[src/trunk]: src/usr.bin/vis Multi-byte docs and fixes for > 2 mblen charsets...

[christos <christos%NetBSD.org@localhost>]

details:   https://anonhg.NetBSD.org/src/rev/350d68579b54
branches:  trunk
changeset: 784902:350d68579b54
user:      christos <christos%NetBSD.org@localhost>
date:      Wed Feb 13 22:24:48 2013 +0000

description:
Multi-byte docs and fixes for > 2 mblen charsets from J.R. Oldroyd

diffstat:

 usr.bin/vis/vis.1 |  30 ++++++++++++++++++++++++++++--
 usr.bin/vis/vis.c |  24 ++++++++++++++++--------
 2 files changed, 44 insertions(+), 10 deletions(-)

diffs (114 lines):

diff -r b2aea053d74a -r 350d68579b54 usr.bin/vis/vis.1
--- a/usr.bin/vis/vis.1 Wed Feb 13 22:19:48 2013 +0000
+++ b/usr.bin/vis/vis.1 Wed Feb 13 22:24:48 2013 +0000
@@ -1,4 +1,4 @@
-.\"    $NetBSD: vis.1,v 1.14 2009/02/10 23:06:31 christos Exp $
+.\"    $NetBSD: vis.1,v 1.15 2013/02/13 22:24:48 christos Exp $
 .\"
 .\" Copyright (c) 1989, 1991, 1993, 1994
 .\"    The Regents of the University of California.  All rights reserved.
@@ -29,7 +29,7 @@
 .\"
 .\"     @(#)vis.1      8.4 (Berkeley) 4/19/94
 .\"
-.Dd February 10, 2009
+.Dd February 13, 2013
 .Dt VIS 1
 .Os
 .Sh NAME
@@ -135,6 +135,28 @@
 White space (space-tab-newline) is also encoded.
 .Pq Dv VIS_WHITE
 .El
+.Sh MULTIBYTE CHARACTER SUPPORT
+.Nm
+supports multibyte character input.
+The encoding conversion is influenced by the setting of the LC_CTYPE
+environment variable which defines the set of characters that can be
+copied without encoding.
+.Pp
+When 8-bit data is present in the input, LC_CTYPE must be set to
+the correct locale or to the C locale.
+If the locales of the data and the conversion are mismatched, multibyte
+character recognition may fail and encoding will be performed byte-by-byte
+instead.
+The result of encoding using
+.Nm
+followed by decoding using
+.Xr unvis 3
+is unlikely to return the same input data in this case.
+.Sh ENVIRONMENT
+.Bl -tag -width ".Ev LC_CTYPE"
+.It Ev LC_CTYPE
+Specify the locale of the input data.
+Set to C if the input data locale is unknown.
 .Sh SEE ALSO
 .Xr unvis 1 ,
 .Xr svis 3 ,
@@ -144,3 +166,7 @@
 .Nm
 command appears in
 .Bx 4.4 .
+Myltibyte character support was added in
+.Nx 6.1
+and
+.Fx 9.2 .
diff -r b2aea053d74a -r 350d68579b54 usr.bin/vis/vis.c
--- a/usr.bin/vis/vis.c Wed Feb 13 22:19:48 2013 +0000
+++ b/usr.bin/vis/vis.c Wed Feb 13 22:24:48 2013 +0000
@@ -1,4 +1,4 @@
-/*     $NetBSD: vis.c,v 1.17 2013/02/13 13:58:44 christos Exp $        */
+/*     $NetBSD: vis.c,v 1.18 2013/02/13 22:24:48 christos Exp $        */
 
 /*-
  * Copyright (c) 1989, 1993
@@ -39,7 +39,7 @@
 #if 0
 static char sccsid[] = "@(#)vis.c      8.1 (Berkeley) 6/6/93";
 #endif
-__RCSID("$NetBSD: vis.c,v 1.17 2013/02/13 13:58:44 christos Exp $");
+__RCSID("$NetBSD: vis.c,v 1.18 2013/02/13 22:24:48 christos Exp $");
 #endif /* not lint */
 
 #include <stdio.h>
@@ -47,6 +47,7 @@
 #include <stdlib.h>
 #include <errno.h>
 #include <wchar.h>
+#include <limits.h>
 #include <unistd.h>
 #include <err.h>
 #include <vis.h>
@@ -159,8 +160,10 @@
        static int col = 0;
        static char nul[] = "\0";
        char *cp = nul + 1;     /* so *(cp-1) starts out != '\n' */
-       wint_t c, rachar; 
-       char buff[5];
+       wint_t c, c1, rachar; 
+       wchar_t ibuff[3]; /* room for c + rachar + NUL */
+       char mbibuff[13]; /* ((sizeof(ibuff) - 1) * MB_LEN_MAX) + NUL */
+       char buff[5]; /* max vis-encoding length for one char + NUL */
        
        c = getwc(fp);
        if (c == WEOF && errno == EILSEQ)
@@ -182,10 +185,15 @@
                        *cp++ = '$';
                        *cp++ = '\n';
                        *cp = '\0';
-               } else if (extra)
-                       (void)svis(buff, c, eflags, rachar, extra);
-               else
-                       (void)vis(buff, c, eflags, rachar);
+               } else {
+                       c1 = rachar;
+                       if (c1 == WEOF)
+                               c1 = L'\0';
+                       swprintf(ibuff, 3, L"%lc%lc", c, c1);
+                       wcstombs(mbibuff, ibuff,
+                           (wcslen(ibuff) * MB_LEN_MAX) + 1);
+                       (void) strsvisx(buff, mbibuff, 1, eflags, extra);
+               }
 
                cp = buff;
                if (fold) {