summaryrefslogtreecommitdiff
path: root/src/shared/utf8.c
diff options
context:
space:
mode:
Diffstat (limited to 'src/shared/utf8.c')
-rw-r--r--src/shared/utf8.c279
1 files changed, 71 insertions, 208 deletions
diff --git a/src/shared/utf8.c b/src/shared/utf8.c
index 1a68394a53..732f0f00ca 100644
--- a/src/shared/utf8.c
+++ b/src/shared/utf8.c
@@ -51,8 +51,6 @@
#include "utf8.h"
#include "util.h"
-#define FILTER_CHAR '_'
-
static inline bool is_unicode_valid(uint32_t ch) {
if (ch >= 0x110000) /* End of unicode space */
@@ -67,17 +65,6 @@ static inline bool is_unicode_valid(uint32_t ch) {
return true;
}
-static inline bool is_continuation_char(uint8_t ch) {
- if ((ch & 0xc0) != 0x80) /* 10xxxxxx */
- return false;
- return true;
-}
-
-static inline void merge_continuation_char(uint32_t *u_ch, uint8_t ch) {
- *u_ch <<= 6;
- *u_ch |= ch & 0x3f;
-}
-
static bool is_unicode_control(uint32_t ch) {
/*
@@ -90,163 +77,97 @@ static bool is_unicode_control(uint32_t ch) {
(0x7F <= ch && ch <= 0x9F);
}
-bool utf8_is_printable(const char* str, size_t length) {
- uint32_t val = 0;
- uint32_t min = 0;
- const uint8_t *p;
+/* count of characters used to encode one unicode char */
+static int utf8_encoded_expected_len(const char *str) {
+ unsigned char c = (unsigned char)str[0];
- assert(str);
+ if (c < 0x80)
+ return 1;
+ if ((c & 0xe0) == 0xc0)
+ return 2;
+ if ((c & 0xf0) == 0xe0)
+ return 3;
+ if ((c & 0xf8) == 0xf0)
+ return 4;
+ if ((c & 0xfc) == 0xf8)
+ return 5;
+ if ((c & 0xfe) == 0xfc)
+ return 6;
+ return 0;
+}
- for (p = (const uint8_t*) str; length; p++, length--) {
- if (*p < 128) {
- val = *p;
- } else {
- if ((*p & 0xe0) == 0xc0) { /* 110xxxxx two-char seq. */
- min = 128;
- val = (uint32_t) (*p & 0x1e);
- goto ONE_REMAINING;
- } else if ((*p & 0xf0) == 0xe0) { /* 1110xxxx three-char seq.*/
- min = (1 << 11);
- val = (uint32_t) (*p & 0x0f);
- goto TWO_REMAINING;
- } else if ((*p & 0xf8) == 0xf0) { /* 11110xxx four-char seq */
- min = (1 << 16);
- val = (uint32_t) (*p & 0x07);
- } else
- return false;
-
- p++;
- length--;
- if (!length || !is_continuation_char(*p))
- return false;
- merge_continuation_char(&val, *p);
-
- TWO_REMAINING:
- p++;
- length--;
- if (!is_continuation_char(*p))
- return false;
- merge_continuation_char(&val, *p);
-
- ONE_REMAINING:
- p++;
- length--;
- if (!is_continuation_char(*p))
- return false;
- merge_continuation_char(&val, *p);
-
- if (val < min)
- return false;
- }
+/* decode one unicode char */
+static int utf8_encoded_to_unichar(const char *str) {
+ int unichar;
+ int len;
+ int i;
- if (is_unicode_control(val))
- return false;
+ len = utf8_encoded_expected_len(str);
+ switch (len) {
+ case 1:
+ return (int)str[0];
+ case 2:
+ unichar = str[0] & 0x1f;
+ break;
+ case 3:
+ unichar = (int)str[0] & 0x0f;
+ break;
+ case 4:
+ unichar = (int)str[0] & 0x07;
+ break;
+ case 5:
+ unichar = (int)str[0] & 0x03;
+ break;
+ case 6:
+ unichar = (int)str[0] & 0x01;
+ break;
+ default:
+ return -1;
}
- return true;
+ for (i = 1; i < len; i++) {
+ if (((int)str[i] & 0xc0) != 0x80)
+ return -1;
+ unichar <<= 6;
+ unichar |= (int)str[i] & 0x3f;
+ }
+
+ return unichar;
}
-static char* utf8_validate(const char *str, char *output) {
- uint32_t val = 0;
- uint32_t min = 0;
- const uint8_t *p, *last;
- int size;
- uint8_t *o;
+bool utf8_is_printable(const char* str, size_t length) {
+ const uint8_t *p;
assert(str);
- o = (uint8_t*) output;
- for (p = (const uint8_t*) str; *p; p++) {
- if (*p < 128) {
- if (o)
- *o = *p;
- } else {
- last = p;
-
- if ((*p & 0xe0) == 0xc0) { /* 110xxxxx two-char seq. */
- size = 2;
- min = 128;
- val = (uint32_t) (*p & 0x1e);
- goto ONE_REMAINING;
- } else if ((*p & 0xf0) == 0xe0) { /* 1110xxxx three-char seq.*/
- size = 3;
- min = (1 << 11);
- val = (uint32_t) (*p & 0x0f);
- goto TWO_REMAINING;
- } else if ((*p & 0xf8) == 0xf0) { /* 11110xxx four-char seq */
- size = 4;
- min = (1 << 16);
- val = (uint32_t) (*p & 0x07);
- } else
- goto error;
-
- p++;
- if (!is_continuation_char(*p))
- goto error;
- merge_continuation_char(&val, *p);
-
- TWO_REMAINING:
- p++;
- if (!is_continuation_char(*p))
- goto error;
- merge_continuation_char(&val, *p);
-
- ONE_REMAINING:
- p++;
- if (!is_continuation_char(*p))
- goto error;
- merge_continuation_char(&val, *p);
-
- if (val < min)
- goto error;
-
- if (!is_unicode_valid(val))
- goto error;
-
- if (o) {
- memcpy(o, last, (size_t) size);
- o += size;
- }
-
- continue;
-
- error:
- if (o) {
- *o = FILTER_CHAR;
- p = last; /* We retry at the next character */
- } else
- goto failure;
- }
+ for (p = (const uint8_t*) str; length; p++) {
+ int encoded_len = utf8_encoded_valid_unichar((const char *)p);
+ int32_t val = utf8_encoded_to_unichar((const char*)p);
- if (o)
- o++;
- }
+ if (encoded_len < 0 || val < 0 || is_unicode_control(val))
+ return false;
- if (o) {
- *o = '\0';
- return output;
+ length -= encoded_len;
}
- return (char*) str;
-
-failure:
- return NULL;
-}
-
-char* utf8_is_valid (const char *str) {
- return utf8_validate(str, NULL);
+ return true;
}
-char* utf8_filter (const char *str) {
- char *new_str;
+const char *utf8_is_valid(const char *str) {
+ const uint8_t *p;
assert(str);
- new_str = malloc(strlen(str) + 1);
- if (!new_str)
- return NULL;
+ for (p = (const uint8_t*) str; *p; ) {
+ int len = utf8_encoded_valid_unichar((const char *)p);
+
+ if (len < 0)
+ return NULL;
+
+ p += len;
+ }
- return utf8_validate(str, new_str);
+ return str;
}
char *ascii_is_valid(const char *str) {
@@ -318,64 +239,6 @@ char *utf16_to_utf8(const void *s, size_t length) {
return r;
}
-/* count of characters used to encode one unicode char */
-static int utf8_encoded_expected_len(const char *str) {
- unsigned char c = (unsigned char)str[0];
-
- if (c < 0x80)
- return 1;
- if ((c & 0xe0) == 0xc0)
- return 2;
- if ((c & 0xf0) == 0xe0)
- return 3;
- if ((c & 0xf8) == 0xf0)
- return 4;
- if ((c & 0xfc) == 0xf8)
- return 5;
- if ((c & 0xfe) == 0xfc)
- return 6;
- return 0;
-}
-
-/* decode one unicode char */
-static int utf8_encoded_to_unichar(const char *str) {
- int unichar;
- int len;
- int i;
-
- len = utf8_encoded_expected_len(str);
- switch (len) {
- case 1:
- return (int)str[0];
- case 2:
- unichar = str[0] & 0x1f;
- break;
- case 3:
- unichar = (int)str[0] & 0x0f;
- break;
- case 4:
- unichar = (int)str[0] & 0x07;
- break;
- case 5:
- unichar = (int)str[0] & 0x03;
- break;
- case 6:
- unichar = (int)str[0] & 0x01;
- break;
- default:
- return -1;
- }
-
- for (i = 1; i < len; i++) {
- if (((int)str[i] & 0xc0) != 0x80)
- return -1;
- unichar <<= 6;
- unichar |= (int)str[i] & 0x3f;
- }
-
- return unichar;
-}
-
/* expected size used to encode one unicode char */
static int utf8_unichar_to_encoded_len(int unichar) {
if (unichar < 0x80)
generated by cgit v1.2.3-54-g00ecf (git 2.45.2) at 2024-11-25 15:29:00 +0000