type: Add missing space for printing complex types.

[cparser] / string_rep.h

diff --git a/string_rep.h b/string_rep.h
index 13a1674..ce4ca3e 100644 (file)
--- a/string_rep.h
+++ b/string_rep.h
@@ -20,66 +20,24 @@
 #ifndef STRING_REP_H
 #define STRING_REP_H
 
-#include <assert.h>
 #include <stdlib.h>
+#include "unicode.h"
+
+enum string_encoding_t {
+       STRING_ENCODING_CHAR,
+       STRING_ENCODING_CHAR16,
+       STRING_ENCODING_CHAR32,
+       STRING_ENCODING_UTF8,
+       STRING_ENCODING_WIDE
+};
+typedef enum string_encoding_t string_encoding_t;
 
 typedef struct string_t {
-       const char *begin; /**< UTF-8 encoded string, the last character is
-                                               * guaranteed to be 0 */
-       size_t      size;  /**< size of string in bytes (not characters) */
+       char const       *begin; /**< UTF-8 encoded string, the last character is guaranteed to be \0. */
+       size_t            size;  /**< size of string in bytes (not characters), without terminating \0. */
+       string_encoding_t encoding;
 } string_t;
 
-typedef unsigned int utf32;
-#define UTF32_PRINTF_FORMAT "%u"
-
-/**
- * "parse" an utf8 character from a string.
- * Warning: This function only works for valid utf-8 inputs. The behaviour
- * is undefined for invalid utf-8 input.
- *
- * @param p    A pointer to a pointer into the string. The pointer
- *             is incremented for each consumed char
- */
-static inline utf32 read_utf8_char(const char **p)
-{
-       const unsigned char *c      = (const unsigned char *) *p;
-       utf32                result;
-
-       if ((*c & 0x80) == 0) {
-               /* 1 character encoding: 0b0??????? */
-               result = *c++;
-       } else if ((*c & 0xE0) == 0xC0) {
-               /* 2 character encoding: 0b110?????, 0b10?????? */
-               result = *c++ & 0x1F;
-               result = (result << 6) | (*c++ & 0x3F);
-       } else if ((*c & 0xF0) == 0xE0) {
-               /* 3 character encoding: 0b1110????, 0b10??????, 0b10?????? */
-               result = *c++ & 0x0F;
-               result = (result << 6) | (*c++ & 0x3F);
-               result = (result << 6) | (*c++ & 0x3F);
-       } else {
-               /* 4 character enc.: 0b11110???, 0b10??????, 0b10??????, 0b10?????? */
-               assert((*c & 0xF8) == 0xF0);
-               result = *c++ & 0x07;
-               result = (result << 6) | (*c++ & 0x3F);
-               result = (result << 6) | (*c++ & 0x3F);
-               result = (result << 6) | (*c++ & 0x3F);
-       }
-
-       *p = (const char*) c;
-       return result;
-}
-
-static inline size_t wstrlen(const string_t *string)
-{
-       size_t      result = 0;
-       const char *p      = string->begin;
-       const char *end    = p + string->size;
-       while (p < end) {
-               read_utf8_char(&p);
-               ++result;
-       }
-       return result;
-}
+size_t get_string_len(string_t const *str);
 
 #endif