Character-encoding checked for all text (arguments/values).

5 files changed, 35 insertions, 15 deletions

diff --git a/argv.c b/argv.c
index 51ef35b2..b72c5696 100644
--- a/argv.c
+++ b/argv.c
@@ -267,7 +267,7 @@ static int
 args(struct mdoc *mdoc, int line, 
 		int *pos, char *buf, int fl, char **v)
 {
-	int		  i, c;
+	int		  i;
 	char		 *p, *pp;
 
 	assert(*pos > 0);
@@ -290,14 +290,11 @@ args(struct mdoc *mdoc, int line,
 	 */
 
 	if ((fl & ARGS_DELIM) && mdoc_iscdelim(buf[*pos])) {
-		for (i = *pos; (c = buf[i]); ) {
-			if ( ! mdoc_iscdelim(c))
+		for (i = *pos; buf[i]; ) {
+			if ( ! mdoc_iscdelim(buf[i]))
 				break;
 			i++;
-			if (0 == buf[i] || ! isspace(c))
-				break;
-			i++;
-			while (buf[i] && isspace(c))
+			while (buf[i] && isspace((int)buf[i]))
 				i++;
 		}
 		if (0 == buf[i]) {
@@ -404,8 +401,8 @@ args(struct mdoc *mdoc, int line,
 		/* Do non-tabsep look-ahead here. */
 		
 		if ( ! (ARGS_TABSEP & fl))
-			while ((c = buf[*pos])) {
-				if (isspace(c))
+			while (buf[*pos]) {
+				if (isspace((int)buf[*pos]))
 					if ('\\' != buf[*pos - 1])
 						break;
 				(*pos)++;
diff --git a/mdoc.3 b/mdoc.3
index fe09e890..7bddf115 100644
--- a/mdoc.3
+++ b/mdoc.3
@@ -89,7 +89,9 @@ This section further defines the
 .Sx Functions 
 and
 .Sx Variables
-available to programmers.  The last sub-section,
+available to programmers.  Following that,
+.Sx Character Encoding
+describes input format.  Lastly, 
 .Sx Abstract Syntax Tree ,
 documents the output tree.
 .\" SUBSECTION
@@ -174,6 +176,21 @@ An array of string-ified token names.
 An array of string-ified token argument names.
 .El
 .\" SUBSECTION
+.Ss Character Encoding
+The
+.Xr mdoc 3
+library accepts only printable ASCII characters as defined by
+.Xr isprint 3 .
+Non-ASCII character sequences are escaped with an escape character 
+.Sq \\
+and followed by either an open-parenthesis 
+.Sq \&(
+for two-character sequences; an open-bracket
+.Sq \&[
+for n-character sequences (terminated at a close-bracket
+.Sq \&] ) ;
+or one of a small set of single characters for other escapes.
+.\" SUBSECTION
 .Ss Abstract Syntax Tree
 The 
 .Nm
diff --git a/strings.c b/strings.c
index 3e9415c5..2ca3de93 100644
--- a/strings.c
+++ b/strings.c
@@ -55,14 +55,16 @@ mdoc_isescape(const char *p)
 		/* FALLTHROUGH */
 	case (' '):
 		/* FALLTHROUGH */
+	case ('&'):
+		/* FALLTHROUGH */
 	case ('.'):
 		/* FALLTHROUGH */
 	case ('e'):
 		return(2);
 	case ('('):
-		if (0 == *++p)
+		if (0 == *++p || ! isgraph(*p))
 			return(0);
-		if (0 == *++p)
+		if (0 == *++p || ! isgraph(*p))
 			return(0);
 		return(4);
 	case ('['):
@@ -72,7 +74,7 @@ mdoc_isescape(const char *p)
 	}
 
 	for (c = 3, p++; *p && ']' != *p; p++, c++)
-		if (isspace(*p))
+		if ( ! isgraph(*p))
 			break;
 
 	return(*p == ']' ? c : 0);
diff --git a/term.c b/term.c
index 28089e2d..8ba4beae 100644
--- a/term.c
+++ b/term.c
@@ -1054,7 +1054,7 @@ termp_sq_pre(DECL_ARGS)
 
 	if (MDOC_BODY != node->type)
 		return(1);
-	word(p, "`");
+	word(p, "\'");
 	p->flags |= TERMP_NOSPACE;
 	return(1);
 }
diff --git a/validate.c b/validate.c
index dcc7f4ce..82f69874 100644
--- a/validate.c
+++ b/validate.c
@@ -17,6 +17,7 @@
  * PERFORMANCE OF THIS SOFTWARE.
  */
 #include <assert.h>
+#include <ctype.h>
 #include <stdlib.h>
 
 #include "private.h"
@@ -396,6 +397,9 @@ check_text(struct mdoc *mdoc, size_t line, size_t pos, const char *p)
 	size_t		 c;
 
 	for ( ; *p; p++) {
+		if ( ! isprint(*p) && '\t' != *p)
+			return(mdoc_perr(mdoc, line, pos,
+					"invalid characters"));
 		if ('\\' != *p)
 			continue;
 		if ((c = mdoc_isescape(p))) {
@@ -403,7 +407,7 @@ check_text(struct mdoc *mdoc, size_t line, size_t pos, const char *p)
 			continue;
 		}
 		return(mdoc_perr(mdoc, line, pos,
-				"invalid escape sequence"));
+					"invalid escape sequence"));
 	}
 
 	return(1);