38 |
|
|
39 |
|
|
40 |
|
|
41 |
|
static int utf8_table1[] = { |
42 |
|
0x0000007f, 0x000007ff, 0x0000ffff, 0x001fffff, 0x03ffffff, 0x7fffffff}; |
43 |
|
|
44 |
|
static int utf8_table2[] = { |
45 |
|
0, 0xc0, 0xe0, 0xf0, 0xf8, 0xfc}; |
46 |
|
|
47 |
|
static int utf8_table3[] = { |
48 |
|
0xff, 0x1f, 0x0f, 0x07, 0x03, 0x01}; |
49 |
|
|
50 |
|
|
51 |
|
/************************************************* |
52 |
|
* Convert character value to UTF-8 * |
53 |
|
*************************************************/ |
54 |
|
|
55 |
|
/* This function takes an integer value in the range 0 - 0x7fffffff |
56 |
|
and encodes it as a UTF-8 character in 0 to 6 bytes. |
57 |
|
|
58 |
|
Arguments: |
59 |
|
cvalue the character value |
60 |
|
buffer pointer to buffer for result - at least 6 bytes long |
61 |
|
|
62 |
|
Returns: number of characters placed in the buffer |
63 |
|
-1 if input character is negative |
64 |
|
0 if input character is positive but too big (only when |
65 |
|
int is longer than 32 bits) |
66 |
|
*/ |
67 |
|
|
68 |
|
static int |
69 |
|
ord2utf8(int cvalue, unsigned char *buffer) |
70 |
|
{ |
71 |
|
register int i, j; |
72 |
|
for (i = 0; i < sizeof(utf8_table1)/sizeof(int); i++) |
73 |
|
if (cvalue <= utf8_table1[i]) break; |
74 |
|
if (i >= sizeof(utf8_table1)/sizeof(int)) return 0; |
75 |
|
if (cvalue < 0) return -1; |
76 |
|
*buffer++ = utf8_table2[i] | (cvalue & utf8_table3[i]); |
77 |
|
cvalue >>= 6 - i; |
78 |
|
for (j = 0; j < i; j++) |
79 |
|
{ |
80 |
|
*buffer++ = 0x80 | (cvalue & 0x3f); |
81 |
|
cvalue >>= 6; |
82 |
|
} |
83 |
|
return i + 1; |
84 |
|
} |
85 |
|
|
86 |
|
|
87 |
|
/************************************************* |
88 |
|
* Convert UTF-8 string to value * |
89 |
|
*************************************************/ |
90 |
|
|
91 |
|
/* This function takes one or more bytes that represents a UTF-8 character, |
92 |
|
and returns the value of the character. |
93 |
|
|
94 |
|
Argument: |
95 |
|
buffer a pointer to the byte vector |
96 |
|
vptr a pointer to an int to receive the value |
97 |
|
|
98 |
|
Returns: > 0 => the number of bytes consumed |
99 |
|
-6 to 0 => malformed UTF-8 character at offset = (-return) |
100 |
|
*/ |
101 |
|
|
102 |
|
int |
103 |
|
utf82ord(unsigned char *buffer, int *vptr) |
104 |
|
{ |
105 |
|
int c = *buffer++; |
106 |
|
int d = c; |
107 |
|
int i, j, s; |
108 |
|
|
109 |
|
for (i = -1; i < 6; i++) /* i is number of additional bytes */ |
110 |
|
{ |
111 |
|
if ((d & 0x80) == 0) break; |
112 |
|
d <<= 1; |
113 |
|
} |
114 |
|
|
115 |
|
if (i == -1) { *vptr = c; return 1; } /* ascii character */ |
116 |
|
if (i == 0 || i == 6) return 0; /* invalid UTF-8 */ |
117 |
|
|
118 |
|
/* i now has a value in the range 1-5 */ |
119 |
|
|
120 |
|
d = c & utf8_table3[i]; |
121 |
|
s = 6 - i; |
122 |
|
|
123 |
|
for (j = 0; j < i; j++) |
124 |
|
{ |
125 |
|
c = *buffer++; |
126 |
|
if ((c & 0xc0) != 0x80) return -(j+1); |
127 |
|
d |= (c & 0x3f) << s; |
128 |
|
s += 6; |
129 |
|
} |
130 |
|
|
131 |
|
/* Check that encoding was the correct unique one */ |
132 |
|
|
133 |
|
for (j = 0; j < sizeof(utf8_table1)/sizeof(int); j++) |
134 |
|
if (d <= utf8_table1[j]) break; |
135 |
|
if (j != i) return -(i+1); |
136 |
|
|
137 |
|
/* Valid value */ |
138 |
|
|
139 |
|
*vptr = d; |
140 |
|
return i+1; |
141 |
|
} |
142 |
|
|
143 |
|
|
144 |
|
|
145 |
|
|
146 |
|
|
147 |
|
|
148 |
/* Debugging function to print the internal form of the regex. This is the same |
/* Debugging function to print the internal form of the regex. This is the same |
149 |
code as contained in pcre.c under the DEBUG macro. */ |
code as contained in pcre.c under the DEBUG macro. */ |
150 |
|
|
372 |
|
|
373 |
|
|
374 |
|
|
375 |
/* Character string printing function. */ |
/* Character string printing function. A "normal" and a UTF-8 version. */ |
376 |
|
|
377 |
static void pchars(unsigned char *p, int length) |
static void pchars(unsigned char *p, int length, int utf8) |
378 |
{ |
{ |
379 |
int c; |
int c; |
380 |
while (length-- > 0) |
while (length-- > 0) |
381 |
|
{ |
382 |
|
if (utf8) |
383 |
|
{ |
384 |
|
int rc = utf82ord(p, &c); |
385 |
|
if (rc > 0) |
386 |
|
{ |
387 |
|
length -= rc - 1; |
388 |
|
p += rc; |
389 |
|
if (c < 256 && isprint(c)) fprintf(outfile, "%c", c); |
390 |
|
else fprintf(outfile, "\\x{%02x}", c); |
391 |
|
continue; |
392 |
|
} |
393 |
|
} |
394 |
|
|
395 |
|
/* Not UTF-8, or malformed UTF-8 */ |
396 |
|
|
397 |
if (isprint(c = *(p++))) fprintf(outfile, "%c", c); |
if (isprint(c = *(p++))) fprintf(outfile, "%c", c); |
398 |
else fprintf(outfile, "\\x%02x", c); |
else fprintf(outfile, "\\x%02x", c); |
399 |
|
} |
400 |
} |
} |
401 |
|
|
402 |
|
|
527 |
int do_g = 0; |
int do_g = 0; |
528 |
int do_showinfo = showinfo; |
int do_showinfo = showinfo; |
529 |
int do_showrest = 0; |
int do_showrest = 0; |
530 |
|
int utf8 = 0; |
531 |
int erroroffset, len, delimiter; |
int erroroffset, len, delimiter; |
532 |
|
|
533 |
if (infile == stdin) printf(" re> "); |
if (infile == stdin) printf(" re> "); |
619 |
case 'S': do_study = 1; break; |
case 'S': do_study = 1; break; |
620 |
case 'U': options |= PCRE_UNGREEDY; break; |
case 'U': options |= PCRE_UNGREEDY; break; |
621 |
case 'X': options |= PCRE_EXTRA; break; |
case 'X': options |= PCRE_EXTRA; break; |
622 |
|
case '8': options |= PCRE_UTF8; utf8 = 1; break; |
623 |
|
|
624 |
case 'L': |
case 'L': |
625 |
ppp = pp; |
ppp = pp; |
759 |
if (backrefmax > 0) |
if (backrefmax > 0) |
760 |
fprintf(outfile, "Max back reference = %d\n", backrefmax); |
fprintf(outfile, "Max back reference = %d\n", backrefmax); |
761 |
if (options == 0) fprintf(outfile, "No options\n"); |
if (options == 0) fprintf(outfile, "No options\n"); |
762 |
else fprintf(outfile, "Options:%s%s%s%s%s%s%s%s\n", |
else fprintf(outfile, "Options:%s%s%s%s%s%s%s%s%s\n", |
763 |
((options & PCRE_ANCHORED) != 0)? " anchored" : "", |
((options & PCRE_ANCHORED) != 0)? " anchored" : "", |
764 |
((options & PCRE_CASELESS) != 0)? " caseless" : "", |
((options & PCRE_CASELESS) != 0)? " caseless" : "", |
765 |
((options & PCRE_EXTENDED) != 0)? " extended" : "", |
((options & PCRE_EXTENDED) != 0)? " extended" : "", |
767 |
((options & PCRE_DOTALL) != 0)? " dotall" : "", |
((options & PCRE_DOTALL) != 0)? " dotall" : "", |
768 |
((options & PCRE_DOLLAR_ENDONLY) != 0)? " dollar_endonly" : "", |
((options & PCRE_DOLLAR_ENDONLY) != 0)? " dollar_endonly" : "", |
769 |
((options & PCRE_EXTRA) != 0)? " extra" : "", |
((options & PCRE_EXTRA) != 0)? " extra" : "", |
770 |
((options & PCRE_UNGREEDY) != 0)? " ungreedy" : ""); |
((options & PCRE_UNGREEDY) != 0)? " ungreedy" : "", |
771 |
|
((options & PCRE_UTF8) != 0)? " utf8" : ""); |
772 |
|
|
773 |
if (((((real_pcre *)re)->options) & PCRE_ICHANGED) != 0) |
if (((((real_pcre *)re)->options) & PCRE_ICHANGED) != 0) |
774 |
fprintf(outfile, "Case state changes\n"); |
fprintf(outfile, "Case state changes\n"); |
923 |
break; |
break; |
924 |
|
|
925 |
case 'x': |
case 'x': |
926 |
|
|
927 |
|
/* Handle \x{..} specially - new Perl thing for utf8 */ |
928 |
|
|
929 |
|
if (*p == '{') |
930 |
|
{ |
931 |
|
unsigned char *pt = p; |
932 |
|
c = 0; |
933 |
|
while (isxdigit(*(++pt))) |
934 |
|
c = c * 16 + tolower(*pt) - ((isdigit(*pt))? '0' : 'W'); |
935 |
|
if (*pt == '}') |
936 |
|
{ |
937 |
|
unsigned char buffer[8]; |
938 |
|
int ii, utn; |
939 |
|
utn = ord2utf8(c, buffer); |
940 |
|
for (ii = 0; ii < utn - 1; ii++) *q++ = buffer[ii]; |
941 |
|
c = buffer[ii]; /* Last byte */ |
942 |
|
p = pt + 1; |
943 |
|
break; |
944 |
|
} |
945 |
|
/* Not correct form; fall through */ |
946 |
|
} |
947 |
|
|
948 |
|
/* Ordinary \x */ |
949 |
|
|
950 |
c = 0; |
c = 0; |
951 |
while (i++ < 2 && isxdigit(*p)) |
while (i++ < 2 && isxdigit(*p)) |
952 |
{ |
{ |
1027 |
{ |
{ |
1028 |
fprintf(outfile, "%2d: ", (int)i); |
fprintf(outfile, "%2d: ", (int)i); |
1029 |
pchars(dbuffer + pmatch[i].rm_so, |
pchars(dbuffer + pmatch[i].rm_so, |
1030 |
pmatch[i].rm_eo - pmatch[i].rm_so); |
pmatch[i].rm_eo - pmatch[i].rm_so, utf8); |
1031 |
fprintf(outfile, "\n"); |
fprintf(outfile, "\n"); |
1032 |
if (i == 0 && do_showrest) |
if (i == 0 && do_showrest) |
1033 |
{ |
{ |
1034 |
fprintf(outfile, " 0+ "); |
fprintf(outfile, " 0+ "); |
1035 |
pchars(dbuffer + pmatch[i].rm_eo, len - pmatch[i].rm_eo); |
pchars(dbuffer + pmatch[i].rm_eo, len - pmatch[i].rm_eo, utf8); |
1036 |
fprintf(outfile, "\n"); |
fprintf(outfile, "\n"); |
1037 |
} |
} |
1038 |
} |
} |
1082 |
else |
else |
1083 |
{ |
{ |
1084 |
fprintf(outfile, "%2d: ", i/2); |
fprintf(outfile, "%2d: ", i/2); |
1085 |
pchars(bptr + offsets[i], offsets[i+1] - offsets[i]); |
pchars(bptr + offsets[i], offsets[i+1] - offsets[i], utf8); |
1086 |
fprintf(outfile, "\n"); |
fprintf(outfile, "\n"); |
1087 |
if (i == 0) |
if (i == 0) |
1088 |
{ |
{ |
1089 |
if (do_showrest) |
if (do_showrest) |
1090 |
{ |
{ |
1091 |
fprintf(outfile, " 0+ "); |
fprintf(outfile, " 0+ "); |
1092 |
pchars(bptr + offsets[i+1], len - offsets[i+1]); |
pchars(bptr + offsets[i+1], len - offsets[i+1], utf8); |
1093 |
fprintf(outfile, "\n"); |
fprintf(outfile, "\n"); |
1094 |
} |
} |
1095 |
} |
} |
1122 |
else |
else |
1123 |
{ |
{ |
1124 |
fprintf(outfile, "%2dG %s (%d)\n", i, substring, rc); |
fprintf(outfile, "%2dG %s (%d)\n", i, substring, rc); |
1125 |
free((void *)substring); |
/* free((void *)substring); */ |
1126 |
|
pcre_free_substring(substring); |
1127 |
} |
} |
1128 |
} |
} |
1129 |
} |
} |
1141 |
fprintf(outfile, "%2dL %s\n", i, stringlist[i]); |
fprintf(outfile, "%2dL %s\n", i, stringlist[i]); |
1142 |
if (stringlist[i] != NULL) |
if (stringlist[i] != NULL) |
1143 |
fprintf(outfile, "string list not terminated by NULL\n"); |
fprintf(outfile, "string list not terminated by NULL\n"); |
1144 |
free((void *)stringlist); |
/* free((void *)stringlist); */ |
1145 |
|
pcre_free_substring_list(stringlist); |
1146 |
} |
} |
1147 |
} |
} |
1148 |
} |
} |