xmltok: Add more in-code documentation about byte types

expat/lib/xmltok.c

@@ -587,11 +587,13 @@ static const struct normal_encoding ascii_encoding
 static int PTRFASTCALL
 unicode_byte_type(char hi, char lo) {
   switch ((unsigned char)hi) {
+  /* 0xD800–0xDBFF first 16-bit code unit or high surrogate (W1) */
   case 0xD8:
   case 0xD9:
   case 0xDA:
   case 0xDB:
     return BT_LEAD4;
+  /* 0xDC00–0xDFFF second 16-bit code unit or low surrogate (W2) */
   case 0xDC:
   case 0xDD:
   case 0xDE:
@@ -599,8 +601,8 @@ unicode_byte_type(char hi, char lo) {
     return BT_TRAIL;
   case 0xFF:
     switch ((unsigned char)lo) {
-    case 0xFF:
-    case 0xFE:
+    case 0xFF: /* noncharacter-FFFF */
+    case 0xFE: /* noncharacter-FFFE */
       return BT_NONXML;
     }
     break;

expat/lib/xmltok_impl.h

@@ -31,43 +31,43 @@
 */
 
 enum {
-  BT_NONXML,
-  BT_MALFORM,
-  BT_LT,
-  BT_AMP,
-  BT_RSQB,
-  BT_LEAD2,
-  BT_LEAD3,
-  BT_LEAD4,
-  BT_TRAIL,
-  BT_CR,
-  BT_LF,
-  BT_GT,
-  BT_QUOT,
-  BT_APOS,
-  BT_EQUALS,
-  BT_QUEST,
-  BT_EXCL,
-  BT_SOL,
-  BT_SEMI,
-  BT_NUM,
-  BT_LSQB,
-  BT_S,
-  BT_NMSTRT,
-  BT_COLON,
-  BT_HEX,
-  BT_DIGIT,
-  BT_NAME,
-  BT_MINUS,
+  BT_NONXML,   /* e.g. noncharacter-FFFF */
+  BT_MALFORM,  /* illegal, with regard to encoding */
+  BT_LT,       /* less than = "<" */
+  BT_AMP,      /* ampersand = "&" */
+  BT_RSQB,     /* right square bracket = "[" */
+  BT_LEAD2,    /* lead byte of a 2-byte UTF-8 character */
+  BT_LEAD3,    /* lead byte of a 3-byte UTF-8 character */
+  BT_LEAD4,    /* lead byte of a 4-byte UTF-8 character */
+  BT_TRAIL,    /* trailing unit, e.g. second 16-bit unit of a 4-byte char. */
+  BT_CR,       /* carriage return = "\r" */
+  BT_LF,       /* line feed = "\n" */
+  BT_GT,       /* greater than = ">" */
+  BT_QUOT,     /* quotation character = "\"" */
+  BT_APOS,     /* aposthrophe = "'" */
+  BT_EQUALS,   /* equal sign = "=" */
+  BT_QUEST,    /* question mark = "?" */
+  BT_EXCL,     /* exclamation mark = "!" */
+  BT_SOL,      /* solidus, slash = "/" */
+  BT_SEMI,     /* semicolon = ";" */
+  BT_NUM,      /* number sign = "#" */
+  BT_LSQB,     /* left square bracket = "[" */
+  BT_S,        /* white space, e.g. "\t", " "[, "\r"] */
+  BT_NMSTRT,   /* non-hex name start letter = "G".."Z" + "g".."z" + "_" */
+  BT_COLON,    /* colon = ":" */
+  BT_HEX,      /* hex letter = "A".."F" + "a".."f" */
+  BT_DIGIT,    /* digit = "0".."9" */
+  BT_NAME,     /* dot and middle dot = "." + chr(0xb7) */
+  BT_MINUS,    /* minus = "-" */
   BT_OTHER,    /* known not to be a name or name start character */
   BT_NONASCII, /* might be a name or name start character */
-  BT_PERCNT,
-  BT_LPAR,
-  BT_RPAR,
-  BT_AST,
-  BT_PLUS,
-  BT_COMMA,
-  BT_VERBAR
+  BT_PERCNT,   /* percent sign = "%" */
+  BT_LPAR,     /* left parenthesis = "(" */
+  BT_RPAR,     /* right parenthesis = "(" */
+  BT_AST,      /* asterisk = "*" */
+  BT_PLUS,     /* plus sign = "+" */
+  BT_COMMA,    /* comma = "," */
+  BT_VERBAR    /* vertical bar = "|" */
 };
 
 #include <stddef.h>