1 | /* pred.c -- execute the expression tree.
|
---|
2 | Copyright (C) 1990, 1991, 1992, 1993, 1994, 2000,
|
---|
3 | 2003, 2004 Free Software Foundation, Inc.
|
---|
4 |
|
---|
5 | This program is free software; you can redistribute it and/or modify
|
---|
6 | it under the terms of the GNU General Public License as published by
|
---|
7 | the Free Software Foundation; either version 2, or (at your option)
|
---|
8 | any later version.
|
---|
9 |
|
---|
10 | This program is distributed in the hope that it will be useful,
|
---|
11 | but WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
12 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
---|
13 | GNU General Public License for more details.
|
---|
14 |
|
---|
15 | You should have received a copy of the GNU General Public License
|
---|
16 | along with this program; if not, write to the Free Software
|
---|
17 | Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301,
|
---|
18 | USA.
|
---|
19 | */
|
---|
20 |
|
---|
21 | #ifdef HAVE_CONFIG_H
|
---|
22 | #include <config.h>
|
---|
23 | #endif
|
---|
24 |
|
---|
25 | # include <stddef.h>
|
---|
26 | # include <stdlib.h>
|
---|
27 | #include <ctype.h>
|
---|
28 |
|
---|
29 | #if HAVE_STRING_H || STDC_HEADERS
|
---|
30 | #include <string.h>
|
---|
31 | #else
|
---|
32 | #include <strings.h>
|
---|
33 | #endif
|
---|
34 |
|
---|
35 |
|
---|
36 | /* Get mbstate_t, mbrtowc(), mbsinit(), wcwidth(). */
|
---|
37 | #if HAVE_WCHAR_H
|
---|
38 | # include <wchar.h>
|
---|
39 | #endif
|
---|
40 |
|
---|
41 | #include "printquoted.h"
|
---|
42 |
|
---|
43 |
|
---|
44 | /*
|
---|
45 | This comment, IN_CTYPE_DOMAIN and ISPRINT were borrowed from
|
---|
46 | coreutils at Sun Jun 5 21:17:40 2005 UTC.
|
---|
47 |
|
---|
48 | Jim Meyering writes:
|
---|
49 |
|
---|
50 | "... Some ctype macros are valid only for character codes that
|
---|
51 | isascii says are ASCII (SGI's IRIX-4.0.5 is one such system --when
|
---|
52 | using /bin/cc or gcc but without giving an ansi option). So, all
|
---|
53 | ctype uses should be through macros like ISPRINT... If
|
---|
54 | STDC_HEADERS is defined, then autoconf has verified that the ctype
|
---|
55 | macros don't need to be guarded with references to isascii. ...
|
---|
56 | Defining isascii to 1 should let any compiler worth its salt
|
---|
57 | eliminate the && through constant folding."
|
---|
58 |
|
---|
59 | Bruno Haible adds:
|
---|
60 |
|
---|
61 | "... Furthermore, isupper(c) etc. have an undefined result if c is
|
---|
62 | outside the range -1 <= c <= 255. One is tempted to write isupper(c)
|
---|
63 | with c being of type `char', but this is wrong if c is an 8-bit
|
---|
64 | character >= 128 which gets sign-extended to a negative value.
|
---|
65 | The macro ISUPPER protects against this as well." */
|
---|
66 |
|
---|
67 |
|
---|
68 |
|
---|
69 |
|
---|
70 | /* ISPRINT is defined in <sys/euc.h> on at least Solaris2.6 systems. */
|
---|
71 | #undef ISPRINT
|
---|
72 | #define ISPRINT(c) (IN_CTYPE_DOMAIN (c) && isprint (c))
|
---|
73 |
|
---|
74 | #if STDC_HEADERS || (!defined (isascii) && !HAVE_ISASCII)
|
---|
75 | # define IN_CTYPE_DOMAIN(c) 1
|
---|
76 | #else
|
---|
77 | # define IN_CTYPE_DOMAIN(c) isascii(c)
|
---|
78 | #endif
|
---|
79 |
|
---|
80 |
|
---|
81 |
|
---|
82 |
|
---|
83 |
|
---|
84 | /* Convert a possibly-signed character to an unsigned character. This is
|
---|
85 | * a bit safer than casting to unsigned char, since it catches some type
|
---|
86 | * errors that the cast doesn't.
|
---|
87 | *
|
---|
88 | * This code taken from coreutils' system.h header at
|
---|
89 | * Sun Jun 5 21:05:21 2005 UTC.
|
---|
90 | */
|
---|
91 | static inline unsigned char to_uchar (char ch)
|
---|
92 | {
|
---|
93 | return ch;
|
---|
94 | }
|
---|
95 |
|
---|
96 |
|
---|
97 |
|
---|
98 | static size_t
|
---|
99 | unibyte_qmark_chars(char *buf, size_t len)
|
---|
100 | {
|
---|
101 | char *p = buf;
|
---|
102 | char const *plimit = buf + len;
|
---|
103 |
|
---|
104 | while (p < plimit)
|
---|
105 | {
|
---|
106 | if (! ISPRINT (to_uchar (*p)))
|
---|
107 | *p = '?';
|
---|
108 | p++;
|
---|
109 | }
|
---|
110 | return len;
|
---|
111 | }
|
---|
112 |
|
---|
113 |
|
---|
114 | #if HAVE_MBRTOWC
|
---|
115 | static size_t
|
---|
116 | multibyte_qmark_chars(char *buf, size_t len)
|
---|
117 | {
|
---|
118 | if (MB_CUR_MAX <= 1)
|
---|
119 | {
|
---|
120 | return unibyte_qmark_chars(buf, len);
|
---|
121 | }
|
---|
122 | else
|
---|
123 | {
|
---|
124 | char const *p = buf;
|
---|
125 | char const *plimit = buf + len;
|
---|
126 | char *q = buf;
|
---|
127 |
|
---|
128 | while (p < plimit)
|
---|
129 | switch (*p)
|
---|
130 | {
|
---|
131 | case ' ': case '!': case '"': case '#': case '%':
|
---|
132 | case '&': case '\'': case '(': case ')': case '*':
|
---|
133 | case '+': case ',': case '-': case '.': case '/':
|
---|
134 | case '0': case '1': case '2': case '3': case '4':
|
---|
135 | case '5': case '6': case '7': case '8': case '9':
|
---|
136 | case ':': case ';': case '<': case '=': case '>':
|
---|
137 | case '?':
|
---|
138 | case 'A': case 'B': case 'C': case 'D': case 'E':
|
---|
139 | case 'F': case 'G': case 'H': case 'I': case 'J':
|
---|
140 | case 'K': case 'L': case 'M': case 'N': case 'O':
|
---|
141 | case 'P': case 'Q': case 'R': case 'S': case 'T':
|
---|
142 | case 'U': case 'V': case 'W': case 'X': case 'Y':
|
---|
143 | case 'Z':
|
---|
144 | case '[': case '\\': case ']': case '^': case '_':
|
---|
145 | case 'a': case 'b': case 'c': case 'd': case 'e':
|
---|
146 | case 'f': case 'g': case 'h': case 'i': case 'j':
|
---|
147 | case 'k': case 'l': case 'm': case 'n': case 'o':
|
---|
148 | case 'p': case 'q': case 'r': case 's': case 't':
|
---|
149 | case 'u': case 'v': case 'w': case 'x': case 'y':
|
---|
150 | case 'z': case '{': case '|': case '}': case '~':
|
---|
151 | /* These characters are printable ASCII characters. */
|
---|
152 | *q++ = *p++;
|
---|
153 | break;
|
---|
154 | default:
|
---|
155 | /* If we have a multibyte sequence, copy it until we
|
---|
156 | reach its end, replacing each non-printable multibyte
|
---|
157 | character with a single question mark. */
|
---|
158 | {
|
---|
159 | mbstate_t mbstate;
|
---|
160 | memset (&mbstate, 0, sizeof mbstate);
|
---|
161 | do
|
---|
162 | {
|
---|
163 | wchar_t wc;
|
---|
164 | size_t bytes;
|
---|
165 | int w;
|
---|
166 |
|
---|
167 | bytes = mbrtowc (&wc, p, plimit - p, &mbstate);
|
---|
168 |
|
---|
169 | if (bytes == (size_t) -1)
|
---|
170 | {
|
---|
171 | /* An invalid multibyte sequence was
|
---|
172 | encountered. Skip one input byte, and
|
---|
173 | put a question mark. */
|
---|
174 | p++;
|
---|
175 | *q++ = '?';
|
---|
176 | break;
|
---|
177 | }
|
---|
178 |
|
---|
179 | if (bytes == (size_t) -2)
|
---|
180 | {
|
---|
181 | /* An incomplete multibyte character
|
---|
182 | at the end. Replace it entirely with
|
---|
183 | a question mark. */
|
---|
184 | p = plimit;
|
---|
185 | *q++ = '?';
|
---|
186 | break;
|
---|
187 | }
|
---|
188 |
|
---|
189 | if (bytes == 0)
|
---|
190 | /* A null wide character was encountered. */
|
---|
191 | bytes = 1;
|
---|
192 |
|
---|
193 | w = wcwidth (wc);
|
---|
194 | if (w >= 0)
|
---|
195 | {
|
---|
196 | /* A printable multibyte character.
|
---|
197 | Keep it. */
|
---|
198 | for (; bytes > 0; --bytes)
|
---|
199 | *q++ = *p++;
|
---|
200 | }
|
---|
201 | else
|
---|
202 | {
|
---|
203 | /* An unprintable multibyte character.
|
---|
204 | Replace it entirely with a question
|
---|
205 | mark. */
|
---|
206 | p += bytes;
|
---|
207 | *q++ = '?';
|
---|
208 | }
|
---|
209 | }
|
---|
210 | while (! mbsinit (&mbstate));
|
---|
211 | }
|
---|
212 | break;
|
---|
213 | }
|
---|
214 |
|
---|
215 | /* The buffer may have shrunk. */
|
---|
216 | len = q - buf;
|
---|
217 | return len;
|
---|
218 | }
|
---|
219 | }
|
---|
220 | #endif
|
---|
221 |
|
---|
222 |
|
---|
223 | /* Scan BUF, replacing any dangerous-looking characters with question
|
---|
224 | * marks. This code is taken from the ls.c file in coreutils as at
|
---|
225 | * Sun Jun 5 20:51:54 2005 UTC.
|
---|
226 | *
|
---|
227 | * This function may shrink the buffer. Either way, the new length
|
---|
228 | * is returned.
|
---|
229 | */
|
---|
230 | size_t
|
---|
231 | qmark_chars(char *buf, size_t len)
|
---|
232 | {
|
---|
233 | #if HAVE_MBRTOWC
|
---|
234 | return multibyte_qmark_chars(buf, len);
|
---|
235 | #else
|
---|
236 | return unibyte_qmark_chars(buf, len);
|
---|
237 | #endif
|
---|
238 | }
|
---|
239 |
|
---|