Implement support for digraphs.

Specifically, the following six punctuator tokens are now supported:

<: :> <% %> %: %:%:

These behave the same as the existing tokens [, ], {, }, #, and ## (respectively), apart from their spelling.

This can be useful when the full ASCII character set cannot easily be displayed or input (e.g. on the IIgs text screen with certain language settings).

CCommon.pas

@@ -191,7 +191,8 @@ type
    charEnum =                           {character kinds}
       (illegal,ch_special,ch_dash,ch_plus,ch_lt,ch_gt,ch_eq,ch_exc,
        ch_and,ch_bar,ch_dot,ch_white,ch_eol,ch_eof,ch_char,ch_string,
-       ch_asterisk,ch_slash,ch_percent,ch_carot,ch_pound,letter,digit);
+       ch_asterisk,ch_slash,ch_percent,ch_carot,ch_pound,ch_colon,
+       letter,digit);
 
    tokenSet = set of tokenEnum;
    tokenClass = (reservedWord,reservedSymbol,identifier,intConstant,longConstant,
@@ -202,7 +203,7 @@ type
       numString: stringPtr;             {chars in number (macros only)}
       case class: tokenClass of         {token info}
          reservedWord  : ();
-         reservedSymbol: ();
+         reservedSymbol: (isDigraph: boolean);
          identifier    : (name: stringPtr;
                           symbolPtr: identPtr);
          intConstant   : (ival: integer);

Header.pas

@@ -18,7 +18,7 @@ uses CCommon, MM, Scanner, Symbol, CGI;
 {$segment 'SCANNER'}
 
 const
-   symFileVersion = 4;                  {version number of .sym file format}
+   symFileVersion = 5;                  {version number of .sym file format}
 
 var
    inhibitHeader: boolean;		{should .sym includes be blocked?}
@@ -717,6 +717,9 @@ procedure EndInclude {chPtr: ptr};
                 		WriteByte(ord(token.ispstring));
                 		end;
             macroParameter:	WriteWord(token.pnum);
+            reservedSymbol:	if token.kind in [lbracech,rbracech,lbrackch,
+                                   rbrackch,poundch,poundpoundop] then 
+                                   WriteByte(ord(token.isDigraph));
 	    otherwise:	;
 	    end; {case}
 	 end; {WriteToken}
@@ -1321,6 +1324,9 @@ var
                 	        token.ispstring := ReadByte <> 0;
                 	        end;
          macroParameter:	token.pnum := ReadWord;
+         reservedSymbol:	if token.kind in [lbracech,rbracech,lbrackch,
+                                   rbrackch,poundch,poundpoundop] then 
+                                   token.isDigraph := boolean(ReadByte);
 	 otherwise:		;
 	 end; {case}
       end; {ReadToken}

Scanner.asm

@@ -115,7 +115,8 @@ cch      equ   13
 
          enum  (illegal,ch_special,ch_dash,ch_plus,ch_lt,ch_gt,ch_eq,ch_exc),0
          enum  (ch_and,ch_bar,ch_dot,ch_white,ch_eol,ch_eof,ch_char,ch_string)
-         enum  (ch_asterisk,ch_slash,ch_percent,ch_carot,ch_pound,letter,digit)
+         enum  (ch_asterisk,ch_slash,ch_percent,ch_carot,ch_pound,ch_colon)
+         enum  (letter,digit)
 
          tsc                            create stack frame
          sec

Scanner.pas

@@ -736,8 +736,7 @@ case token.kind of
    unsignedsy,voidsy,volatilesy,whilesy:
                      write(reservedWords[token.kind]);
 
-   tildech,questionch,lparench,rparench,lbrackch,rbrackch,lbracech,
+   tildech,questionch,lparench,rparench,commach,semicolonch,colonch:
-   rbracech,commach,semicolonch,colonch,poundch:
                      begin
                      for i := minChar to maxChar do
                         if charSym[i] = token.kind then begin
@@ -746,6 +745,31 @@ case token.kind of
                            end; {if}
                      end;
 
+   lbrackch:         if not token.isDigraph then
+                        write('[')
+                     else
+                        write('<:');
+
+   rbrackch:         if not token.isDigraph then
+                        write(']')
+                     else
+                        write(':>');
+
+   lbracech:         if not token.isDigraph then
+                        write('{')
+                     else
+                        write('<%');
+
+   rbracech:         if not token.isDigraph then
+                        write('}')
+                     else
+                        write('%>');
+
+   poundch:          if not token.isDigraph then
+                        write('#')
+                     else
+                        write('%:');
+
    minusch:          write('-');
 
    plusch:           write('+');
@@ -2580,7 +2604,6 @@ reportEOL := true;
 tSkipping := skipping;                  {don't skip the directive name!}
 skipping := false;
 nextLineNumber := -1;
-NextCh;                                 {skip the '#' char}
 while charKinds[ord(ch)] = ch_white do  {skip white space}
    NextCh;
 if ch in ['a','d','e','i','l','p','u','w'] then begin
@@ -3671,7 +3694,7 @@ procedure NextToken;
 
 { Read the next token from the file.                            }
 
-label 1,2,3,4;
+label 1,2,3,4,5;
 
 type
    three = (s100,s1000,s4000);          {these declarations are used for a}
@@ -3701,6 +3724,7 @@ var
    tPtr: tokenListRecordPtr;            {for removing tokens from putback buffer}
    tToken: tokenType;                   {for merging tokens}
    sPtr,tsPtr: gstringPtr;              {for forming string constants}
+   lLastWasReturn: boolean;             {local copy of lastWasReturn}
 
 
    function EscapeCh: integer;
@@ -3899,11 +3923,13 @@ if tokenList <> nil then begin          {get a token put back by a macro}
          end; {if}
    goto 2;
    end; {if}
-                                        {skip white space}
+5:                                      {skip white space}
 while charKinds[ord(ch)] in [illegal,ch_white,ch_eol] do begin
    if charKinds[ord(ch)] = illegal then begin
-      if (ch = '#') and (lastWasReturn or (token.kind = eolsy)) then
+      if (ch = '#') and (lastWasReturn or (token.kind = eolsy)) then begin
+         NextCh;                        {skip the '#' char}
          PreProcess                     {call the preprocessor}
+         end {if}
       else begin
          tokenLine := lineNumber;       {record a # token}
          tokenColumn := ord(ord4(chPtr)-ord4(firstPtr));
@@ -3941,6 +3967,7 @@ case charKinds[ord(ch)] of
 
    ch_special  : begin
       token.kind := charSym[ord(ch)];
+      token.isDigraph := false;
       NextCh;
       end;
 
@@ -3949,6 +3976,7 @@ case charKinds[ord(ch)] of
 
    ch_pound : begin                     {tokens that start with '#'}
       NextCh;
+      token.isDigraph := false;
       if ch = '#' then begin
          token.kind := poundpoundop;
          NextCh;
@@ -4004,6 +4032,16 @@ case charKinds[ord(ch)] of
          token.kind := lteqop;
          NextCh;
          end
+      else if ch = ':' then begin
+         token.kind := lbrackch;        { <: digraph }
+         token.isDigraph := true;
+         NextCh;
+         end
+      else if ch = '%' then begin
+         token.kind := lbracech;        { <% digraph }
+         token.isDigraph := true;
+         NextCh;
+         end
       else
          token.kind := ltch;
       end;
@@ -4075,12 +4113,34 @@ case charKinds[ord(ch)] of
          token.kind := barch;
       end;
 
-   ch_percent: begin                      {tokens that start with '%'}
+   ch_percent: begin                    {tokens that start with '%'}
+      lLastWasReturn := lastWasReturn or (token.kind = eolsy);
       NextCh;
       if ch = '=' then begin
          token.kind := percenteqop;
          NextCh;
          end
+      else if ch = '>' then begin
+         token.kind := rbracech;        {%> digraph}
+         token.isDigraph := true;
+         NextCh;
+         end
+      else if ch = ':' then begin
+         NextCh;
+         token.isDigraph := true;
+         if (ch = '%') and (chPtr <> eofPtr) and (chr(chPtr^) = ':') then begin
+            token.kind := poundpoundop; {%:%: digraph}
+            NextCh;
+            NextCh;
+            end
+         else begin
+            token.kind := poundch;      {%: digraph}
+            if lLastWasReturn then begin
+               PreProcess;
+               goto 5;
+               end;
+            end;
+         end
       else
          token.kind := percentch;
       end;
@@ -4124,6 +4184,17 @@ case charKinds[ord(ch)] of
          end; {else}
       end;
 
+   ch_colon : begin                     {tokens that start with ':'}
+      NextCh;
+      if ch = '>' then begin
+         token.kind := rbrackch;        {:> digraph}
+         token.isDigraph := true;
+         NextCh;
+         end
+      else
+         token.kind := colonch;
+      end;
+
    ch_char  : CharConstant;		{character constants}
 
    ch_string: begin                     {string constants}

Table.asm

@@ -18,7 +18,8 @@ root     start                          dummy (.root) segment
 charKinds start                         character set
          enum  (illegal,ch_special,ch_dash,ch_plus,ch_lt,ch_gt,ch_eq,ch_exc),0
          enum  (ch_and,ch_bar,ch_dot,ch_white,ch_eol,ch_eof,ch_char,ch_string)
-         enum  (ch_asterisk,ch_slash,ch_percent,ch_carot,ch_pound,letter,digit)
+         enum  (ch_asterisk,ch_slash,ch_percent,ch_carot,ch_pound,ch_colon)
+         enum  (letter,digit)
 
 ! STANDARD
          dc    i'ch_eof'                nul
@@ -79,7 +80,7 @@ charKinds start                         character set
          dc    i'digit'                 7
          dc    i'digit'                 8
          dc    i'digit'                 9
-         dc    i'ch_special'            :
+         dc    i'ch_colon'              :
          dc    i'ch_special'            ;
          dc    i'ch_lt'                 <
          dc    i'ch_eq'                 =