Several improvements to escape sequence handling.

* Add the missing special character \_ (underscore).
* Partial implementations of \a (leader character)
and \E (uninterpreted escape character).
* Parse and ignore \r (reverse line feed).
* Add a WARNING message about undefined escape sequences.
* Add an UNSUPP message about unsupported escape sequences.
* Mark \! and \? (transparent throughput)
and \O (suppress output) as unsupported.
* Treat the various variants of zero-width spaces as one-byte escape
sequences rather than as special characters, to avoid defining bogus
forms with square brackets.
* For special characters with one-byte names, do not define bogus
forms with square brackets, except for \[-], which is valid.
* In the form with square brackets, undefined special characters do not
fall back to printing the name verbatim, not even for one-byte names.
* Starting a special character name with a blank is an error.
* Undefined escape sequences never abort formatting of the input
string, not even in HTML output mode.
* Document the newly handled escapes, and a few that were missing.
* Regression tests for most of the above.

1 files changed, 33 insertions, 14 deletions

diff --git a/roff.c b/roff.c
index 99d33c84..02dae757 100644
--- a/roff.c
+++ b/roff.c
@@ -1154,6 +1154,7 @@ roff_res(struct roff *r, struct buf *buf, int ln, int pos)
 	struct roff_node *n;	/* used for header comments */
 	const char	*start;	/* start of the string to process */
 	char		*stesc;	/* start of an escape sequence ('\\') */
+	const char	*esct;	/* type of esccape sequence */
 	char		*ep;	/* end of comment string */
 	const char	*stnam;	/* start of the name, after "[(*" */
 	const char	*cp;	/* end of the name, e.g. before ']' */
@@ -1163,7 +1164,6 @@ roff_res(struct roff *r, struct buf *buf, int ln, int pos)
 	size_t		 naml;	/* actual length of the escape name */
 	size_t		 asz;	/* length of the replacement */
 	size_t		 rsz;	/* length of the rest of the string */
-	enum mandoc_esc	 esc;	/* type of the escape sequence */
 	int		 inaml;	/* length returned from mandoc_escape() */
 	int		 expand_count;	/* to avoid infinite loops */
 	int		 npos;	/* position in numeric expression */
@@ -1172,6 +1172,7 @@ roff_res(struct roff *r, struct buf *buf, int ln, int pos)
 	int		 done;	/* no more input available */
 	int		 deftype; /* type of definition to paste */
 	int		 rcsid;	/* kind of RCS id seen */
+	enum mandocerr	 err;	/* for escape sequence problems */
 	char		 sign;	/* increment number register */
 	char		 term;	/* character terminating the escape */
 
@@ -1304,7 +1305,10 @@ roff_res(struct roff *r, struct buf *buf, int ln, int pos)
 
 		term = '\0';
 		cp = stesc + 1;
-		switch (*cp) {
+		if (*cp == 'E')
+			cp++;
+		esct = cp;
+		switch (*esct) {
 		case '*':
 		case '$':
 			res = NULL;
@@ -1320,12 +1324,26 @@ roff_res(struct roff *r, struct buf *buf, int ln, int pos)
 			res = ubuf;
 			break;
 		default:
-			esc = mandoc_escape(&cp, &stnam, &inaml);
-			if (esc == ESCAPE_ERROR ||
-			    (esc == ESCAPE_SPECIAL &&
-			     mchars_spec2cp(stnam, inaml) < 0))
-				mandoc_msg(MANDOCERR_ESC_BAD,
-				    ln, (int)(stesc - buf->buf),
+			err = MANDOCERR_OK;
+			switch(mandoc_escape(&cp, &stnam, &inaml)) {
+			case ESCAPE_SPECIAL:
+				if (mchars_spec2cp(stnam, inaml) >= 0)
+					break;
+				/* FALLTHROUGH */
+			case ESCAPE_ERROR:
+				err = MANDOCERR_ESC_BAD;
+				break;
+			case ESCAPE_UNDEF:
+				err = MANDOCERR_ESC_UNDEF;
+				break;
+			case ESCAPE_UNSUPP:
+				err = MANDOCERR_ESC_UNSUPP;
+				break;
+			default:
+				break;
+			}
+			if (err != MANDOCERR_OK)
+				mandoc_msg(err, ln, (int)(stesc - buf->buf),
 				    "%.*s", (int)(cp - stesc), stesc);
 			stesc--;
 			continue;
@@ -1382,7 +1400,7 @@ roff_res(struct roff *r, struct buf *buf, int ln, int pos)
 				cp++;
 				break;
 			}
-			if (*cp++ != '\\' || stesc[1] != 'w') {
+			if (*cp++ != '\\' || *esct != 'w') {
 				naml++;
 				continue;
 			}
@@ -1390,6 +1408,7 @@ roff_res(struct roff *r, struct buf *buf, int ln, int pos)
 			case ESCAPE_SPECIAL:
 			case ESCAPE_UNICODE:
 			case ESCAPE_NUMBERED:
+			case ESCAPE_UNDEF:
 			case ESCAPE_OVERSTRIKE:
 				naml++;
 				break;
@@ -1403,7 +1422,7 @@ roff_res(struct roff *r, struct buf *buf, int ln, int pos)
 		 * undefined, resume searching for escapes.
 		 */
 
-		switch (stesc[1]) {
+		switch (*esct) {
 		case '*':
 			if (arg_complete) {
 				deftype = ROFFDEF_USER | ROFFDEF_PRE;
@@ -1430,15 +1449,15 @@ roff_res(struct roff *r, struct buf *buf, int ln, int pos)
 				break;
 			}
 			ctx = r->mstack + r->mstackpos;
-			npos = stesc[2] - '1';
+			npos = esct[1] - '1';
 			if (npos >= 0 && npos <= 8) {
 				res = npos < ctx->argc ?
 				    ctx->argv[npos] : "";
 				break;
 			}
-			if (stesc[2] == '*')
+			if (esct[1] == '*')
 				quote_args = 0;
-			else if (stesc[2] == '@')
+			else if (esct[1] == '@')
 				quote_args = 1;
 			else {
 				mandoc_msg(MANDOCERR_ARG_NONUM, ln,
@@ -1500,7 +1519,7 @@ roff_res(struct roff *r, struct buf *buf, int ln, int pos)
 		}
 
 		if (res == NULL) {
-			if (stesc[1] == '*')
+			if (*esct == '*')
 				mandoc_msg(MANDOCERR_STR_UNDEF,
 				    ln, (int)(stesc - buf->buf),
 				    "%.*s", (int)naml, stnam);