1 /*++
2 /* NAME
3 /* printable 3
4 /* SUMMARY
5 /* mask non-printable characters
6 /* SYNOPSIS
7 /* #include <stringops.h>
8 /*
9 /* int util_utf8_enable;
10 /*
11 /* char *printable(buffer, replacement)
12 /* char *buffer;
13 /* int replacement;
14 /*
15 /* char *printable_except(buffer, replacement, except)
16 /* char *buffer;
17 /* int replacement;
18 /* const char *except;
19 /* DESCRIPTION
20 /* printable() replaces non-printable characters
21 /* in its input with the given replacement.
22 /*
23 /* util_utf8_enable controls whether UTF8 is considered printable.
24 /* With util_utf8_enable equal to zero, non-ASCII text is replaced.
25 /*
26 /* Arguments:
27 /* .IP buffer
28 /* The null-terminated input string.
29 /* .IP replacement
30 /* Replacement value for characters in \fIbuffer\fR that do not
31 /* pass the ASCII isprint(3) test or that are not valid UTF8.
32 /* .IP except
33 /* Null-terminated sequence of non-replaced ASCII characters.
34 /* LICENSE
35 /* .ad
36 /* .fi
37 /* The Secure Mailer license must be distributed with this software.
38 /* AUTHOR(S)
39 /* Wietse Venema
40 /* IBM T.J. Watson Research
41 /* P.O. Box 704
42 /* Yorktown Heights, NY 10598, USA
43 /*
44 /* Wietse Venema
45 /* Google, Inc.
46 /* 111 8th Avenue
47 /* New York, NY 10011, USA
48 /*--*/
49
50 /* System library. */
51
52 #include "sys_defs.h"
53 #include <ctype.h>
54 #include <string.h>
55
56 /* Utility library. */
57
58 #include "stringops.h"
59
60 int util_utf8_enable = 0;
61
62 /* printable - binary compatibility */
63
64 #undef printable
65
66 char *printable(char *, int);
67
printable(char * string,int replacement)68 char *printable(char *string, int replacement)
69 {
70 return (printable_except(string, replacement, (char *) 0));
71 }
72
73 /* printable_except - pass through printable or other preserved characters */
74
printable_except(char * string,int replacement,const char * except)75 char *printable_except(char *string, int replacement, const char *except)
76 {
77 unsigned char *cp;
78 int ch;
79
80 /*
81 * XXX Replace invalid UTF8 sequences (too short, over-long encodings,
82 * out-of-range code points, etc). See valid_utf8_string.c.
83 */
84 cp = (unsigned char *) string;
85 while ((ch = *cp) != 0) {
86 if (ISASCII(ch) && (ISPRINT(ch) || (except && strchr(except, ch)))) {
87 /* ok */
88 } else if (util_utf8_enable && ch >= 194 && ch <= 254
89 && cp[1] >= 128 && cp[1] < 192) {
90 /* UTF8; skip the rest of the bytes in the character. */
91 while (cp[1] >= 128 && cp[1] < 192)
92 cp++;
93 } else {
94 /* Not ASCII and not UTF8. */
95 *cp = replacement;
96 }
97 cp++;
98 }
99 return (string);
100 }
101