WI #2632 Use correct default encoding for alphanumeric literals writt…

…en using hex notation (#2633) * WI #2632 Use correct default encoding for alphanumeric literals written using hex notation * WI #2632 Use CODEPAGE option to get encoding for alphanumeric literals
TypeCobolTeam · Jul 9, 2024 · 63cb8f7 · 63cb8f7
1 parent f568ebe
commit 63cb8f7
Show file tree

Hide file tree

Showing 13 changed files with 89 additions and 47 deletions.
diff --git a/TypeCobol.LanguageServer/Workspace.cs b/TypeCobol.LanguageServer/Workspace.cs
@@ -212,7 +212,8 @@ internal void BindFileCompilerSourceTextDocument(DocumentContext docContext, str
             StopDocumentBackgroundCompilation(docContext);
             CompilationProject compilationProject = docContext.Project.Project;
             string fileName = Path.GetFileName(docContext.Uri.LocalPath);
-            ITextDocument initialTextDocumentLines = new ReadOnlyTextDocument(fileName, Configuration.Format.Encoding,
+            var encodingForAlphanumericLiterals = compilationProject.CompilationOptions.GetEncodingForAlphanumericLiterals();
+            ITextDocument initialTextDocumentLines = new ReadOnlyTextDocument(fileName, encodingForAlphanumericLiterals,
                 Configuration.Format.ColumnsLayout, docContext.IsCopy, sourceText);
             FileCompiler fileCompiler = null;
 

diff --git a/TypeCobol.Test/Parser/CodeElements/DISPLAY.CodeElements.txt b/TypeCobol.Test/Parser/CodeElements/DISPLAY.CodeElements.txt
@@ -93,7 +93,7 @@ Line 46[24,27] <27, Error, Syntax> - Syntax error : mismatched input 'Var1' expe
 - OutputDeviceName = SYSOUT
 - WithNoAdvancing
 
-[[DisplayStatement]] [12,18:display]<DISPLAY> --> [24,28+:X'40']<HexadecimalAlphanumericLiteral>(',Y,Y){@}
+[[DisplayStatement]] [12,18:display]<DISPLAY> --> [24,28+:X'40']<HexadecimalAlphanumericLiteral>(',Y,Y){ }
 - variables = X'40'
 
 [[DisplayStatement]] [12,18:display]<DISPLAY> --> [24,33:HIGH-VALUE]<HIGH_VALUE>

diff --git a/TypeCobol.Test/Parser/FileFormat/TestCobolFile.cs b/TypeCobol.Test/Parser/FileFormat/TestCobolFile.cs
@@ -1,15 +1,19 @@
-using System;
-using System.IO;
-using System.Linq;
-using System.Text;
+using System.Text;
 using TypeCobol.Compiler;
 using TypeCobol.Compiler.File;
 using TypeCobol.Compiler.Text;
 using TypeCobol.Test.Parser.Scanner;
 
 namespace TypeCobol.Test.Parser.FileFormat
 {
-    static class TestCobolFile {
+    static class TestCobolFile
+    {
+        private static readonly Encoding _EncodingForAlphanumericLiterals =
+#if EUROINFO_RULES
+            IBMCodePages.GetDotNetEncodingFromIBMCCSID(1147);
+#else
+            IBMCodePages.GetDotNetEncodingFromIBMCCSID(1140);
+#endif
         public static readonly string SampleFolder = "Parser" + Path.DirectorySeparatorChar + "FileFormat" + Path.DirectorySeparatorChar + "Samples";
         public static void Check_EBCDICCobolFile()
         {
@@ -27,7 +31,7 @@ public static void Check_EBCDICCobolFile()
             if (fileProvider.TryGetFile("EbcdicRefFormat", out cobolFile))
             {
                 // Load the CobolFile in a TextDocument
-                ReadOnlyTextDocument textDocument = new ReadOnlyTextDocument("EbcdicRefFormat.TXT", docFormat.Encoding, docFormat.ColumnsLayout, false, cobolFile.ReadChars());
+                ReadOnlyTextDocument textDocument = new ReadOnlyTextDocument("EbcdicRefFormat.TXT", _EncodingForAlphanumericLiterals, docFormat.ColumnsLayout, false, cobolFile.ReadChars());
                 // Send all text lines in one batch to the test observer
                 textDocument.TextChanged += textSourceListener.OnTextChanged;
                 textDocument.StartSendingChangeEvents();
@@ -84,7 +88,7 @@ public static void Check_EBCDICCobolFileWithUnsupportedChar()
                 try
                 {
                     // Load the CobolFile in a TextDocument;
-                    ReadOnlyTextDocument textDocument = new ReadOnlyTextDocument("EbcdicRefFormatWithBadChars.TXT", docFormat.Encoding, docFormat.ColumnsLayout, false, cobolFile.ReadChars());
+                    ReadOnlyTextDocument textDocument = new ReadOnlyTextDocument("EbcdicRefFormatWithBadChars.TXT", _EncodingForAlphanumericLiterals, docFormat.ColumnsLayout, false, cobolFile.ReadChars());
                 }
                 catch(Exception e)
                 {
@@ -117,7 +121,7 @@ public static void Check_ASCIICobolFile_ReferenceFormat()
             if (fileProvider.TryGetFile("AsciiRefFormat", out cobolFile))
             {
                 // Load the CobolFile in a TextDocument
-                ReadOnlyTextDocument textDocument = new ReadOnlyTextDocument("MSVCOUT.cpy", docFormat.Encoding, docFormat.ColumnsLayout, true, cobolFile.ReadChars());
+                ReadOnlyTextDocument textDocument = new ReadOnlyTextDocument("MSVCOUT.cpy", _EncodingForAlphanumericLiterals, docFormat.ColumnsLayout, true, cobolFile.ReadChars());
                 // Send all text lines in one batch to the test observer
                 textDocument.TextChanged += textSourceListener.OnTextChanged;
                 textDocument.StartSendingChangeEvents();
@@ -172,7 +176,7 @@ public static void Check_ASCIICobolFile_LinuxReferenceFormat()
             if (fileProvider.TryGetFile("AsciiLinuxFormat.14", out cobolFile))
             {
                 // Load the CobolFile in a TextDocument
-                ReadOnlyTextDocument textDocument = new ReadOnlyTextDocument("AsciiLinuxFormat.14", docFormat.Encoding, docFormat.ColumnsLayout, false, cobolFile.ReadChars());
+                ReadOnlyTextDocument textDocument = new ReadOnlyTextDocument("AsciiLinuxFormat.14", _EncodingForAlphanumericLiterals, docFormat.ColumnsLayout, false, cobolFile.ReadChars());
                 // Send all text lines in one batch to the test observer
                 textDocument.TextChanged += textSourceListener.OnTextChanged;
                 textDocument.StartSendingChangeEvents();
@@ -227,7 +231,7 @@ public static void Check_ASCIICobolFile_FreeTextFormat()
             if (fileProvider.TryGetFile("AsciiFreeFormat", out cobolFile))
             {
                 // Load the CobolFile in a TextDocument
-                ReadOnlyTextDocument textDocument = new ReadOnlyTextDocument("AsciiFreeFormat.cpy", docFormat.Encoding, docFormat.ColumnsLayout, true, cobolFile.ReadChars());
+                ReadOnlyTextDocument textDocument = new ReadOnlyTextDocument("AsciiFreeFormat.cpy", _EncodingForAlphanumericLiterals, docFormat.ColumnsLayout, true, cobolFile.ReadChars());
                 // Send all text lines in one batch to the test observer
                 textDocument.TextChanged += textSourceListener.OnTextChanged;
                 textDocument.StartSendingChangeEvents();
@@ -307,7 +311,7 @@ public static void Check_UTF8File()
             if (fileProvider.TryGetFile(filename, out cobolFile))
             {
                 // Load the CobolFile in a TextDocument
-                ReadOnlyTextDocument textDocument = new ReadOnlyTextDocument(filename, docFormat.Encoding, docFormat.ColumnsLayout, false, cobolFile.ReadChars());
+                ReadOnlyTextDocument textDocument = new ReadOnlyTextDocument(filename, _EncodingForAlphanumericLiterals, docFormat.ColumnsLayout, false, cobolFile.ReadChars());
                 // Send all text lines in one batch to the test observer
                 textDocument.TextChanged += textSourceListener.OnTextChanged;
                 textDocument.StartSendingChangeEvents();

diff --git a/TypeCobol.Test/Parser/Scanner/ResultFiles/AlphanumericLiterals4-1140.txt b/TypeCobol.Test/Parser/Scanner/ResultFiles/AlphanumericLiterals4-1140.txt
@@ -0,0 +1,3 @@
+-- Line 1 --
+[1,9+:X'C085D0']<HexadecimalAlphanumericLiteral>(',Y,Y){{e}}
+
diff --git a/TypeCobol.Test/Parser/Scanner/ResultFiles/AlphanumericLiterals4-1147.txt b/TypeCobol.Test/Parser/Scanner/ResultFiles/AlphanumericLiterals4-1147.txt
@@ -0,0 +1,3 @@
+-- Line 1 --
+[1,9+:X'C085D0']<HexadecimalAlphanumericLiteral>(',Y,Y){éeè}
+
diff --git a/TypeCobol.Test/Parser/Scanner/ScannerUtils.cs b/TypeCobol.Test/Parser/Scanner/ScannerUtils.cs
@@ -1,6 +1,4 @@
-using System.Collections.Generic;
-using System.IO;
-using System.Text;
+using System.Text;
 using TypeCobol.Compiler.Concurrency;
 using TypeCobol.Compiler.Diagnostics;
 using TypeCobol.Compiler.Directives;
@@ -44,8 +42,8 @@ public TextChangeMap(TextChange change, ColumnsLayout columnsLayout)
 
     internal static class ScannerUtils
     {
-        public static TextSourceInfo TextSourceInfo = new TextSourceInfo("test", IBMCodePages.GetDotNetEncodingFromIBMCCSID(1147), ColumnsLayout.FreeTextFormat, false);//Assuming a program here, not a copy.
         public static TypeCobolOptions CompilerOptions = new TypeCobolOptions();
+        public static TextSourceInfo TextSourceInfo = new TextSourceInfo("test", CompilerOptions.GetEncodingForAlphanumericLiterals(), ColumnsLayout.FreeTextFormat, false);//Assuming a program here, not a copy.
         public static List<RemarksDirective.TextNameVariation> CopyTextNameVariations = new List<RemarksDirective.TextNameVariation>();
 
         public static string ScanLine(string testLine)

diff --git a/TypeCobol.Test/Parser/Scanner/TestTokenTypes.cs b/TypeCobol.Test/Parser/Scanner/TestTokenTypes.cs
@@ -1,5 +1,4 @@
-using System;
-using TypeCobol.Compiler.Scanner;
+using TypeCobol.Compiler.Scanner;
 
 namespace TypeCobol.Test.Parser.Scanner
 {
@@ -139,6 +138,15 @@ public static void CheckAlphanumericLiterals()
             result = ScannerUtils.ScanLines(testLines);
             ScannerUtils.CheckWithResultFile(result, testName);
 
+#if EUROINFO_RULES
+            testName = "AlphanumericLiterals4-1147";
+#else
+            testName = "AlphanumericLiterals4-1140";
+#endif
+            testLines = new string[] { "X'C085D0'" }; // 'éeè' in EBCDIC 1147, '{e}' in EBCDIC 1140
+            result = ScannerUtils.ScanLines(testLines);
+            ScannerUtils.CheckWithResultFile(result, testName);
+
             testName = "UTF8Literals";
             testLines = new string[] {
                 @"U""This text does not include any escaped char""",

diff --git a/TypeCobol.Test/Parser/TestParserRobustness.cs b/TypeCobol.Test/Parser/TestParserRobustness.cs
@@ -1,9 +1,4 @@
-using System;
-using System.Collections.Generic;
-using System.Linq;
-using System.Text;
-using System.Threading.Tasks;
-using TypeCobol.Compiler;
+using TypeCobol.Compiler;
 using TypeCobol.Compiler.CodeElements;
 using TypeCobol.Compiler.Diagnostics;
 using TypeCobol.Compiler.Directives;
@@ -22,11 +17,11 @@ public static void CheckProgramCodeElements()
         private static CodeElement[] ParseCodeElements(string cobolString, bool asPartOfACopy, out Diagnostic[] parserDiagnostics)
         {
             // Load text document from string
-            var textDocument = new ReadOnlyTextDocument("test string", Encoding.Default, ColumnsLayout.FreeTextFormat, asPartOfACopy, string.Empty);
+            var typeCobolOptions = new TypeCobolOptions();
+            var textDocument = new ReadOnlyTextDocument("test string", typeCobolOptions.GetEncodingForAlphanumericLiterals(), ColumnsLayout.FreeTextFormat, asPartOfACopy, string.Empty);
             textDocument.LoadChars(cobolString);
 
             // Create a compilation project and a compiler for this document
-            var typeCobolOptions = new TypeCobolOptions();
             var project = new CompilationProject("test project", ".", new[] { ".cbl", ".cpy" },
                 DocumentFormat.FreeTextFormat, typeCobolOptions, null);
             var compiler = new FileCompiler(textDocument, project.SourceFileProvider, project, typeCobolOptions, project);

diff --git a/TypeCobol.Test/Parser/Text/TestReadOnlyTextDocument.cs b/TypeCobol.Test/Parser/Text/TestReadOnlyTextDocument.cs
@@ -1,8 +1,6 @@
-using System;
-using System.IO;
-using System.Linq;
-using System.Text;
+using System.Text;
 using TypeCobol.Compiler;
+using TypeCobol.Compiler.Directives;
 using TypeCobol.Compiler.File;
 using TypeCobol.Compiler.Text;
 
@@ -49,7 +47,8 @@ public static void Check_DocumentFormatExceptions()
 
         public static void Check_EmptyDocument()
         {
-            ReadOnlyTextDocument textDocument = new ReadOnlyTextDocument("empty", Encoding.Default, ColumnsLayout.CobolReferenceFormat, false, String.Empty);
+            var options = new TypeCobolOptions();
+            ReadOnlyTextDocument textDocument = new ReadOnlyTextDocument("empty", options.GetEncodingForAlphanumericLiterals(), ColumnsLayout.CobolReferenceFormat, false, String.Empty);
 
             Exception resultException = null;
             try
@@ -176,7 +175,8 @@ public static void Check_ReferenceFormatDocument()
             }
 
             // Load the CobolFile in a TextDocument
-            ReadOnlyTextDocument textDocument = new ReadOnlyTextDocument("MSVCOUT.cpy", docFormat.Encoding, docFormat.ColumnsLayout, true, cobolFile.ReadChars());
+            var options = new TypeCobolOptions();
+            ReadOnlyTextDocument textDocument = new ReadOnlyTextDocument("MSVCOUT.cpy", options.GetEncodingForAlphanumericLiterals(), docFormat.ColumnsLayout, true, cobolFile.ReadChars());
 
             if(textDocument.CharAt(0) != '0')
             {
@@ -269,7 +269,8 @@ public static void Check_FreeFormatDocument()
             }
 
             // Load the CobolFile in a TextDocument
-            ReadOnlyTextDocument textDocument = new ReadOnlyTextDocument("MSVCINP free format.cpy", docFormat.Encoding, docFormat.ColumnsLayout, true, cobolFile.ReadChars());
+            var options = new TypeCobolOptions();
+            ReadOnlyTextDocument textDocument = new ReadOnlyTextDocument("MSVCINP free format.cpy", options.GetEncodingForAlphanumericLiterals(), docFormat.ColumnsLayout, true, cobolFile.ReadChars());
 
             if (textDocument.CharAt(0) != '/')
             {

diff --git a/TypeCobol.Test/Utils/ParserUtils.cs b/TypeCobol.Test/Utils/ParserUtils.cs
@@ -1,8 +1,4 @@
-using System;
-using System.Collections.Generic;
-using System.IO;
-using System.Linq;
-using System.Text;
+using System.Text;
 using Antlr4.Runtime;
 using TypeCobol.Compiler;
 using TypeCobol.Compiler.AntlrUtils;
@@ -60,10 +56,10 @@ public static CompilationUnit ParseCobolFile(string textName, string folder, boo
         public static CompilationUnit ParseCobolString(string cobolString, bool asPartOfACopy)
         {
             //Prepare
-            var textDocument = new ReadOnlyTextDocument("Empty doc", Encoding.Default, ColumnsLayout.FreeTextFormat, asPartOfACopy, string.Empty);
+            var typeCobolOptions = new TypeCobolOptions();
+            var textDocument = new ReadOnlyTextDocument("Empty doc", typeCobolOptions.GetEncodingForAlphanumericLiterals(), ColumnsLayout.FreeTextFormat, asPartOfACopy, string.Empty);
             textDocument.LoadChars(cobolString);
 
-            var typeCobolOptions = new TypeCobolOptions();
             var project = new CompilationProject("Empty project", ".", new[] { ".cbl", ".cpy" },
                 DocumentFormat.FreeTextFormat, typeCobolOptions, null);
 

diff --git a/TypeCobol/Compiler/Directives/IBMCompilerOptions.cs b/TypeCobol/Compiler/Directives/IBMCompilerOptions.cs
@@ -1,6 +1,8 @@
 #nullable enable
 
 using System.Diagnostics.CodeAnalysis;
+using System.Text;
+using TypeCobol.Compiler.File;
 
 namespace TypeCobol.Compiler.Directives
 {
@@ -137,7 +139,14 @@ internal IBMCompilerOptionStatus(IBMCompilerOptionName name)
                 case IBMCompilerOptionName.BLOCK0: IsActivated = false; Value = null; break;
                 case IBMCompilerOptionName.BUFSIZE: IsActivated = true; Value = "4096"; break;
                 case IBMCompilerOptionName.CICS: IsActivated = false; Value = null; break;
-                case IBMCompilerOptionName.CODEPAGE: IsActivated = true; Value = "1140"; break;
+                case IBMCompilerOptionName.CODEPAGE:
+                    IsActivated = true;
+#if EUROINFO_RULES
+                    Value = "1147"; //IBM EBCDIC (France-Euro)
+#else
+                    Value = "1140"; //IBM EBCDIC (EU-Canada-Euro)
+#endif
+                    break;
                 case IBMCompilerOptionName.COMPILE: IsActivated = false; Value = "S"; break;
                 case IBMCompilerOptionName.COPYLOC: IsActivated = false; Value = null; break;
                 case IBMCompilerOptionName.CURRENCY: IsActivated = false; Value = null; break;
@@ -1015,4 +1024,29 @@ public enum IBMCompilerOptionName
         /* If you compile using ZWB, the compiler removes the sign from a signed zoned decimal (DISPLAY) field before comparing this field to an alphanumeric elementary field during execution. */
         ZWB
     }
+
+    public static class IBMCompilerOptionsExtensions
+    {
+        /// <summary>
+        /// Get from the CODEPAGE compiler option:
+        /// • The encoding of literals in the source program
+        /// • The encoding for data items described with USAGE DISPLAY or DISPLAY-1
+        /// • The encoding for XML parsing and XML generation
+        ///
+        /// The encoding of national and UTF-8 data is not affected by the CODEPAGE compiler option. The encoding
+        /// for national literals and data items described with usage NATIONAL is UTF-16BE (big endian), CCSID
+        /// 1200. A reference to UTF-16 in this document is a reference to UTF-16BE. The encoding for UTF-8 literals
+        /// and data items described with usage UTF-8 is UTF-8, CCSID 1208.
+        /// </summary>
+        public static Encoding GetEncodingForAlphanumericLiterals(this IBMCompilerOptions ibmCompilerOptions)
+        {
+            string? codePageOption = ibmCompilerOptions.CODEPAGE.Value;
+            if (int.TryParse(codePageOption, out int codePage))
+            {
+                return IBMCodePages.GetDotNetEncodingFromIBMCCSID(codePage);
+            }
+
+            throw new ArgumentException($"Invalid CODEPAGE compiler option: '{codePageOption}'.");
+        }
+    }
 }
diff --git a/TypeCobol/Compiler/FileCompiler.cs b/TypeCobol/Compiler/FileCompiler.cs
@@ -1,5 +1,4 @@
-using System;
-using System.Diagnostics;
+using System.Diagnostics;
 using JetBrains.Annotations;
 using TypeCobol.Compiler.CodeModel;
 using TypeCobol.Compiler.Directives;
@@ -165,7 +164,8 @@ private FileCompiler(Tuple<string, string, ColumnsLayout, bool> fileInfo, ITextD
             {
                 // 2.a Load it in a new text document in memory
                 Debug.Assert(sourceFile != null);
-                TextDocument = new ReadOnlyTextDocument(sourceFile.Name, sourceFile.Encoding, fileInfo.Item3, fileInfo.Item4, sourceFile.ReadChars());
+                var encodingForAlphanumericLiterals = compilerOptions.GetEncodingForAlphanumericLiterals();
+                TextDocument = new ReadOnlyTextDocument(sourceFile.Name, encodingForAlphanumericLiterals, fileInfo.Item3, fileInfo.Item4, sourceFile.ReadChars());
             }
             else
             {

diff --git a/TypeCobol/Compiler/Text/CobolTextLine.cs b/TypeCobol/Compiler/Text/CobolTextLine.cs
@@ -2,7 +2,6 @@
 
 using TypeCobol.Compiler.Concurrency;
 using TypeCobol.Compiler.Directives;
-using TypeCobol.Compiler.File;
 using TypeCobol.Compiler.Scanner;
 
 namespace TypeCobol.Compiler.Text
@@ -193,7 +192,7 @@ private static IList<Tuple<string, bool> > Split(string line, int max, int min,
                 }
             }
             TokensLine tempTokensLine = TokensLine.CreateVirtualLineForInsertedToken(0, line, layout);
-            tempTokensLine.InitializeScanState(new MultilineScanState(IBMCodePages.GetDotNetEncodingFromIBMCCSID(1147)));
+            tempTokensLine.InitializeScanState(new MultilineScanState(scannerOptions.GetEncodingForAlphanumericLiterals()));
 
             Scanner.Scanner scanner = new Scanner.Scanner(line, 0, line.Length - 1, tempTokensLine, scannerOptions, false);
             Token? t;
Original file line number	Diff line number	Diff line change
		@@ -0,0 +1,3 @@
		-- Line 1 --
		[1,9+:X'C085D0']<HexadecimalAlphanumericLiteral>(',Y,Y){{e}}