Originally we only supported ucs2, so move the ucs4 version from libsystemd-terminal to shared
and use that everywhere.
#include "sd-event.h"
#include "term-internal.h"
#include "util.h"
+#include "utf8.h"
typedef struct Output Output;
typedef struct Terminal Terminal;
output_printf(o, " ");
} else {
for (k = 0; k < n_ch; ++k) {
- ulen = term_utf8_encode(utf8, ch[k]);
+ ulen = utf8_encode_unichar(utf8, ch[k]);
output_write(o, utf8, ulen);
}
}
assert(t);
- len = term_utf8_encode(buf, ucs4);
+ len = utf8_encode_unichar(buf, ucs4);
if (len < 1)
return 0;
*bg = b;
}
-/**
- * term_utf8_encode() - Encode single UCS-4 character as UTF-8
- * @out_utf8: output buffer of at least 4 bytes or NULL
- * @g: UCS-4 character to encode
- *
- * This encodes a single UCS-4 character as UTF-8 and writes it into @out_utf8.
- * The length of the character is returned. It is not zero-terminated! If the
- * output buffer is NULL, only the length is returned.
- *
- * Returns: The length in bytes that the UTF-8 representation does or would
- * occupy.
- */
-size_t term_utf8_encode(char *out_utf8, uint32_t g) {
- if (g < (1 << 7)) {
- if (out_utf8)
- out_utf8[0] = g & 0x7f;
- return 1;
- } else if (g < (1 << 11)) {
- if (out_utf8) {
- out_utf8[0] = 0xc0 | ((g >> 6) & 0x1f);
- out_utf8[1] = 0x80 | (g & 0x3f);
- }
- return 2;
- } else if (g < (1 << 16)) {
- if (out_utf8) {
- out_utf8[0] = 0xe0 | ((g >> 12) & 0x0f);
- out_utf8[1] = 0x80 | ((g >> 6) & 0x3f);
- out_utf8[2] = 0x80 | (g & 0x3f);
- }
- return 3;
- } else if (g < (1 << 21)) {
- if (out_utf8) {
- out_utf8[0] = 0xf0 | ((g >> 18) & 0x07);
- out_utf8[1] = 0x80 | ((g >> 12) & 0x3f);
- out_utf8[2] = 0x80 | ((g >> 6) & 0x3f);
- out_utf8[3] = 0x80 | (g & 0x3f);
- }
- return 4;
- } else {
- return 0;
- }
-}
-
/**
* term_utf8_decode() - Try decoding the next UCS-4 character
* @p: decoder object to operate on or NULL
#include "macro.h"
#include "term-internal.h"
#include "util.h"
+#include "utf8.h"
int term_screen_new(term_screen **out, term_screen_write_fn write_fn, void *write_fn_data, term_screen_cmd_fn cmd_fn, void *cmd_fn_data) {
_cleanup_(term_screen_unrefp) term_screen *screen = NULL;
/* map unicode keys */
for (i = 0; i < n_syms; ++i)
- p += term_utf8_encode(p, ucs4[i]);
+ p += utf8_encode_unichar(p, ucs4[i]);
return p;
}
unsigned int valid : 1;
};
-size_t term_utf8_encode(char *out_utf8, uint32_t g);
size_t term_utf8_decode(term_utf8 *p, uint32_t **out_buf, char c);
/*
#include "macro.h"
#include "term-internal.h"
#include "util.h"
+#include "utf8.h"
static void test_term_utf8_invalid(void) {
term_utf8 p = { };
/* Convert all ucs-4 chars to utf-8 and back */
for (i = 0; i < 0x10FFFF; ++i) {
- ulen = term_utf8_encode(u8, i);
+ ulen = utf8_encode_unichar(u8, i);
if (!ulen)
continue;
if (!GREEDY_REALLOC(s, allocated, n + 4))
return -ENOMEM;
- n += utf8_encode_unichar(x, s + n);
+ n += utf8_encode_unichar(s + n, x);
c += 5;
continue;
} else
return (char*) str;
}
-int utf8_encode_unichar(uint16_t c, char *p) {
- uint8_t *t = (uint8_t*) p;
-
- if (c < 0x80) {
- t[0] = (uint8_t) c;
+/**
+ * utf8_encode_unichar() - Encode single UCS-4 character as UTF-8
+ * @out_utf8: output buffer of at least 4 bytes or NULL
+ * @g: UCS-4 character to encode
+ *
+ * This encodes a single UCS-4 character as UTF-8 and writes it into @out_utf8.
+ * The length of the character is returned. It is not zero-terminated! If the
+ * output buffer is NULL, only the length is returned.
+ *
+ * Returns: The length in bytes that the UTF-8 representation does or would
+ * occupy.
+ */
+size_t utf8_encode_unichar(char *out_utf8, uint32_t g) {
+ if (g < (1 << 7)) {
+ if (out_utf8)
+ out_utf8[0] = g & 0x7f;
return 1;
- } else if (c < 0x800) {
- t[0] = (uint8_t) (0xc0 | (c >> 6));
- t[1] = (uint8_t) (0x80 | (c & 0x3f));
+ } else if (g < (1 << 11)) {
+ if (out_utf8) {
+ out_utf8[0] = 0xc0 | ((g >> 6) & 0x1f);
+ out_utf8[1] = 0x80 | (g & 0x3f);
+ }
return 2;
- } else {
- t[0] = (uint8_t) (0xe0 | (c >> 12));
- t[1] = (uint8_t) (0x80 | ((c >> 6) & 0x3f));
- t[2] = (uint8_t) (0x80 | (c & 0x3f));
+ } else if (g < (1 << 16)) {
+ if (out_utf8) {
+ out_utf8[0] = 0xe0 | ((g >> 12) & 0x0f);
+ out_utf8[1] = 0x80 | ((g >> 6) & 0x3f);
+ out_utf8[2] = 0x80 | (g & 0x3f);
+ }
return 3;
+ } else if (g < (1 << 21)) {
+ if (out_utf8) {
+ out_utf8[0] = 0xf0 | ((g >> 18) & 0x07);
+ out_utf8[1] = 0x80 | ((g >> 12) & 0x3f);
+ out_utf8[2] = 0x80 | ((g >> 6) & 0x3f);
+ out_utf8[3] = 0x80 | (g & 0x3f);
+ }
+ return 4;
+ } else {
+ return 0;
}
}
return NULL;
for (f = s, t = r; f < (const uint8_t*) s + length; f += 2)
- t += utf8_encode_unichar((f[1] << 8) | f[0], t);
+ t += utf8_encode_unichar(t, (f[1] << 8) | f[0]);
*t = 0;
return r;
char *utf8_escape_invalid(const char *s);
char *utf8_escape_non_printable(const char *str);
-int utf8_encode_unichar(uint16_t c, char *p);
+size_t utf8_encode_unichar(char *out_utf8, uint32_t g);
char *utf16_to_utf8(const void *s, size_t length);
int utf8_encoded_valid_unichar(const char *str);
test_one("{\"foo\" : [true, false]}", JSON_OBJECT_OPEN, JSON_STRING, "foo", JSON_COLON, JSON_ARRAY_OPEN, JSON_BOOLEAN, true, JSON_COMMA, JSON_BOOLEAN, false, JSON_ARRAY_CLOSE, JSON_OBJECT_CLOSE, JSON_END);
test_one("\"\xef\xbf\xbd\"", JSON_STRING, "\xef\xbf\xbd", JSON_END);
test_one("\"\\ufffd\"", JSON_STRING, "\xef\xbf\xbd", JSON_END);
+ test_one("\"\\uf\"", -EINVAL);
return 0;
}