/*
 * Finds the next whitespace-delimited word within the 'in' string.
 * Returns a pointer to the first character of the word, and a pointer
 * to the next byte after the last character in the word (in *end).
 */
static char *
findwrd(char *in, char **end)
{
	char	   *start;

	/* Skip leading spaces */
	while (*in && t_isspace(in))
		in += pg_mblen(in);

	/* Return NULL on empty lines */
	if (*in == '\0')
	{
		*end = NULL;
		return NULL;
	}

	start = in;

	/* Find end of word */
	while (*in && !t_isspace(in))
		in += pg_mblen(in);

	*end = in;
	return start;
}

Datum
dabs_int_lexize(PG_FUNCTION_ARGS)
{
	char *in = (char *) PG_GETARG_POINTER(1);
	char *out = pnstrdup(in, PG_GETARG_INT32(2));
	char *start;
	char *end;
	TSLexeme *res = palloc0(sizeof(TSLexeme) * 2);

	res[1].lexeme = NULL;

	while (*out && t_iseq(out, '-'))
  		out += pg_mblen(out);

	start = out;

	while (*out)
    		out += pg_mblen(out);

	end = out;
	out = pnstrdup(start, end - start);

  res[0].lexeme = out;

	PG_RETURN_POINTER(res);
}

static bool
mb_strchr(char *str, char *c)
{
	int			clen,
				plen,
				i;
	char	   *ptr = str;
	bool		res = false;

	clen = pg_mblen(c);
	while (*ptr && !res)
	{
		plen = pg_mblen(ptr);
		if (plen == clen)
		{
			i = plen;
			res = true;
			while (i--)
				if (*(ptr + i) != *(c + i))
				{
					res = false;
					break;
				}
		}

		ptr += plen;
	}

	return res;
}

static int pgq_quote_literal(char *dst, const uint8 *src, int srclen)
{
    const uint8 *cp1 = src, *src_end = src + srclen;
    char *cp2 = dst;
    bool is_ext = false;

    *cp2++ = '\'';
    while (cp1 < src_end) {
        int wl = pg_mblen((const char *)cp1);
        if (wl != 1) {
            while (wl-- > 0 && cp1 < src_end)
                *cp2++ = *cp1++;
            continue;
        }

        if (*cp1 == '\'') {
            *cp2++ = '\'';
        } else if (*cp1 == '\\') {
            if (!is_ext) {
                /* make room for 'E' */
                memmove(dst + 1, dst, cp2 - dst);
                *dst = 'E';
                is_ext = true;
                cp2++;
            }
            *cp2++ = '\\';
        }
        *cp2++ = *cp1++;
    }
    *cp2++ = '\'';

    return cp2 - dst;
}

/*
 * returns the common prefix length of a node key
*/
static int32
gbt_var_node_cp_len(const GBT_VARKEY *node, const gbtree_vinfo *tinfo)
{

	GBT_VARKEY_R r = gbt_var_key_readable(node);
	int32		i = 0;
	int32		l = 0;
	int32		t1len = VARSIZE(r.lower) - VARHDRSZ;
	int32		t2len = VARSIZE(r.upper) - VARHDRSZ;
	int32		ml = Min(t1len, t2len);

	char	   *p1 = VARDATA(r.lower);
	char	   *p2 = VARDATA(r.upper);

	if (ml == 0)
		return 0;

	while (i < ml)
	{
		if (tinfo->eml > 1 && l == 0)
		{

			if ((l = pg_mblen(p1)) != pg_mblen(p2))
			{
				return i;
			}
		}
		if (*p1 != *p2)
		{
			if (tinfo->eml > 1)
			{
				return (i - l + 1);
			}
			else
			{
				return i;
			}
		}

		p1++;
		p2++;
		l--;
		i++;
	}
	return (ml);				/* lower == upper */
}

bool
RS_execute(Regis *r, char *str)
{
	RegisNode  *ptr = r->node;
	char	   *c = str;
	int			len = 0;

	while (*c)
	{
		len++;
		c += pg_mblen(c);
	}

	if (len < r->nchar)
		return 0;

	c = str;
	if (r->issuffix)
	{
		len -= r->nchar;
		while (len-- > 0)
			c += pg_mblen(c);
	}


	while (ptr)
	{
		switch (ptr->type)
		{
			case RSF_ONEOF:
				if (!mb_strchr((char *) ptr->data, c))
					return false;
				break;
			case RSF_NONEOF:
				if (mb_strchr((char *) ptr->data, c))
					return false;
				break;
			default:
				elog(ERROR, "unrecognized regis node type: %d", ptr->type);
		}
		ptr = ptr->next;
		c += pg_mblen(c);
	}

	return true;
}

/*
 * Try to match the given source text to a single-quoted literal.
 * If successful, adjust newcursorpos to correspond to the character
 * (not byte) index corresponding to cursorpos in the source text.
 *
 * At entry, literal points just past a ' character.  We must check for the
 * trailing quote.
 */
static bool
match_prosrc_to_literal(const char *prosrc, const char *literal,
						int cursorpos, int *newcursorpos)
{
	int			newcp = cursorpos;
	int			chlen;

	/*
	 * This implementation handles backslashes and doubled quotes in the
	 * string literal.	It does not handle the SQL syntax for literals
	 * continued across line boundaries.
	 *
	 * We do the comparison a character at a time, not a byte at a time, so
	 * that we can do the correct cursorpos math.
	 */
	while (*prosrc)
	{
		cursorpos--;			/* characters left before cursor */

		/*
		 * Check for backslashes and doubled quotes in the literal; adjust
		 * newcp when one is found before the cursor.
		 */
		if (*literal == '\\')
		{
			literal++;
			if (cursorpos > 0)
				newcp++;
		}
		else if (*literal == '\'')
		{
			if (literal[1] != '\'')
				goto fail;
			literal++;
			if (cursorpos > 0)
				newcp++;
		}
		chlen = pg_mblen(prosrc);
		if (strncmp(prosrc, literal, chlen) != 0)
			goto fail;
		prosrc += chlen;
		literal += chlen;
	}

	if (*literal == '\'' && literal[1] != '\'')
	{
		/* success */
		*newcursorpos = newcp;
		return true;
	}

fail:
	/* Must set *newcursorpos to suppress compiler warning */
	*newcursorpos = newcp;
	return false;
}

static char *
getlexeme(char *start, char *end, int *len)
{
	char	   *ptr;
	int			charlen;

	while (start < end && (charlen = pg_mblen(start)) == 1 && t_iseq(start, '_'))
		start += charlen;

	ptr = start;
	if (ptr >= end)
		return NULL;

	while (ptr < end && !((charlen = pg_mblen(ptr)) == 1 && t_iseq(ptr, '_')))
		ptr += charlen;

	*len = ptr - start;
	return start;
}

/*
 * Adds trigrams from words (already padded).
 */
static trgm *
make_trigrams(trgm *tptr, char *str, int bytelen, int charlen)
{
	char	   *ptr = str;

	if (charlen < 3)
		return tptr;

	if (bytelen > charlen)
	{
		/* Find multibyte character boundaries and apply compact_trigram */
		int			lenfirst = pg_mblen(str),
					lenmiddle = pg_mblen(str + lenfirst),
					lenlast = pg_mblen(str + lenfirst + lenmiddle);

		while ((ptr - str) + lenfirst + lenmiddle + lenlast <= bytelen)
		{
			compact_trigram(tptr, ptr, lenfirst + lenmiddle + lenlast);

			ptr += lenfirst;
			tptr++;

			lenfirst = lenmiddle;
			lenmiddle = lenlast;
			lenlast = pg_mblen(ptr + lenfirst + lenmiddle);
		}
	}
	else
	{
		/* Fast path when there are no multibyte characters */
		Assert(bytelen == charlen);

		while (ptr - str < bytelen - 2 /* number of trigrams = strlen - 2 */ )
		{
			CPTRGM(tptr, ptr);
			ptr++;
			tptr++;
		}
	}

	return tptr;
}

/*
 * Adds trigrams from words (already padded).
 */
static trgm *
make_trigrams(trgm *tptr, char *str, int bytelen, int charlen)
{
	char	   *ptr = str;

	if (charlen < 3)
		return tptr;

#ifdef USE_WIDE_UPPER_LOWER
	if (pg_database_encoding_max_length() > 1)
	{
		int			lenfirst = pg_mblen(str),
					lenmiddle = pg_mblen(str + lenfirst),
					lenlast = pg_mblen(str + lenfirst + lenmiddle);

		while ((ptr - str) + lenfirst + lenmiddle + lenlast <= bytelen)
		{
			cnt_trigram(tptr, ptr, lenfirst + lenmiddle + lenlast);

			ptr += lenfirst;
			tptr++;

			lenfirst = lenmiddle;
			lenmiddle = lenlast;
			lenlast = pg_mblen(ptr + lenfirst + lenmiddle);
		}
	}
	else
#endif
	{
		Assert(bytelen == charlen);

		while (ptr - str < bytelen - 2 /* number of trigrams = strlen - 2 */ )
		{
			CPTRGM(tptr, ptr);
			ptr++;
			tptr++;
		}
	}

	return tptr;
}

/*
 * Finds first word in string, returns pointer to the word,
 * endword points to the character after word
 */
static char *
find_word(char *str, int lenstr, char **endword, int *charlen)
{
	char	   *beginword = str;

	while (beginword - str < lenstr && !iswordchr(beginword))
		beginword += pg_mblen(beginword);

	if (beginword - str >= lenstr)
		return NULL;

	*endword = beginword;
	*charlen = 0;
	while (*endword - str < lenstr && iswordchr(*endword))
	{
		*endword += pg_mblen(*endword);
		(*charlen)++;
	}

	return beginword;
}

/*--------------------
 * Support routine for MatchText. Compares given multibyte streams
 * as wide characters. If they match, returns 1 otherwise returns 0.
 *--------------------
 */
static inline int
wchareq(char *p1, char *p2)
{
	int			p1_len;

	/* Optimization:  quickly compare the first byte. */
	if (*p1 != *p2)
		return 0;

	p1_len = pg_mblen(p1);
	if (pg_mblen(p2) != p1_len)
		return 0;

	/* They are the same length */
	while (p1_len--)
	{
		if (*p1++ != *p2++)
			return 0;
	}
	return 1;
}

/*
 * Finds the next whitespace-delimited word within the 'in' string.
 * Returns a pointer to the first character of the word, and a pointer
 * to the next byte after the last character in the word (in *end).
 * Character '*' at the end of word will not be threated as word
 * charater if flags is not null.
 */
static char *
findwrd(char *in, char **end, uint16 *flags)
{
	char	   *start;
	char	   *lastchar;

	/* Skip leading spaces */
	while (*in && t_isspace(in))
		in += pg_mblen(in);

	/* Return NULL on empty lines */
	if (*in == '\0')
	{
		*end = NULL;
		return NULL;
	}

	lastchar = start = in;

	/* Find end of word */
	while (*in && !t_isspace(in))
	{
		lastchar = in;
		in += pg_mblen(in);
	}

	if (in - lastchar == 1 && t_iseq(lastchar, '*') && flags)
	{
		*flags = TSL_PREFIX;
		*end = lastchar;
	}
	else
	{
		if (flags)
			*flags = 0;
		*end = in;
	}

	return start;
}

/*
 * Extract a single, possibly multi-byte char from the input string.
 */
static char *
extract_mb_char(char *s)
{
	char   *res;
	int		len;

	len = pg_mblen(s);
	res = palloc(len + 1);
	memcpy(res, s, len);
	res[len] = '\0';

	return res;
}

int
t_isprint(const char *ptr)
{
	int			clen = pg_mblen(ptr);
	wchar_t		character[2];

	if (clen == 1 || lc_ctype_is_c())
		return isprint(TOUCHAR(ptr));

	char2wchar(character, 2, ptr, clen);

	return iswprint((wint_t) character[0]);
}

/* returns the length (counted as a wchar) of a multibyte string
   (not necessarily  NULL terminated) */
int
pg_mbstrlen_with_len(const unsigned char *mbstr, int limit)
{
	int			len = 0;
	int			l;

	while (limit > 0 && *mbstr)
	{
		l = pg_mblen(mbstr);
		limit -= l;
		mbstr += l;
		len++;
	}
	return (len);
}

int
t_isprint(const char *ptr)
{
	int			clen = pg_mblen(ptr);
	wchar_t		character[2];
	Oid			collation = DEFAULT_COLLATION_OID;		/* TODO */
	pg_locale_t mylocale = 0;	/* TODO */

	if (clen == 1 || lc_ctype_is_c(collation))
		return isprint(TOUCHAR(ptr));

	char2wchar(character, 2, ptr, clen, mylocale);

	return iswprint((wint_t) character[0]);
}

/* returns the length (counted in wchars) of a multibyte string */
int
pg_mbstrlen(const char *mbstr)
{
	int			len = 0;

	/* optimization for single byte encoding */
	if (pg_database_encoding_max_length() == 1)
		return strlen(mbstr);

	while (*mbstr)
	{
		mbstr += pg_mblen(mbstr);
		len++;
	}
	return len;
}

static char *
percent_encode(unsigned char *s, int srclen)
{
	unsigned char  *end;
	StringInfoData  buf;
	int				len;

	initStringInfo(&buf);

	if (srclen < 0)
		srclen = strlen((char *) s);

	end = s + srclen;

	for (; s < end; s += len)
	{
		unsigned char  *utf;
		int				ulen;

		len = pg_mblen((const char *) s);

		if (len == 1)
		{
			if (('0' <= s[0] && s[0] <= '9') ||
				('A' <= s[0] && s[0] <= 'Z') ||
				('a' <= s[0] && s[0] <= 'z') ||
				(s[0] == '-') || (s[0] == '.') ||
				(s[0] == '_') || (s[0] == '~'))
			{
				appendStringInfoChar(&buf, s[0]);
				continue;
			}
		}

		utf = pg_do_encoding_conversion(s, len, GetDatabaseEncoding(), PG_UTF8);
		ulen = pg_encoding_mblen(PG_UTF8, (const char *) utf);
		while(ulen--)
		{
			appendStringInfo(&buf, "%%%2X", *utf);
			utf++;
		}
	}

	return buf.data;
}

/* returns the length (counted in wchars) of a multibyte string
 * (not necessarily NULL terminated)
 */
int
pg_mbstrlen_with_len(const char *mbstr, int limit)
{
	int			len = 0;

	/* optimization for single byte encoding */
	if (pg_database_encoding_max_length() == 1)
		return limit;

	while (limit > 0 && *mbstr)
	{
		int			l = pg_mblen(mbstr);

		limit -= l;
		mbstr += l;
		len++;
	}
	return len;
}