1. Add the flag MKD_STRICT, which disables any configured superscripting

of A^B and non-emphasis on A_B.
2. markdown(1) now has the `strict` (enable `MKD_STRICT`) and `relax` (disable
   `MKD_STRICT`) flags.
3. truncate the input buffer at the end of every call to `text()` (corrects
   defect with --relaxed-emphasis where it was incorrectly looking back past
   the start of the input and triggering on the contents of the previous
   input (not reliably, so only detected when I put MKD_STRICT into the
   code.)
4. `markdown.1` manpage renamed to `markdown.1.in` and processed according
   to whether --relaxed-emphasis and/or --with-superscript are configured.

VERSION

@@ -1 +1 @@
-1.2.10
+1.3.0

configure.sh

@@ -106,7 +106,7 @@ AC_DEFINE 'TABSTOP' $TABSTOP
 AC_SUB    'TABSTOP' $TABSTOP
 
 test -z "$WITH_SUPERSCRIPT" || AC_DEFINE 'SUPERSCRIPT'	1
-
+test -z "$RELAXED_EMPHASIS" || AC_DEFINE 'RELAXED_EMPHASIS'	1
 
 
 if [ "$WITH_AMALLOC" ]; then
@@ -116,13 +116,16 @@ else
     AC_SUB	'AMALLOC'	''
 fi
 
-if [ "$RELAXED_EMPHASIS" ]; then
-    AC_DEFINE	'RELAXED_EMPHASIS'	1
+if [ "$RELAXED_EMPHASIS" -o "$WITH_SUPERSCRIPT" ]; then
+    AC_SUB      'STRICT'	''
+else
+    AC_SUB	'STRICT'	'.\"'
 fi
 
+
 [ "$OS_FREEBSD" -o "$OS_DRAGONFLY" ] || AC_CHECK_HEADERS malloc.h
 
 [ "$WITH_DL_TAG" ] && AC_DEFINE 'DL_TAG_EXTENSION' '1'
 [ "$WITH_PANDOC_HEADER" ] && AC_DEFINE 'PANDOC_HEADER' '1'
 
-AC_OUTPUT Makefile version.c
+AC_OUTPUT Makefile version.c markdown.1

generate.c

@@ -74,6 +74,31 @@ cursor(MMIOT *f)
 }
 
 
+static int
+isthisspace(MMIOT *f, int i)
+{
+    int c = peek(f, i);
+
+    return isspace(c) || (c == EOF);
+}
+
+
+static int
+isthisalnum(MMIOT *f, int i)
+{
+    int c = peek(f, i);
+
+    return (c != EOF) && isalnum(c);
+}
+
+
+static int
+isthisnonword(MMIOT *f, int i)
+{
+    return isthisspace(f, i) || ispunct(peek(f,i));
+}
+
+
 /* return/set the current cursor position
  */
 #define mmiotseek(f,x)	(f->isp = x)
@@ -662,11 +687,11 @@ forbidden_tag(MMIOT *f)
     if ( f->flags & DENY_HTML )
 	return 1;
 
-    if ( c == 'A' && (f->flags & DENY_A) && !isalnum(peek(f,2)) )
+    if ( c == 'A' && (f->flags & DENY_A) && !isthisalnum(f,2) )
 	return 1;
     if ( c == 'I' && (f->flags & DENY_IMG)
 		  && strncasecmp(cursor(f)+1, "MG", 2) == 0
-		  && !isalnum(peek(f,4)) )
+		  && !isthisalnum(f,4) )
 	return 1;
     return 0;
 }
@@ -746,22 +771,6 @@ maybe_tag_or_link(MMIOT *f)
 } /* maybe_tag_or_link */
 
 
-static int
-isthisspace(MMIOT *f, int i)
-{
-    int c = peek(f, i);
-
-    return isspace(c) || (c == EOF);
-}
-
-
-static int
-isthisnonword(MMIOT *f, int i)
-{
-    return isthisspace(f, i) || ispunct(peek(f,i));
-}
-
-
 /* smartyquote code that's common for single and double quotes
  */
 static int
@@ -929,7 +938,8 @@ text(MMIOT *f)
 			Qchar(c, f);
 		    break;
 #if SUPERSCRIPT
-	case '^':   if ( isthisspace(f,-1) || isthisspace(f,1) )
+	/* A^B -> A<sup>B</sup> */
+	case '^':   if ( (f->flags & STRICT) || isthisspace(f,-1) || isthisspace(f,1) )
 			Qchar(c,f);
 		    else {
 			char *sup = cursor(f);
@@ -944,13 +954,12 @@ text(MMIOT *f)
 		    }
 		    break;
 #endif
-	case '_':
 #if RELAXED_EMPHASIS
-		    /* If RELAXED_EMPHASIS, underscores don't count when
-		     * they're in the middle of a word.
-		     */
-		    if ( (isthisspace(f,-1) && isthisspace(f,1))
-			    || (isalnum(peek(f,-1)) && isalnum(peek(f,1))) ) {
+	/* Underscores don't count if they're in the middle of a word */
+	case '_':
+		    if ( (!(f->flags & STRICT))
+			     && ((isthisspace(f,-1) && isthisspace(f,1))
+			      || (isthisalnum(f,-1) && isthisalnum(f,1))) ){
 			Qchar(c, f);
 			break;
 		    }
@@ -1004,7 +1013,7 @@ text(MMIOT *f)
 		    break;
 
 	case '&':   j = (peek(f,1) == '#' ) ? 2 : 1;
-		    while ( isalnum(peek(f,j)) )
+		    while ( isthisalnum(f,j) )
 			++j;
 
 		    if ( peek(f,j) != ';' )
@@ -1017,6 +1026,8 @@ text(MMIOT *f)
 		    break;
 	}
     }
+    /* truncate the input string after we've finished processing it */
+    S(f->in) = f->isp = 0;
 } /* text */
 
 

main.c

@@ -41,12 +41,14 @@ static struct {
     int off;
     int flag;
 } opts[] = {
-    { "tabstop", 0, MKD_TABSTOP },
-    { "image",   1, MKD_NOIMAGE },
-    { "links",   1, MKD_NOLINKS },
+    { "tabstop", 0, MKD_TABSTOP  },
+    { "image",   1, MKD_NOIMAGE  },
+    { "links",   1, MKD_NOLINKS  },
+    { "relax",   1, MKD_STRICT   },
+    { "strict",  0, MKD_STRICT   },
     { "header",  1, MKD_NOHEADER },
-    { "html",    1, MKD_NOHTML },
-    { "cdata",   0, MKD_CDATA },
+    { "html",    1, MKD_NOHTML   },
+    { "cdata",   0, MKD_CDATA    },
 } ;
 
 #define NR(x)	(sizeof x / sizeof x[0])
@@ -138,7 +140,6 @@ main(int argc, char **argv)
     argc -= optind;
     argv += optind;
 
-
     if ( text ) {
 	rc = mkd_text( text, strlen(text), stdout, flags);
 	adump();

markdown.1.in

@@ -50,6 +50,10 @@ Generate valid XML output.
 Do not process pandoc headers.
 .It Ar tabstops
 Use markdown-standard 4-space tabstops.
+@STRICT@.It Ar strict
+@STRICT@Disable superscript and relaxed emphasis.
+@STRICT@.It Ar relax
+@STRICT@Enable superscript and relaxed emphasis (this is the default.)
 .El
 .Pp
 As an example, the option

markdown.h

@@ -68,6 +68,7 @@ typedef struct mmiot {
 #define DENY_IMG	0x0002
 #define DENY_SMARTY	0x0004
 #define DENY_HTML	0x0008
+#define STRICT		0x0010
 #define INSIDE_TAG	0x0020
 #define NO_PSEUDO_PROTO	0x0040
 #define CDATA_OUTPUT	0x0080

mkdio.h

@@ -46,6 +46,7 @@ extern char markdown_version[];
 #define MKD_NOIMAGE	0x0002	/* don't do image processing, block <img> */
 #define MKD_NOPANTS	0x0004	/* don't run smartypants() */
 #define MKD_NOHTML	0x0008	/* don't allow raw html through AT ALL */
+#define MKD_STRICT	0x0010	/* disable SUPERSCRIPT, RELAXED_EMPHASIS */
 #define MKD_TAGTEXT	0x0020	/* don't expand `_` and `*` */
 #define MKD_NO_EXT	0x0040	/* don't allow pseudo-protocols */
 #define MKD_CDATA	0x0080	/* generate code for xml ![CDATA[...]] */
@@ -56,4 +57,5 @@ extern char markdown_version[];
 #define MKD_NOHEADER	0x0100	/* don't process header blocks */
 #define MKD_TABSTOP	0x0200	/* expand tabs to 4 spaces */
 
+
 #endif/*_MKDIO_D*/

tests/emphasis.t

@@ -58,23 +58,24 @@ else
     rc=1
 fi
 
-./echo -n '  A_B .............................. '
-case "$MARKDOWN_VERSION" in
-*RELAXED*)
-    if ./echo '_A_B' | ./markdown | grep -i 'A_B' > /dev/null; then
+if ./markdown -V | grep RELAXED >/dev/null; then
+    ./echo -n '  _A_B with -frelax ................ '
+
+    if ./echo '_A_B' | ./markdown -frelax | grep -i 'A_B' > /dev/null; then
 	./echo "ok"
     else
 	./echo "FAILED"
 	rc=1
     fi
-    ;;
-*)  if ./echo '_A_B' | ./markdown | grep -i 'A</em>B' > /dev/null; then
+    
+    ./echo -n '  _A_B with -fstrict ............... '
+
+    if ./echo '_A_B' | ./markdown -fstrict | grep -i 'A</em>B' > /dev/null; then
 	./echo "ok"
     else
 	./echo "FAILED"
 	rc=1
     fi
-    ;;
-esac
+fi
 
 exit $rc

tests/smarty.t

@@ -114,13 +114,20 @@ else
 fi
 
 if ./markdown -V | grep SUPERSCRIPT >/dev/null; then
-    ./echo -n "  A^B -> A<sup>B</sup> ............. "
-    if ./echo "A^B" | ./markdown | grep '<sup>B</sup>' >/dev/null; then
+    ./echo -n "  A^B -> A<sup>B</sup> (-frelax) ... "
+    if ./echo "A^B" | ./markdown -frelax | grep '<sup>B</sup>' >/dev/null; then
 	./echo "ok"
     else
 	./echo "FAILED"
 	rc=1
     fi
+    ./echo -n "  A^B != A<sup>B</sup> (-fstrict) .. "
+    if ./echo "A^B" | ./markdown -fstrict | grep '<sup>B</sup>' >/dev/null; then
+	./echo "FAILED"
+	rc=1
+    else
+	./echo "ok"
+    fi
 fi
 
 exit $rc

version.c.in

@@ -10,12 +10,12 @@ char markdown_version[] = VERSION
 #if @TABSTOP@ != 4
 		" TAB=@TABSTOP@"
 #endif
-#if SUPERSCRIPT
-		" SUPERSCRIPT"
-#endif
 #if USE_AMALLOC
 		" DEBUG"
 #endif
+#if SUPERSCRIPT
+		" SUPERSCRIPT"
+#endif
 #if RELAXED_EMPHASIS
 		" RELAXED"
 #endif