Update vis(3) the latest from NetBSD.

This adds VIS_DQ for compatiblity with OpenBSD.

Correct by an off-by-one error and a read buffer overflow detected using
asan.

MFC after:	1 day
This commit is contained in:
Brooks Davis 2017-11-28 01:35:28 +00:00
commit ff88ef4183
4 changed files with 39 additions and 21 deletions

View File

@ -1,4 +1,4 @@
.\" $NetBSD: unvis.3,v 1.27 2012/12/15 07:34:36 wiz Exp $
.\" $NetBSD: unvis.3,v 1.29 2017/10/24 19:14:55 abhinav Exp $
.\" $FreeBSD$
.\"
.\" Copyright (c) 1989, 1991, 1993
@ -35,7 +35,10 @@
.Os
.Sh NAME
.Nm unvis ,
.Nm strunvis
.Nm strunvis ,
.Nm strnunvis ,
.Nm strunvisx ,
.Nm strnunvisx
.Nd decode a visual representation of characters
.Sh LIBRARY
.Lb libc
@ -184,7 +187,7 @@ char out;
while ((ch = getchar()) != EOF) {
again:
switch(unvis(\*[Am]out, ch, \*[Am]state, 0)) {
switch(unvis(&out, ch, &state, 0)) {
case 0:
case UNVIS_NOCHAR:
break;
@ -198,7 +201,7 @@ again:
errx(EXIT_FAILURE, "Bad character sequence!");
}
}
if (unvis(\*[Am]out, '\e0', \*[Am]state, UNVIS_END) == UNVIS_VALID)
if (unvis(&out, '\e0', &state, UNVIS_END) == UNVIS_VALID)
(void)putchar(out);
.Ed
.Sh ERRORS

View File

@ -1,4 +1,4 @@
.\" $NetBSD: vis.3,v 1.45 2016/06/08 15:00:04 wiz Exp $
.\" $NetBSD: vis.3,v 1.49 2017/08/05 20:22:29 wiz Exp $
.\" $FreeBSD$
.\"
.\" Copyright (c) 1989, 1991, 1993
@ -30,7 +30,7 @@
.\"
.\" @(#)vis.3 8.1 (Berkeley) 6/9/93
.\"
.Dd January 14, 2015
.Dd April 22, 2017
.Dt VIS 3
.Os
.Sh NAME
@ -243,6 +243,8 @@ except space, tab, and newline are encoded (see
The following flags
alter this:
.Bl -tag -width VIS_WHITEX
.It Dv VIS_DQ
Also encode double quotes
.It Dv VIS_GLOB
Also encode the magic characters
.Ql ( * ,
@ -310,7 +312,7 @@ warning on the use of the
.Dv VIS_NOSLASH
flag below).
.Pp
There are four forms of encoding.
There are six forms of encoding.
All forms use the backslash character
.Ql \e
to introduce a special
@ -401,6 +403,9 @@ If
.Fa nextc
is an octal digit, the latter representation is used to
avoid ambiguity.
.Pp
Non-printable characters without C-style
backslash sequences use the default representation.
.It Dv VIS_OCTAL
Use a three digit octal sequence.
The form is
@ -408,6 +413,11 @@ The form is
where
.Em d
represents an octal digit.
.It Dv VIS_CSTYLE \&| Dv VIS_OCTAL
Same as
.Dv VIS_CSTYLE
except that non-printable characters without C-style
backslash sequences use a three digit octal sequence.
.It Dv VIS_HTTPSTYLE
Use URI encoding as described in RFC 1738.
The form is

View File

@ -1,4 +1,4 @@
/* $NetBSD: vis.c,v 1.71 2016/01/14 20:41:23 christos Exp $ */
/* $NetBSD: vis.c,v 1.74 2017/11/27 16:37:21 christos Exp $ */
/*-
* Copyright (c) 1989, 1993
@ -57,7 +57,7 @@
#include <sys/cdefs.h>
#if defined(LIBC_SCCS) && !defined(lint)
__RCSID("$NetBSD: vis.c,v 1.71 2016/01/14 20:41:23 christos Exp $");
__RCSID("$NetBSD: vis.c,v 1.74 2017/11/27 16:37:21 christos Exp $");
#endif /* LIBC_SCCS and not lint */
#ifdef __FBSDID
__FBSDID("$FreeBSD$");
@ -379,6 +379,7 @@ makeextralist(int flags, const char *src)
if (flags & VIS_SP) *d++ = L' ';
if (flags & VIS_TAB) *d++ = L'\t';
if (flags & VIS_NL) *d++ = L'\n';
if (flags & VIS_DQ) *d++ = L'"';
if ((flags & VIS_NOSLASH) == 0) *d++ = L'\\';
*d = L'\0';
@ -408,6 +409,14 @@ istrsenvisx(char **mbdstp, size_t *dlen, const char *mbsrc, size_t mblength,
_DIAGASSERT(mbsrc != NULL || mblength == 0);
_DIAGASSERT(mbextra != NULL);
mbslength = (ssize_t)mblength;
/*
* When inputing a single character, must also read in the
* next character for nextc, the look-ahead character.
*/
if (mbslength == 1)
mbslength++;
/*
* Input (mbsrc) is a char string considered to be multibyte
* characters. The input loop will read this string pulling
@ -424,12 +433,12 @@ istrsenvisx(char **mbdstp, size_t *dlen, const char *mbsrc, size_t mblength,
/* Allocate space for the wide char strings */
psrc = pdst = extra = NULL;
mdst = NULL;
if ((psrc = calloc(mblength + 1, sizeof(*psrc))) == NULL)
if ((psrc = calloc(mbslength + 1, sizeof(*psrc))) == NULL)
return -1;
if ((pdst = calloc((4 * mblength) + 1, sizeof(*pdst))) == NULL)
if ((pdst = calloc((16 * mbslength) + 1, sizeof(*pdst))) == NULL)
goto out;
if (*mbdstp == NULL) {
if ((mdst = calloc((4 * mblength) + 1, sizeof(*mdst))) == NULL)
if ((mdst = calloc((16 * mbslength) + 1, sizeof(*mdst))) == NULL)
goto out;
*mbdstp = mdst;
}
@ -452,13 +461,6 @@ istrsenvisx(char **mbdstp, size_t *dlen, const char *mbsrc, size_t mblength,
* stop at NULs because we may be processing a block of data
* that includes NULs.
*/
mbslength = (ssize_t)mblength;
/*
* When inputing a single character, must also read in the
* next character for nextc, the look-ahead character.
*/
if (mbslength == 1)
mbslength++;
bzero(&mbstate, sizeof(mbstate));
while (mbslength > 0) {
/* Convert one multibyte character to wchar_t. */
@ -470,12 +472,13 @@ istrsenvisx(char **mbdstp, size_t *dlen, const char *mbsrc, size_t mblength,
clen = 1;
cerr = 1;
}
if (clen == 0)
if (clen == 0) {
/*
* NUL in input gives 0 return value. process
* as single NUL byte and keep going.
*/
clen = 1;
}
/* Advance buffer character pointer. */
src++;
/* Advance input pointer by number of bytes read. */
@ -485,6 +488,7 @@ istrsenvisx(char **mbdstp, size_t *dlen, const char *mbsrc, size_t mblength,
}
len = src - psrc;
src = psrc;
/*
* In the single character input case, we will have actually
* processed two characters, c and nextc. Reset len back to

View File

@ -1,4 +1,4 @@
/* $NetBSD: vis.h,v 1.24 2016/01/14 20:42:14 christos Exp $ */
/* $NetBSD: vis.h,v 1.25 2017/04/23 01:57:36 christos Exp $ */
/* $FreeBSD$ */
/*-
@ -52,6 +52,7 @@
#define VIS_NL 0x0010 /* also encode newline */
#define VIS_WHITE (VIS_SP | VIS_TAB | VIS_NL)
#define VIS_SAFE 0x0020 /* only encode "unsafe" characters */
#define VIS_DQ 0x8000 /* also encode double quotes */
/*
* other