===================================================================
RCS file: /cvsrepo/anoncvs/cvs/src/usr.bin/mandoc/term.c,v
retrieving revision 1.92
retrieving revision 1.93
diff -c -r1.92 -r1.93
*** src/usr.bin/mandoc/term.c	2014/10/28 18:48:56	1.92
--- src/usr.bin/mandoc/term.c	2014/10/29 00:17:01	1.93
***************
*** 1,4 ****
! /*	$OpenBSD: term.c,v 1.92 2014/10/28 18:48:56 schwarze Exp $ */
  /*
   * Copyright (c) 2008, 2009, 2010, 2011 Kristaps Dzonsons <kristaps@bsd.lv>
   * Copyright (c) 2010-2014 Ingo Schwarze <schwarze@openbsd.org>
--- 1,4 ----
! /*	$OpenBSD: term.c,v 1.93 2014/10/29 00:17:01 schwarze Exp $ */
  /*
   * Copyright (c) 2008, 2009, 2010, 2011 Kristaps Dzonsons <kristaps@bsd.lv>
   * Copyright (c) 2010-2014 Ingo Schwarze <schwarze@openbsd.org>
***************
*** 389,395 ****
  {
  	const char	 nbrsp[2] = { ASCII_NBRSP, 0 };
  	const char	*seq, *cp;
- 	char		 c;
  	int		 sz, uc;
  	size_t		 ssz;
  	enum mandoc_esc	 esc;
--- 389,394 ----
***************
*** 441,456 ****
  		switch (esc) {
  		case ESCAPE_UNICODE:
  			uc = mchars_num2uc(seq + 1, sz - 1);
- 			if (p->enc == TERMENC_ASCII) {
- 				cp = ascii_uc2str(uc);
- 				encode(p, cp, strlen(cp));
- 			} else
- 				encode1(p, uc);
  			break;
  		case ESCAPE_NUMBERED:
! 			c = mchars_num2char(seq, sz);
! 			if ('\0' != c)
! 				encode(p, &c, 1);
  			break;
  		case ESCAPE_SPECIAL:
  			if (p->enc == TERMENC_ASCII) {
--- 440,450 ----
  		switch (esc) {
  		case ESCAPE_UNICODE:
  			uc = mchars_num2uc(seq + 1, sz - 1);
  			break;
  		case ESCAPE_NUMBERED:
! 			uc = mchars_num2char(seq, sz);
! 			if (uc < 0)
! 				continue;
  			break;
  		case ESCAPE_SPECIAL:
  			if (p->enc == TERMENC_ASCII) {
***************
*** 463,498 ****
  				if (uc > 0)
  					encode1(p, uc);
  			}
! 			break;
  		case ESCAPE_FONTBOLD:
  			term_fontrepl(p, TERMFONT_BOLD);
! 			break;
  		case ESCAPE_FONTITALIC:
  			term_fontrepl(p, TERMFONT_UNDER);
! 			break;
  		case ESCAPE_FONTBI:
  			term_fontrepl(p, TERMFONT_BI);
! 			break;
  		case ESCAPE_FONT:
  			/* FALLTHROUGH */
  		case ESCAPE_FONTROMAN:
  			term_fontrepl(p, TERMFONT_NONE);
! 			break;
  		case ESCAPE_FONTPREV:
  			term_fontlast(p);
! 			break;
  		case ESCAPE_NOSPACE:
  			if (TERMP_SKIPCHAR & p->flags)
  				p->flags &= ~TERMP_SKIPCHAR;
  			else if ('\0' == *word)
  				p->flags |= TERMP_NOSPACE;
! 			break;
  		case ESCAPE_SKIPCHAR:
  			p->flags |= TERMP_SKIPCHAR;
! 			break;
  		default:
! 			break;
  		}
  	}
  	p->flags &= ~TERMP_NBRWORD;
  }
--- 457,507 ----
  				if (uc > 0)
  					encode1(p, uc);
  			}
! 			continue;
  		case ESCAPE_FONTBOLD:
  			term_fontrepl(p, TERMFONT_BOLD);
! 			continue;
  		case ESCAPE_FONTITALIC:
  			term_fontrepl(p, TERMFONT_UNDER);
! 			continue;
  		case ESCAPE_FONTBI:
  			term_fontrepl(p, TERMFONT_BI);
! 			continue;
  		case ESCAPE_FONT:
  			/* FALLTHROUGH */
  		case ESCAPE_FONTROMAN:
  			term_fontrepl(p, TERMFONT_NONE);
! 			continue;
  		case ESCAPE_FONTPREV:
  			term_fontlast(p);
! 			continue;
  		case ESCAPE_NOSPACE:
  			if (TERMP_SKIPCHAR & p->flags)
  				p->flags &= ~TERMP_SKIPCHAR;
  			else if ('\0' == *word)
  				p->flags |= TERMP_NOSPACE;
! 			continue;
  		case ESCAPE_SKIPCHAR:
  			p->flags |= TERMP_SKIPCHAR;
! 			continue;
  		default:
! 			continue;
  		}
+ 
+ 		/*
+ 		 * Common handling for Unicode and numbered
+ 		 * character escape sequences.
+ 		 */
+ 
+ 		if (p->enc == TERMENC_ASCII) {
+ 			cp = ascii_uc2str(uc);
+ 			encode(p, cp, strlen(cp));
+ 		} else {
+ 			if ((uc < 0x20 && uc != 0x09) ||
+ 			    (uc > 0x7E && uc < 0xA0))
+ 				uc = 0xFFFD;
+ 			encode1(p, uc);
+ 		}
  	}
  	p->flags &= ~TERMP_NBRWORD;
  }
***************
*** 643,649 ****
  term_strlen(const struct termp *p, const char *cp)
  {
  	size_t		 sz, rsz, i;
! 	int		 ssz, skip, c;
  	const char	*seq, *rhs;
  	enum mandoc_esc	 esc;
  	static const char rej[] = { '\\', ASCII_NBRSP, ASCII_HYPH,
--- 652,658 ----
  term_strlen(const struct termp *p, const char *cp)
  {
  	size_t		 sz, rsz, i;
! 	int		 ssz, skip, uc;
  	const char	*seq, *rhs;
  	enum mandoc_esc	 esc;
  	static const char rej[] = { '\\', ASCII_NBRSP, ASCII_HYPH,
***************
*** 673,715 ****
  
  			switch (esc) {
  			case ESCAPE_UNICODE:
! 				c = mchars_num2uc(seq + 1, sz - 1);
! 				if (p->enc == TERMENC_ASCII) {
! 					rhs = ascii_uc2str(c);
! 					rsz = strlen(rhs);
! 				} else
! 					sz += cond_width(p, c, &skip);
  				break;
  			case ESCAPE_NUMBERED:
! 				c = mchars_num2char(seq, ssz);
! 				if ('\0' != c)
! 					sz += cond_width(p, c, &skip);
  				break;
  			case ESCAPE_SPECIAL:
! 				if (p->enc == TERMENC_ASCII)
  					rhs = mchars_spec2str(p->symtab,
  					    seq, ssz, &rsz);
! 				else {
! 					c = mchars_spec2cp(p->symtab,
  					    seq, ssz);
! 					if (c > 0)
! 						sz += cond_width(p, c, &skip);
  				}
! 				break;
  			case ESCAPE_SKIPCHAR:
  				skip = 1;
! 				break;
  			default:
! 				break;
  			}
  
! 			if (NULL == rhs)
! 				break;
  
  			if (skip) {
  				skip = 0;
  				break;
  			}
  
  			for (i = 0; i < rsz; i++)
  				sz += (*p->width)(p, *rhs++);
--- 682,741 ----
  
  			switch (esc) {
  			case ESCAPE_UNICODE:
! 				uc = mchars_num2uc(seq + 1, sz - 1);
  				break;
  			case ESCAPE_NUMBERED:
! 				uc = mchars_num2char(seq, ssz);
! 				if (uc < 0)
! 					continue;
  				break;
  			case ESCAPE_SPECIAL:
! 				if (p->enc == TERMENC_ASCII) {
  					rhs = mchars_spec2str(p->symtab,
  					    seq, ssz, &rsz);
! 					if (rhs != NULL)
! 						break;
! 				} else {
! 					uc = mchars_spec2cp(p->symtab,
  					    seq, ssz);
! 					if (uc > 0)
! 						sz += cond_width(p, uc, &skip);
  				}
! 				continue;
  			case ESCAPE_SKIPCHAR:
  				skip = 1;
! 				continue;
  			default:
! 				continue;
  			}
  
! 			/*
! 			 * Common handling for Unicode and numbered
! 			 * character escape sequences.
! 			 */
  
+ 			if (rhs == NULL) {
+ 				if (p->enc == TERMENC_ASCII) {
+ 					rhs = ascii_uc2str(uc);
+ 					rsz = strlen(rhs);
+ 				} else {
+ 					if ((uc < 0x20 && uc != 0x09) ||
+ 					    (uc > 0x7E && uc < 0xA0))
+ 						uc = 0xFFFD;
+ 					sz += cond_width(p, uc, &skip);
+ 					continue;
+ 				}
+ 			}
+ 
  			if (skip) {
  				skip = 0;
  				break;
  			}
+ 
+ 			/*
+ 			 * Common handling for all escape sequences
+ 			 * printing more than one character.
+ 			 */
  
  			for (i = 0; i < rsz; i++)
  				sz += (*p->width)(p, *rhs++);