Change the way ASCII is handled for 64tass

The documentation for 64tass says you're required to pass "--ascii" when the source file is ASCII (as opposed to PETSCII). We were ignoring this, but it turns out that everything works a bit better if we don't. So we now pass "--ascii" on the command line, and add a two-line character encoding definition to every file that is generated with ASCII as the default encoding. The sg_petscii and sg_screen encodings go away, as PETSCII is now the default, and we can use the built-in "screen" encoding.
2025-08-07 22:25:09 +00:00 · 2019-08-20 11:21:30 -07:00
parent 0abea2beac
commit 149e763821
26 changed files with 77 additions and 41 deletions
--- a/SourceGen/AsmGen/AsmTass64.cs
+++ b/SourceGen/AsmGen/AsmTass64.cs
@@ -279,26 +279,30 @@ namespace SourceGen.AsmGen {
            TextScanMode textMode = Project.ProjectProps.AnalysisParams.DefaultTextScanMode;
            switch (textMode) {
                case TextScanMode.C64Petscii:
-                    OutputLine(string.Empty, ".enc", "sg_petscii", string.Empty);
-                    OutputLine(string.Empty, ".cdef", "\" @\", $20", string.Empty);
-                    OutputLine(string.Empty, ".cdef", "\"AZ\", $c1", string.Empty);
-                    OutputLine(string.Empty, ".cdef", "\"az\", $41", string.Empty);
-                    OutputLine(string.Empty, ".cdef", "\"[[\", $5b", string.Empty);
-                    OutputLine(string.Empty, ".cdef", "\"]]\", $5d", string.Empty);
+                    // With "--ascii", this is the default.
+                    //OutputLine(string.Empty, ".enc", "sg_petscii", string.Empty);
+                    //OutputLine(string.Empty, ".cdef", "\" @\", $20", string.Empty);
+                    //OutputLine(string.Empty, ".cdef", "\"AZ\", $c1", string.Empty);
+                    //OutputLine(string.Empty, ".cdef", "\"az\", $41", string.Empty);
+                    //OutputLine(string.Empty, ".cdef", "\"[[\", $5b", string.Empty);
+                    //OutputLine(string.Empty, ".cdef", "\"]]\", $5d", string.Empty);
                    break;
                case TextScanMode.C64ScreenCode:
-                    OutputLine(string.Empty, ".enc", "sg_screen", string.Empty);
-                    OutputLine(string.Empty, ".cdef", "\" ?\", $20", string.Empty);
-                    OutputLine(string.Empty, ".cdef", "\"@@\", $00", string.Empty);
-                    OutputLine(string.Empty, ".cdef", "\"AZ\", $41", string.Empty);
-                    OutputLine(string.Empty, ".cdef", "\"az\", $01", string.Empty);
-                    OutputLine(string.Empty, ".cdef", "\"[[\", $1b", string.Empty);
-                    OutputLine(string.Empty, ".cdef", "\"]]\", $1d", string.Empty);
+                    // With "--ascii", we can use the built-in screen encoding.
+                    OutputLine(string.Empty, ".enc", "screen", string.Empty);
+                    //OutputLine(string.Empty, ".enc", "sg_screen", string.Empty);
+                    //OutputLine(string.Empty, ".cdef", "\" ?\", $20", string.Empty);
+                    //OutputLine(string.Empty, ".cdef", "\"@@\", $00", string.Empty);
+                    //OutputLine(string.Empty, ".cdef", "\"AZ\", $41", string.Empty);
+                    //OutputLine(string.Empty, ".cdef", "\"az\", $01", string.Empty);
+                    //OutputLine(string.Empty, ".cdef", "\"[[\", $1b", string.Empty);
+                    //OutputLine(string.Empty, ".cdef", "\"]]\", $1d", string.Empty);
                    break;
                case TextScanMode.LowAscii:
                case TextScanMode.LowHighAscii:
                default:
-                    // ASCII in, ASCII out
+                    OutputLine(string.Empty, ".enc", "sg_ascii", string.Empty);
+                    OutputLine(string.Empty, ".cdef", "$20,$7e,$20", string.Empty);
                    break;
            }
        }
@@ -718,11 +722,12 @@ namespace SourceGen.AsmGen {
    /// Cross-assembler execution interface.
    /// </summary>
    public class AsmTass64 : IAssembler {
-        // Standard options.  Note we're not using --ascii, which causes all character data
-        // to be converted to PETSCII by default.  By keeping things "raw" we can define our
-        // character encoding explicitly.  Anybody who wants to move the code to native
-        // assembly can generate for PETSCII and then just delete the sg_petscii .cdefs.
-        public const string OPTIONS = "--case-sensitive --nostart --long-address -Wall";
+        // Standard options.  For historical reasons the assembler expects PETSCII input by
+        // default, and requires "--ascii" for ASCII/UTF-8 input.  This flag switches the
+        // default "none" encoding from "raw" to something that converts characters to
+        // PETSCII, so if you want to output strings in another format (such as ASCII) an
+        // explicit encoding must be specified.
+        public const string OPTIONS = "--ascii --case-sensitive --nostart --long-address -Wall";

        // Paths from generator.
        private List<string> mPathNames;
--- a/SourceGen/RuntimeData/Help/codegen.html
+++ b/SourceGen/RuntimeData/Help/codegen.html
@@ -159,15 +159,14 @@ code, but also needs to know how to handle the corner cases.</p>
    that they plan to move to carets.)</li>
  <li>The arguments to COP and BRK require immediate-mode syntax
    (<code>COP #$03</code> rather than <code>COP $03</code>).
-  <li>By default, the assembler assumes that the input is PETSCII, but
-    doesn't convert characters in text strings.  So PETSCII source files
-    generate PETSCII strings, and ASCII source files generate ASCII
-    strings.  However, if you use the built-in "screen" encoding, you will
-    get the wrong behavior if you compile an ASCII source without the
-    "--ascii" command-line flag, because it expects to convert from
-    PETSCII.  To get the behavior expected of a cross-assembler, the
-    recommended approach seems to be to pass "--ascii" and explicitly
-    define an ASCII encoding for use with ASCII text strings.</li>
+  <li>For historical reasons, the default behavior of the assembler is to
+    assume that the source file is PETSCII, and the desired encoding for
+    strings is also PETSCII.  No character conversion is done, so anybody
+    assembling ASCII files will get ASCII strings (which works out pretty
+    well if you're assembling code for a non-Commodore target).  However,
+    the documentation says you're required to pass the "--ascii" flag when
+    the input is ASCII/UTF-8, so to build files that want ASCII operands
+    an explicit character encoding definition must be provided.</li>
 </ul>

 <p>Notes:</p>
--- a/SourceGen/SGTestData/Expected/1000-allops-value-65816_64tass.S
+++ b/SourceGen/SGTestData/Expected/1000-allops-value-65816_64tass.S
@@ -1,4 +1,6 @@
        .cpu    "65816"
+        .enc    sg_ascii
+        .cdef   $20,$7e,$20
 *       =       $1000
        .as
        .xs
--- a/SourceGen/SGTestData/Expected/1001-allops-zero-65816_64tass.S
+++ b/SourceGen/SGTestData/Expected/1001-allops-zero-65816_64tass.S
@@ -1,4 +1,6 @@
        .cpu    "65816"
+        .enc    sg_ascii
+        .cdef   $20,$7e,$20
 *       =       $1000
        .as
        .xs
--- a/SourceGen/SGTestData/Expected/1002-embedded-instructions_64tass.S
+++ b/SourceGen/SGTestData/Expected/1002-embedded-instructions_64tass.S
@@ -1,4 +1,6 @@
        .cpu    "65816"
+        .enc    sg_ascii
+        .cdef   $20,$7e,$20
 *       =       $1000
        .as
        .xs
--- a/SourceGen/SGTestData/Expected/1003-flags-and-branches_64tass.S
+++ b/SourceGen/SGTestData/Expected/1003-flags-and-branches_64tass.S
@@ -1,4 +1,6 @@
        .cpu    "65816"
+        .enc    sg_ascii
+        .cdef   $20,$7e,$20
 *       =       $1000
        .as
        .xs
--- a/SourceGen/SGTestData/Expected/1004-data-recognition_64tass.S
+++ b/SourceGen/SGTestData/Expected/1004-data-recognition_64tass.S
@@ -1,4 +1,6 @@
        .cpu    "65816"
+        .enc    sg_ascii
+        .cdef   $20,$7e,$20
 *       =       $1000
        .as
        .xs
--- a/SourceGen/SGTestData/Expected/2000-allops-value-6502_64tass.S
+++ b/SourceGen/SGTestData/Expected/2000-allops-value-6502_64tass.S
@@ -1,4 +1,6 @@
        .cpu    "6502i"
+        .enc    sg_ascii
+        .cdef   $20,$7e,$20
 *       =       $1000
        jsr     L1035
        jsr     L1038
--- a/SourceGen/SGTestData/Expected/2001-allops-zero-6502_64tass.S
+++ b/SourceGen/SGTestData/Expected/2001-allops-zero-6502_64tass.S
@@ -1,4 +1,6 @@
        .cpu    "6502i"
+        .enc    sg_ascii
+        .cdef   $20,$7e,$20
 *       =       $1000
        jsr     L1035
        jsr     L1038
--- a/SourceGen/SGTestData/Expected/2002-allops-value-65C02_64tass.S
+++ b/SourceGen/SGTestData/Expected/2002-allops-value-65C02_64tass.S
@@ -1,4 +1,6 @@
        .cpu    "65c02"
+        .enc    sg_ascii
+        .cdef   $20,$7e,$20
 *       =       $1000
        jsr     L1014
        jsr     L108A
--- a/SourceGen/SGTestData/Expected/2003-allops-zero-65C02_64tass.S
+++ b/SourceGen/SGTestData/Expected/2003-allops-zero-65C02_64tass.S
@@ -1,4 +1,6 @@
        .cpu    "65c02"
+        .enc    sg_ascii
+        .cdef   $20,$7e,$20
 *       =       $1000
        jsr     L1014
        jsr     L108A
--- a/SourceGen/SGTestData/Expected/2004-numeric-types_64tass.S
+++ b/SourceGen/SGTestData/Expected/2004-numeric-types_64tass.S
@@ -1,5 +1,7 @@
 ;Project file was edited to get all big-endian data types.
        .cpu    "6502"
+        .enc    sg_ascii
+        .cdef   $20,$7e,$20
 *       =       $1000
        rts

--- a/SourceGen/SGTestData/Expected/2005-string-types_64tass.S
+++ b/SourceGen/SGTestData/Expected/2005-string-types_64tass.S
@@ -1,5 +1,7 @@
 ;Project file was edited to get zero-length strings and reverse DCI.
        .cpu    "6502"
+        .enc    sg_ascii
+        .cdef   $20,$7e,$20
 *       =       $1000
        rts

--- a/SourceGen/SGTestData/Expected/2006-operand-formats_64tass.S
+++ b/SourceGen/SGTestData/Expected/2006-operand-formats_64tass.S
@@ -1,5 +1,7 @@
 ;Project file was edited for some ASCII operands.
        .cpu    "65816"
+        .enc    sg_ascii
+        .cdef   $20,$7e,$20
 *       =       $1000
        .as
        .xs
--- a/SourceGen/SGTestData/Expected/2007-labels-and-symbols_64tass.S
+++ b/SourceGen/SGTestData/Expected/2007-labels-and-symbols_64tass.S
@@ -1,6 +1,8 @@
 ;Project was edited to add a label in the middle of a dense hex region, and add
 ;a duplicate label.
        .cpu    "65816"
+        .enc    sg_ascii
+        .cdef   $20,$7e,$20
 zip     =       $cd
 absl    =       $1029
 plataddr =      $3000      ;address only in platform file
--- a/SourceGen/SGTestData/Expected/2008-address-changes_64tass.S
+++ b/SourceGen/SGTestData/Expected/2008-address-changes_64tass.S
@@ -1,4 +1,6 @@
        .cpu    "65816"
+        .enc    sg_ascii
+        .cdef   $20,$7e,$20
 *       =       $1000
        .as
        .xs
--- a/SourceGen/SGTestData/Expected/2009-branches-and-banks_64tass.S
+++ b/SourceGen/SGTestData/Expected/2009-branches-and-banks_64tass.S
@@ -1,4 +1,6 @@
        .cpu    "65816"
+        .enc    sg_ascii
+        .cdef   $20,$7e,$20
 longsym =       $123456

 *       =       $1000
--- a/SourceGen/SGTestData/Expected/2010-target-adjustment_64tass.S
+++ b/SourceGen/SGTestData/Expected/2010-target-adjustment_64tass.S
@@ -1,4 +1,6 @@
        .cpu    "65816"
+        .enc    sg_ascii
+        .cdef   $20,$7e,$20
 *       =       $1000
        .as
        .xs
--- a/SourceGen/SGTestData/Expected/2011-hinting_64tass.S
+++ b/SourceGen/SGTestData/Expected/2011-hinting_64tass.S
@@ -1,4 +1,6 @@
        .cpu    "6502"
+        .enc    sg_ascii
+        .cdef   $20,$7e,$20
 *       =       $1000
        .byte   $03
        .byte   $02
--- a/SourceGen/SGTestData/Expected/2012-label-localizer_64tass.S
+++ b/SourceGen/SGTestData/Expected/2012-label-localizer_64tass.S
@@ -1,4 +1,6 @@
        .cpu    "65816"
+        .enc    sg_ascii
+        .cdef   $20,$7e,$20
 REALLYLONGLABELNAME = $8888 ;that's a long name

 *       =       $1000
--- a/SourceGen/SGTestData/Expected/2013-notes-and-comments_64tass.S
+++ b/SourceGen/SGTestData/Expected/2013-notes-and-comments_64tass.S
@@ -14,6 +14,8 @@
 ;*                                     *
 ;***************************************
        .cpu    "6502"
+        .enc    sg_ascii
+        .cdef   $20,$7e,$20
 plataddr =      $3000      ;address only in platform file

 ;Short, unboxed comment here!!
--- a/SourceGen/SGTestData/Expected/2014-label-dp_64tass.S
+++ b/SourceGen/SGTestData/Expected/2014-label-dp_64tass.S
@@ -1,5 +1,7 @@
 ;6502bench SourceGen v1.1.0-dev1
        .cpu    "65816"
+        .enc    sg_ascii
+        .cdef   $20,$7e,$20
 *       =       $1000
        .as
        .xs
--- a/SourceGen/SGTestData/Expected/2015-64k-nops_64tass.S
+++ b/SourceGen/SGTestData/Expected/2015-64k-nops_64tass.S
@@ -1,4 +1,6 @@
        .cpu    "6502"
+        .enc    sg_ascii
+        .cdef   $20,$7e,$20
 *       =       $0000
        nop
        nop
--- a/SourceGen/SGTestData/Expected/2016-char-encoding-a_64tass.S
+++ b/SourceGen/SGTestData/Expected/2016-char-encoding-a_64tass.S
@@ -1,5 +1,7 @@
 ;Project edited to mark some non-ASCII operands as ASCII.
        .cpu    "65816"
+        .enc    sg_ascii
+        .cdef   $20,$7e,$20
 *       =       $1000
        .as
        .xs
--- a/SourceGen/SGTestData/Expected/2017-char-encoding-p_64tass.S
+++ b/SourceGen/SGTestData/Expected/2017-char-encoding-p_64tass.S
@@ -1,11 +1,5 @@
 ;Project edited to mark some non-ASCII operands as ASCII.
        .cpu    "65816"
-        .enc    sg_petscii
-        .cdef   " @", $20
-        .cdef   "AZ", $c1
-        .cdef   "az", $41
-        .cdef   "[[", $5b
-        .cdef   "]]", $5d
 *       =       $1000
        .as
        .xs
--- a/SourceGen/SGTestData/Expected/2018-char-encoding-s_64tass.S
+++ b/SourceGen/SGTestData/Expected/2018-char-encoding-s_64tass.S
@@ -1,12 +1,6 @@
 ;Project edited to mark some non-ASCII operands as ASCII.
        .cpu    "65816"
-        .enc    sg_screen
-        .cdef   " ?", $20
-        .cdef   "@@", $00
-        .cdef   "AZ", $41
-        .cdef   "az", $01
-        .cdef   "[[", $1b
-        .cdef   "]]", $1d
+        .enc    screen
 *       =       $1000
        .as
        .xs