Improve surrogate handling readability

- add inline function to test and convert surrogates
  is_surrogate(c), is_hi_surrogate(c), is_lo_surrogate(c),
  get_hi_surrogate(c), get_lo_surrogate(c), from_surrogate(hi, lo)
- use names for BC header offsets and lengths in libregexp.c
- remove strict aliasing violations in `lre_exec_backtrack()`
- pass all context variables to XXX_CHAR macros in `lre_exec_backtrack()`

cutils.h

@@ -45,9 +45,10 @@
 #ifndef countof
 #define countof(x) (sizeof(x) / sizeof((x)[0]))
 #endif
-
+#ifndef container_of
 /* return the pointer of type 'type *' containing 'ptr' as field 'member' */
 #define container_of(ptr, type, member) ((type *)((uint8_t *)(ptr) - offsetof(type, member)))
+#endif
 
 typedef int BOOL;
 
@@ -207,17 +208,22 @@ static inline void put_u8(uint8_t *tab, uint8_t val)
     *tab = val;
 }
 
+#ifndef bswap16
 static inline uint16_t bswap16(uint16_t x)
 {
     return (x >> 8) | (x << 8);
 }
+#endif
 
+#ifndef bswap32
 static inline uint32_t bswap32(uint32_t v)
 {
     return ((v & 0xff000000) >> 24) | ((v & 0x00ff0000) >>  8) |
         ((v & 0x0000ff00) <<  8) | ((v & 0x000000ff) << 24);
 }
+#endif
 
+#ifndef bswap64
 static inline uint64_t bswap64(uint64_t v)
 {
     return ((v & ((uint64_t)0xff << (7 * 8))) >> (7 * 8)) |
@@ -229,6 +235,7 @@ static inline uint64_t bswap64(uint64_t v)
         ((v & ((uint64_t)0xff << (1 * 8))) << (5 * 8)) |
         ((v & ((uint64_t)0xff << (0 * 8))) << (7 * 8));
 }
+#endif
 
 /* XXX: should take an extra argument to pass slack information to the caller */
 typedef void *DynBufReallocFunc(void *opaque, void *ptr, size_t size);
@@ -278,6 +285,36 @@ static inline void dbuf_set_error(DynBuf *s)
 int unicode_to_utf8(uint8_t *buf, unsigned int c);
 int unicode_from_utf8(const uint8_t *p, int max_len, const uint8_t **pp);
 
+static inline BOOL is_surrogate(uint32_t c)
+{
+    return (c >> 11) == (0xD800 >> 11); // 0xD800-0xDFFF
+}
+
+static inline BOOL is_hi_surrogate(uint32_t c)
+{
+    return (c >> 10) == (0xD800 >> 10); // 0xD800-0xDBFF
+}
+
+static inline BOOL is_lo_surrogate(uint32_t c)
+{
+    return (c >> 10) == (0xDC00 >> 10); // 0xDC00-0xDFFF
+}
+
+static inline uint32_t get_hi_surrogate(uint32_t c)
+{
+    return (c >> 10) - (0x10000 >> 10) + 0xD800;
+}
+
+static inline uint32_t get_lo_surrogate(uint32_t c)
+{
+    return (c & 0x3FF) | 0xDC00;
+}
+
+static inline uint32_t from_surrogate(uint32_t hi, uint32_t lo)
+{
+    return 0x10000 + 0x400 * (hi - 0xD800) + (lo - 0xDC00);
+}
+
 static inline int from_hex(int c)
 {
     if (c >= '0' && c <= '9')