PETSCII does DCI

I didn't think it made sense, but I found something that used it,
so apparently it's a thing.  This updates the operand editor to
let you choose PETSCII+DCI, and updates the assemblers to handle
it correctly (really just 64tass, since the others either don't
have a DCI directive or don't deal with PETSCII at all).

Changed the char-encoding sample from "bad dcI" to "pet dcI", and
updated the documentation.

Asm65/CharEncoding.cs

@@ -217,6 +217,15 @@ namespace Asm65 {
         public static char ConvertC64Petscii(byte val) {
             return sPetsciiToUnicode[val];
         }
+        public static char ConvertLowAndHighC64Petscii(byte val) {
+            // This is an odd one.  Some programs use DCI with PETSCII, which means the
+            // string is allow lower case except for the last letteR.
+            //
+            // There's no such thing as "high PETSCII", in the same sense that ASCII or
+            // C64 screen codes have it, but I'm giving the method a similar name for
+            // the sake of consistency.
+            return ConvertC64Petscii((byte)(val & 0x7f));
+        }
 
         //
         // C64 Screen Codes

SourceGen/AsmGen/AsmAcme.cs

@@ -573,7 +573,7 @@ namespace SourceGen.AsmGen {
                 case FormatDescriptor.Type.StringReverse:
                 case FormatDescriptor.Type.StringNullTerm:
                 case FormatDescriptor.Type.StringDci:
-                    // Last byte will be output as hex.
+                    // Last byte may be output as hex.
                     break;
                 case FormatDescriptor.Type.StringL8:
                     // Length byte will be output as hex.

SourceGen/AsmGen/AsmTass64.cs

@@ -601,10 +601,7 @@ namespace SourceGen.AsmGen {
                 case FormatDescriptor.SubType.C64Petscii:
                     if (textMode == TextScanMode.C64Petscii) {
                         charConv = CharEncoding.ConvertC64Petscii;
-                        // DCI not supported for PETSCII; make sure it doesn't get tried
+                        dciConv = CharEncoding.ConvertLowAndHighC64Petscii;
-                        if (dfd.FormatType == FormatDescriptor.Type.StringDci) {
-                            charConv = null;
-                        }
                     }
                     break;
                 case FormatDescriptor.SubType.C64Screen:

SourceGen/DataAnalysis.cs

@@ -1044,7 +1044,8 @@ namespace SourceGen {
         /// that all strings have the same hi/lo pattern.
         /// </summary>
         /// <remarks>
-        /// Not useful for C64Petscii, which mixes high/low characters.
+        /// For C64Petscii, this will identify strings that are entirely in lower case except
+        /// for the last letteR, or vice-versa.
         /// </remarks>
         /// <param name="fileData">Raw data.</param>
         /// <param name="start">Offset of first byte in range.</param>

SourceGen/PseudoOp.cs

@@ -410,7 +410,11 @@ namespace SourceGen {
                     delDef = delSet.Get(CharEncoding.Encoding.HighAscii);
                     break;
                 case FormatDescriptor.SubType.C64Petscii:
-                    charConv = CharEncoding.ConvertC64Petscii;
+                    if (dfd.FormatType == FormatDescriptor.Type.StringDci) {
+                        charConv = CharEncoding.ConvertLowAndHighC64Petscii;
+                    } else {
+                        charConv = CharEncoding.ConvertC64Petscii;
+                    }
                     delDef = delSet.Get(CharEncoding.Encoding.C64Petscii);
                     break;
                 case FormatDescriptor.SubType.C64Screen:

SourceGen/RuntimeData/Help/editors.html

@@ -136,9 +136,8 @@ you change the encoding, your available options may change.  The
 low + high ASCII setting will accept both, configuring the appropriate
 encoding based on the data values, but when identifying multiple strings
 it requires that each individual string be entirely one or the other.</p>
-<p>Due to fundamental limitations of the character sets, C64 PETSCII
+<p>Due to fundamental limitations of the character set, C64 screen code
-does not support DCI, and C64 screen code strings cannot be null
+strings cannot be null terminated.</p>
-terminated.</p>
 
 <p>To avoid burying a label in the middle of a data item, contiguous areas
 are split at labels.  This can sometimes have unexpected effects.  For

SourceGen/SGTestData/2016-char-encoding-a.dis65

@@ -1,6 +1,6 @@
 ### 6502bench SourceGen dis65 v1.0 ###
 {
-"_ContentVersion":2,"FileDataLength":1093,"FileDataCrc32":589095233,"ProjectProps":{
+"_ContentVersion":2,"FileDataLength":1093,"FileDataCrc32":-1027700879,"ProjectProps":{
 "CpuName":"65816","IncludeUndocumentedInstr":false,"EntryFlags":32702671,"AutoLabelStyle":"Simple","AnalysisParams":{
 "AnalyzeUncategorizedData":true,"DefaultTextScanMode":"LowAscii","MinCharsForString":4,"SeekNearbyTargets":true},
 "PlatformSymbolFileIdentifiers":[],"ExtensionScriptFileIdentifiers":[],"ProjectSyms":{

SourceGen/SGTestData/2017-char-encoding-p.dis65

@@ -1,6 +1,6 @@
 ### 6502bench SourceGen dis65 v1.0 ###
 {
-"_ContentVersion":2,"FileDataLength":1093,"FileDataCrc32":589095233,"ProjectProps":{
+"_ContentVersion":2,"FileDataLength":1093,"FileDataCrc32":-1027700879,"ProjectProps":{
 "CpuName":"65816","IncludeUndocumentedInstr":false,"EntryFlags":32702671,"AutoLabelStyle":"Simple","AnalysisParams":{
 "AnalyzeUncategorizedData":true,"DefaultTextScanMode":"C64Petscii","MinCharsForString":4,"SeekNearbyTargets":true},
 "PlatformSymbolFileIdentifiers":[],"ExtensionScriptFileIdentifiers":[],"ProjectSyms":{

SourceGen/SGTestData/2018-char-encoding-s.dis65

@@ -1,6 +1,6 @@
 ### 6502bench SourceGen dis65 v1.0 ###
 {
-"_ContentVersion":2,"FileDataLength":1093,"FileDataCrc32":589095233,"ProjectProps":{
+"_ContentVersion":2,"FileDataLength":1093,"FileDataCrc32":-1027700879,"ProjectProps":{
 "CpuName":"65816","IncludeUndocumentedInstr":false,"EntryFlags":32702671,"AutoLabelStyle":"Simple","AnalysisParams":{
 "AnalyzeUncategorizedData":true,"DefaultTextScanMode":"C64ScreenCode","MinCharsForString":4,"SeekNearbyTargets":true},
 "PlatformSymbolFileIdentifiers":[],"ExtensionScriptFileIdentifiers":[],"ProjectSyms":{

SourceGen/SGTestData/Expected/2016-char-encoding-a_64tass.S

@@ -115,7 +115,7 @@
         .byte   $14,$00,$d0,$c5,$d4,$d3,$c3,$c9,$c9,$20,$57,$49,$54,$48,$20,$4c
         .byte   $45,$4e,$47,$54,$48,$32
         .byte   $84
-        .byte   $42,$41,$44,$20,$44,$43,$c9
+        .byte   $50,$45,$54,$20,$44,$43,$c9
         .byte   $84
         .byte   $05,$04,$0f,$43,$20,$0e,$05,$05,$12,$03,$53,$20,$05,$13,$12,$05
         .byte   $16,$05,$12

SourceGen/SGTestData/Expected/2016-char-encoding-a_Merlin32.S

@@ -90,7 +90,7 @@
          dfb   $84
          hex   1400d0c5d4d3c3c9c92057495448204c454e47544832
          dfb   $84
-         hex   424144204443c9
+         hex   504554204443c9
          dfb   $84
          hex   05040f43200e05051203532005131205160512
          dfb   $84

SourceGen/SGTestData/Expected/2016-char-encoding-a_acme.S

@@ -97,7 +97,7 @@
         !byte   $84
         !pet    $14,$00,"PETSCII with length2"
         !byte   $84
-        !pet    "bad dcI"
+        !pet    "pet dcI"
         !byte   $84
         !scr    "edoC neercS esrever"
         !byte   $84

SourceGen/SGTestData/Expected/2016-char-encoding-a_cc65.S

@@ -119,7 +119,7 @@
          .byte   $14,$00,$d0,$c5,$d4,$d3,$c3,$c9,$c9,$20,$57,$49,$54,$48,$20,$4c
          .byte   $45,$4e,$47,$54,$48,$32
          .byte   $84
-         .byte   $42,$41,$44,$20,$44,$43,$c9
+         .byte   $50,$45,$54,$20,$44,$43,$c9
          .byte   $84
          .byte   $05,$04,$0f,$43,$20,$0e,$05,$05,$12,$03,$53,$20,$05,$13,$12,$05
          .byte   $16,$05,$12

SourceGen/SGTestData/Expected/2017-char-encoding-p_64tass.S

@@ -113,7 +113,7 @@
         .byte   $84
         .text   $14,$00,"PETSCII with length2"
         .byte   $84
-        .byte   $42,$41,$44,$20,$44,$43,$c9
+        .shift  "pet dci"
         .byte   $84
         .byte   $05,$04,$0f,$43,$20,$0e,$05,$05,$12,$03,$53,$20,$05,$13,$12,$05
         .byte   $16,$05,$12

SourceGen/SGTestData/Expected/2017-char-encoding-p_Merlin32.S

@@ -96,7 +96,7 @@
          dfb   $84
          hex   1400d0c5d4d3c3c9c92057495448204c454e47544832
          dfb   $84
-         hex   424144204443c9
+         hex   504554204443c9
          dfb   $84
          hex   05040f43200e05051203532005131205160512
          dfb   $84

SourceGen/SGTestData/Expected/2017-char-encoding-p_acme.S

@@ -103,7 +103,7 @@
         !byte   $84
         !pet    $14,$00,"PETSCII with length2"
         !byte   $84
-        !pet    "bad dcI"
+        !pet    "pet dcI"
         !byte   $84
         !scr    "edoC neercS esrever"
         !byte   $84

SourceGen/SGTestData/Expected/2017-char-encoding-p_cc65.S

@@ -125,7 +125,7 @@
          .byte   $14,$00,$d0,$c5,$d4,$d3,$c3,$c9,$c9,$20,$57,$49,$54,$48,$20,$4c
          .byte   $45,$4e,$47,$54,$48,$32
          .byte   $84
-         .byte   $42,$41,$44,$20,$44,$43,$c9
+         .byte   $50,$45,$54,$20,$44,$43,$c9
          .byte   $84
          .byte   $05,$04,$0f,$43,$20,$0e,$05,$05,$12,$03,$53,$20,$05,$13,$12,$05
          .byte   $16,$05,$12

SourceGen/SGTestData/Expected/2018-char-encoding-s_64tass.S

@@ -122,7 +122,7 @@
         .byte   $14,$00,$d0,$c5,$d4,$d3,$c3,$c9,$c9,$20,$57,$49,$54,$48,$20,$4c
         .byte   $45,$4e,$47,$54,$48,$32
         .byte   $84
-        .byte   $42,$41,$44,$20,$44,$43,$c9
+        .byte   $50,$45,$54,$20,$44,$43,$c9
         .byte   $84
         .text   "edoC neercS esrever"
         .byte   $84

SourceGen/SGTestData/Expected/2018-char-encoding-s_Merlin32.S

@@ -96,7 +96,7 @@
          dfb   $84
          hex   1400d0c5d4d3c3c9c92057495448204c454e47544832
          dfb   $84
-         hex   424144204443c9
+         hex   504554204443c9
          dfb   $84
          hex   05040f43200e05051203532005131205160512
          dfb   $84

SourceGen/SGTestData/Expected/2018-char-encoding-s_acme.S

@@ -103,7 +103,7 @@
         !byte   $84
         !pet    $14,$00,"PETSCII with length2"
         !byte   $84
-        !pet    "bad dcI"
+        !pet    "pet dcI"
         !byte   $84
         !scr    "edoC neercS esrever"
         !byte   $84

SourceGen/SGTestData/Expected/2018-char-encoding-s_cc65.S

@@ -125,7 +125,7 @@
          .byte   $14,$00,$d0,$c5,$d4,$d3,$c3,$c9,$c9,$20,$57,$49,$54,$48,$20,$4c
          .byte   $45,$4e,$47,$54,$48,$32
          .byte   $84
-         .byte   $42,$41,$44,$20,$44,$43,$c9
+         .byte   $50,$45,$54,$20,$44,$43,$c9
          .byte   $84
          .byte   $05,$04,$0f,$43,$20,$0e,$05,$05,$12,$03,$53,$20,$05,$13,$12,$05
          .byte   $16,$05,$12

SourceGen/SGTestData/Source/2016-char-encoding.S

@@ -107,7 +107,7 @@
 
 ; The 2005 test exercises low/high ASCII strings, so no need to do that here.
 ; Do a quick test with C64 characters.  Note Screen Code can't be null-terminated
-; by definition, and PETSCII can't effectively be DCI.
+; by definition.
         !byte   $84
         !pet    "IICSTEP esrever"               ;format as StringReverse
         !byte   $84
@@ -119,7 +119,7 @@
         !byte   $84
         !pet    20,0,"PETSCII with length2"     ;format as StringL16
         !byte   $84
-        !pet    "bad dcI"                       ;EDIT: format as StringDCI
+        !pet    "pet dcI"                       ;format as StringDCI
 
         !byte   $84
         !scr    "edoC neercS esrever"           ;format as StringReverse

SourceGen/WpfGui/EditDataOperand.xaml.cs

@@ -405,7 +405,7 @@ namespace SourceGen.WpfGui {
             radioStringNullTerm.IsEnabled = (scanMode != TextScanMode.C64ScreenCode);
             radioStringLen8.IsEnabled = true;
             radioStringLen16.IsEnabled = true;
-            radioStringDci.IsEnabled = (scanMode != TextScanMode.C64Petscii);
+            radioStringDci.IsEnabled = true;
 
             IEnumerator<TypedRangeSet.TypedRange> iter = mSelection.RangeListIterator;
             while (iter.MoveNext()) {