Flip on unicode output (via \[uNNNN]) in -T[x]html. Here we go!

4 files changed, 42 insertions, 9 deletions

diff --git a/chars.c b/chars.c
index 06a2a923..808d70d7 100644
--- a/chars.c
+++ b/chars.c
@@ -1,4 +1,4 @@
-/*	$Id: chars.c,v 1.43 2011/05/15 22:29:50 kristaps Exp $ */
+/*	$Id: chars.c,v 1.44 2011/05/17 11:50:20 kristaps Exp $ */
 /*
  * Copyright (c) 2009, 2010 Kristaps Dzonsons <kristaps@bsd.lv>
  * Copyright (c) 2011 Ingo Schwarze <schwarze@openbsd.org>
@@ -138,7 +138,7 @@ mchars_res2cp(struct mchars *arg, const char *p, size_t sz)
 }
 
 /*
- * Numbered character to literal character.
+ * Numbered character string to ASCII codepoint.
  * This can only be a printable character (i.e., alnum, punct, space) so
  * prevent the character from ruining our state (backspace, newline, and
  * so on).
@@ -151,10 +151,24 @@ mchars_num2char(const char *p, size_t sz)
 
 	if ((i = mandoc_strntou(p, sz, 10)) < 0)
 		return('\0');
-
 	return(isprint(i) ? i : '\0');
 }
 
+/*
+ * Hex character string to Unicode codepoint.
+ * If the character is illegal, returns '\0'.
+ */
+int
+mchars_num2uc(const char *p, size_t sz)
+{
+	int               i;
+
+	if ((i = mandoc_strntou(p, sz, 16)) < 0)
+		return('\0');
+	/* FIXME: make sure we're not in a bogus range. */
+	return(i > 0x80 && i <= 0x10FFFF ? i : '\0');
+}
+
 /* 
  * Special character to string array.
  */
diff --git a/html.c b/html.c
index a50d2eb0..fd696284 100644
--- a/html.c
+++ b/html.c
@@ -1,4 +1,4 @@
-/*	$Id: html.c,v 1.143 2011/05/17 11:38:18 kristaps Exp $ */
+/*	$Id: html.c,v 1.144 2011/05/17 11:50:20 kristaps Exp $ */
 /*
  * Copyright (c) 2008, 2009, 2010, 2011 Kristaps Dzonsons <kristaps@bsd.lv>
  * Copyright (c) 2011 Ingo Schwarze <schwarze@openbsd.org>
@@ -315,6 +315,8 @@ html_strlen(const char *cp)
 		switch (mandoc_escape(&cp, &seq, &ssz)) {
 		case (ESCAPE_ERROR):
 			return(sz);
+		case (ESCAPE_UNICODE):
+			/* FALLTHROUGH */
 		case (ESCAPE_NUMBERED):
 			/* FALLTHROUGH */
 		case (ESCAPE_PREDEF):
@@ -373,6 +375,12 @@ print_encode(struct html *h, const char *p, int norecurse)
 			break;
 
 		switch (esc) {
+		case (ESCAPE_UNICODE):
+			/* Skip passed "u" header. */
+			c = mchars_num2uc(seq + 1, len - 1);
+			if ('\0' != c)
+				printf("&#x%x;", c);
+			break;
 		case (ESCAPE_NUMBERED):
 			c = mchars_num2char(seq, len);
 			if ('\0' != c)
diff --git a/mandoc.3 b/mandoc.3
index 66148c4d..0521e391 100644
--- a/mandoc.3
+++ b/mandoc.3
@@ -1,4 +1,4 @@
-.\"	$Id: mandoc.3,v 1.6 2011/05/01 10:40:52 kristaps Exp $
+.\"	$Id: mandoc.3,v 1.7 2011/05/17 11:50:20 kristaps Exp $
 .\"
 .\" Copyright (c) 2009, 2010, 2011 Kristaps Dzonsons <kristaps@bsd.lv>
 .\" Copyright (c) 2010 Ingo Schwarze <schwarze@openbsd.org>
@@ -15,7 +15,7 @@
 .\" ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
 .\" OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
 .\"
-.Dd $Mdocdate: May 1 2011 $
+.Dd $Mdocdate: May 17 2011 $
 .Dt MANDOC 3
 .Os
 .Sh NAME
@@ -26,6 +26,7 @@
 .Nm mchars_alloc ,
 .Nm mchars_free ,
 .Nm mchars_num2char ,
+.Nm mchars_num2uc ,
 .Nm mchars_res2cp ,
 .Nm mchars_res2str ,
 .Nm mchars_spec2cp ,
@@ -64,6 +65,8 @@
 .Fn mchars_free "struct mchars *p"
 .Ft char
 .Fn mchars_num2char "const char *cp" "size_t sz"
+.Ft int
+.Fn mchars_num2uc "const char *cp" "size_t sz"
 .Ft "const char *"
 .Fo mchars_res2str
 .Fa "struct mchars *p"
@@ -188,6 +191,9 @@ library also contains routines for translating character strings into glyphs
 .Pq see Fn mchars_alloc
 and parsing escape sequences from strings
 .Pq see Fn mandoc_escape .
+.Pp
+This library is
+.Ud
 .Sh REFERENCE
 This section documents the functions, types, and variables available
 via
@@ -247,8 +253,12 @@ The object must be freed with
 Free an object created with
 .Fn mchars_alloc .
 .It Fn mchars_num2char
-Convert a character index as found in \eN\(aq\(aq into a printable
-character.
+Convert a character index (e.g., the \eN\(aq\(aq escape) into a
+printable ASCII character.
+Returns \e0 (the nil character) if the input sequence is malformed.
+.It Fn mchars_num2uc
+Convert a hexadecimal character index (e.g., the \e[uNNNN] escape) into
+a Unicode codepoint.
 Returns \e0 (the nil character) if the input sequence is malformed.
 .It Fn mchars_res2cp
 Convert a predefined character into a valid Unicode codepoint.
diff --git a/mandoc.h b/mandoc.h
index db7b30bf..5f01644c 100644
--- a/mandoc.h
+++ b/mandoc.h
@@ -1,4 +1,4 @@
-/*	$Id: mandoc.h,v 1.75 2011/05/15 15:30:33 kristaps Exp $ */
+/*	$Id: mandoc.h,v 1.76 2011/05/17 11:50:20 kristaps Exp $ */
 /*
  * Copyright (c) 2010, 2011 Kristaps Dzonsons <kristaps@bsd.lv>
  *
@@ -330,6 +330,7 @@ enum mandoc_esc	  mandoc_escape(const char **, const char **, int *);
 
 struct mchars	 *mchars_alloc(void);
 char	 	  mchars_num2char(const char *, size_t);
+int		  mchars_num2uc(const char *, size_t);
 const char	 *mchars_spec2str(struct mchars *, const char *, size_t, size_t *);
 int		  mchars_spec2cp(struct mchars *, const char *, size_t);
 const char	 *mchars_res2str(struct mchars *, const char *, size_t, size_t *);