖЗИЙКЛМНОПРСТУФХЦЧШЩЪЫЬЭЮЯабвгдежзийклмнопрстуфхцчшщъыьэюяABCDEFGHIJKLMNOPQRSTUVWXYZ123456789[]\r\n"
+        val ebcdicBytes = Array(
+          0xB9, 0xBA, 0xED, 0xBF, 0xBC, 0xBD, 0xEC, 0xFA, 0xCB, 0xCC, 0xCD, 0xCE, 0xCF, 0xDA, 0xDB, 0xDC,
+          0xDE, 0xDF, 0xEA, 0xEB, 0xBE, 0xCA, 0xBB, 0xFE, 0xFB, 0xFD, 0x57, 0xEF, 0xEE, 0xFC, 0xB8, 0xDD,
+          0x77, 0x78, 0xAF, 0x8D, 0x8A, 0x8B, 0xAE, 0xB2, 0x8F, 0x90, 0x9A, 0x9B, 0x9C, 0x9D, 0x9E, 0x9F,
+          0xAA, 0xAB, 0xAC, 0xAD, 0x8C, 0x8E, 0x80, 0xB6, 0xB3, 0xB5, 0xB7, 0xB1, 0xB0, 0xB4, 0x76, 0xA0,
+          0xC1, 0xC2, 0xC3, 0xC4, 0xC5, 0xC6, 0xC7, 0xC8, 0xC9, 0xD1, 0xD2, 0xD3, 0xD4, 0xD5, 0xD6, 0xD7,
+          0xD8, 0xD9, 0xE2, 0xE3, 0xE4, 0xE5, 0xE6, 0xE7, 0xE8, 0xE9, 0xF1, 0xF2, 0xF3, 0xF4, 0xF5, 0xF6,
+          0xF7, 0xF8, 0xF9, 0x4A, 0x5A, 0x25, 0x0D,
+        ).map(_.toByte)
+
+        val enc = new CodePage1025
+        val actualUnicode = decodeEbcdicString(ebcdicBytes, KeepAll, enc, improvedNullDetection = false)
+        val actualEbcdicBytes = enc.convert(expectedUnicode, expectedUnicode.length)
+
+        assert(actualUnicode == expectedUnicode)
+        assert(actualEbcdicBytes.sameElements(ebcdicBytes))
+      }
+
       "decode a CP1140 string special characters" in {
-        val expected = "âäàáãåçñ¢.<(+|&éêëèíîïìß!$*);¬-/ÂÄÀÁÃÅÇѦ,%_>?øÉÊËÈÍÎÏÌ`:#@'=\"Øabcdefghi«»ðýþ±°jklmnopqrªºæ¸Æ€µ~stuvwxyz¡¿ÐÝÞ®^£¥·©§¶¼½¾[]¯¨´×{ABCDEFGHI\u00ADôöòóõ}JKLMNOPQR¹ûüùúÿ\\÷STUVWXYZ²ÔÖÒÓÕ0123456789³ÛÜÙÚ"
-        val bytes = Array(
+        val expectedUnicode = "âäàáãåçñ¢.<(+|&éêëèíîïìß!$*);¬-/ÂÄÀÁÃÅÇѦ,%_>?øÉÊËÈÍÎÏÌ`:#@'=\"Øabcdefghi«»ðýþ±°jklmnopqrªºæ¸Æ€µ~stuvwxyz¡¿ÐÝÞ®^£¥·©§¶¼½¾[]¯¨´×{ABCDEFGHI\u00ADôöòóõ}JKLMNOPQR¹ûüùúÿ\\÷STUVWXYZ²ÔÖÒÓÕ0123456789³ÛÜÙÚ"
+        val ebcdicBytes = Array(
                       0x42, 0x43, 0x44, 0x45, 0x46, 0x47, 0x48, 0x49, 0x4A, 0x4B, 0x4C, 0x4D, 0x4E, 0x4F,
           0x50, 0x51, 0x52, 0x53, 0x54, 0x55, 0x56, 0x57, 0x58, 0x59, 0x5A, 0x5B, 0x5C, 0x5D, 0x5E, 0x5F,
           0x60, 0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67, 0x68, 0x69, 0x6A, 0x6B, 0x6C, 0x6D, 0x6E, 0x6F,
@@ -268,9 +288,12 @@ class StringDecodersSpec extends AnyWordSpec {
           0xF0, 0xF1, 0xF2, 0xF3, 0xF4, 0xF5, 0xF6, 0xF7, 0xF8, 0xF9, 0xFA, 0xFB, 0xFC, 0xFD, 0xFE
         ).map(_.toByte)
 
-        val actual = decodeEbcdicString(bytes, KeepAll, new CodePage1140, improvedNullDetection = false)
+        val enc = new CodePage1140
+        val actualUnicode = decodeEbcdicString(ebcdicBytes, KeepAll, enc, improvedNullDetection = false)
+        val actualEbcdicBytes = enc.convert(expectedUnicode, expectedUnicode.length)
 
-        assert(actual == expected)
+        assert(actualUnicode == expectedUnicode)
+        assert(actualEbcdicBytes.sameElements(ebcdicBytes))
       }
 
       "decode a CP1141 string special characters" in {