/* 
 * Copyright (C) 2005-2006 Evgeniy <dushistov@mail.ru>
 * Copyright 2011 kubtek <kubtek@mail.com>
 *
 * This file is part of StarDict.
 *
 * StarDict is free software: you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation, either version 3 of the License, or
 * (at your option) any later version.
 *
 * StarDict is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.
 *
 * You should have received a copy of the GNU General Public License
 * along with StarDict.  If not, see <http://www.gnu.org/licenses/>.
 */

#ifdef HAVE_CONFIG_H
#  include "config.h"
#endif

#include <glib.h>
#include <glib/gi18n.h>
#include <glib/gstdio.h>
#include <gtk/gtk.h>
#include <cstring>
#include <cstdlib>
#include <fcntl.h>
#include <cerrno>

#ifdef CONFIG_GNOME
#  include <libgnome/libgnome.h>
#elif defined(_WIN32)
#  include <gdk/gdkwin32.h>
#  include <Shlwapi.h>
#  include <io.h>
#  include <ERRNO.H>
#endif

#ifndef O_BINARY
#define O_BINARY 0
#endif

#include <list>

#include "libcommon.h"
#include "utils.h"


void fontname_to_family_and_size(const gchar *fontname, std::string &family, gint &size)
{
	const gchar *a = strrchr(fontname, ' ');
	if (a) {
		family.assign(fontname, a - fontname);
		a++;
		size = atoi(a);
	}
}

/*
button, entry {
  color: #ff00ea;
  font: 12px "Comic Sans";
}

font_name:
Comic Sans 12

px_font:
12px "Comic Sans"

*/

#if GTK_MAJOR_VERSION >= 3
void font_name_to_px_font(const char *font_name, std::string &px_font)
{
	const char *ab;
	ab = strrchr(font_name, ' ');
	if (ab) {
		std::string str2(font_name, ab-font_name);
		ab++;
		px_font = ab;
		px_font += "px \"";
		px_font += str2;
		px_font += '\"';
	}
}
#endif

void ProcessGtkEvent()
{
#if GTK_MAJOR_VERSION >= 4
	g_main_context_iteration(g_main_context_default(), FALSE);
#else
	while (gtk_events_pending())
		gtk_main_iteration();
#endif
}

std::string combnum2str(gint comb_code)
{
  switch (comb_code) {
#ifdef _WIN32
  case 0:
    return "Shift";
  case 1:
    return "Alt";
  case 2:
    return "Ctrl";
  case 3:
    return "Ctrl+Alt";
#else
  case 0:
    return "Win";
  case 1:
    return "Shift";
  case 2:
    return "Alt";
  case 3:
    return "Ctrl";
  case 4:
    return "Ctrl+Alt";
  case 5:
    return "Ctrl+e";
  case 6:
    return "Win+d";
  case 7:
    return "F1";
  case 8:
    return "F2";
  case 9:
    return "F3";
  case 10:
    return "F4";
#endif
  default:
    return "";
  }
}

std::vector<std::string> split(const std::string& str, char sep)
{
	std::vector<std::string> res;
	std::string::size_type prev_pos=0, pos = 0;
	while ((pos=str.find(sep, prev_pos))!=std::string::npos) {
		res.push_back(std::string(str, prev_pos, pos-prev_pos));
		prev_pos=pos+1;
	}
	res.push_back(std::string(str, prev_pos, str.length()-prev_pos));

	return res;
}

GdkPixbuf *load_image_from_file(const std::string& filename)
{
	GError *err=NULL;
	GdkPixbuf *res=gdk_pixbuf_new_from_file(filename.c_str(), &err);
	if (!res) {
		g_error(_("Can not load image. %s"), err->message);
		g_error_free(err);
	}

	return res;
}

static gchar * byte_to_hex(unsigned char nr) {
	gchar *result = NULL;

	result = g_strdup_printf("%%%x%x", nr / 0x10, nr % 0x10);
	return result;
}

char *common_build_dictdata(char type, const char *definition)
{
	size_t len = strlen(definition);
	guint32 size;
	size = sizeof(char) + len + 1;
	char *data = (char *)g_malloc(sizeof(guint32) + size);
	char *p = data;
	*((guint32 *)p)= size;
	p += sizeof(guint32);
	*p = type;
	p++;
	memcpy(p, definition, len+1);
	return data;
}

char *common_encode_uri_string(const char *string)
{
	gchar		*newURIString;
	gchar		*hex, *tmp = NULL;
	int		i, j, len, bytes;

	/* the UTF-8 string is casted to ASCII to treat
	   the characters bytewise and convert non-ASCII
	   compatible chars to URI hexcodes */
	newURIString = g_strdup("");
	len = strlen(string);
	for(i = 0; i < len; i++) {
		if(g_ascii_isalnum(string[i]) || strchr("-_.!~*'()", (int)string[i]))
		   	tmp = g_strdup_printf("%s%c", newURIString, string[i]);
		else if(string[i] == ' ')
			tmp = g_strdup_printf("%s%%20", newURIString);
		else if((unsigned char)string[i] <= 127) {
			tmp = g_strdup_printf("%s%s", newURIString, hex = byte_to_hex(string[i]));g_free(hex);
		} else {
			bytes = 0;
			if(((unsigned char)string[i] >= 192) && ((unsigned char)string[i] <= 223))
				bytes = 2;
			else if(((unsigned char)string[i] > 223) && ((unsigned char)string[i] <= 239))
				bytes = 3;
			else if(((unsigned char)string[i] > 239) && ((unsigned char)string[i] <= 247))
				bytes = 4;
			else if(((unsigned char)string[i] > 247) && ((unsigned char)string[i] <= 251))
				bytes = 5;
			else if(((unsigned char)string[i] > 247) && ((unsigned char)string[i] <= 251))
				bytes = 6;
				
			if(0 != bytes) {
				if((i + (bytes - 1)) > len) {
					g_warning(("Unexpected end of character sequence or corrupt UTF-8 encoding! Some characters were dropped!"));
					break;
				}

				for(j=0; j < (bytes - 1); j++) {
					tmp = g_strdup_printf("%s%s", newURIString, hex = byte_to_hex((unsigned char)string[i++]));
					g_free(hex);
					g_free(newURIString);
					newURIString = tmp;
				}
				tmp = g_strdup_printf("%s%s", newURIString, hex = byte_to_hex((unsigned char)string[i]));
				g_free(hex);
			} else {
				/* sh..! */
				g_error("Internal error while converting UTF-8 chars to HTTP URI!");
			}
		}
		g_free(newURIString); 
		newURIString = tmp;
	}
	return newURIString;
}

void common_terminal2pango(const char *t, std::string &pango)
{
	pango.clear();
	const char *p1;
	std::string color_num;
	std::string color;
	std::string tmp_str;
	while (*t) {
		if ((*t == '_') && (t[1] == '')) {
			color = "#FFFA00"; // should be #FFFFFF in fact.
			pango += "<span foreground=\"";
			pango += color;
			pango += "\">";
			t+=2;
			pango += *t;
			pango += "</span>";
			if (*t) {
				t++;
				continue;
			} else {
				break;
			}
		}
		if ((*t == '') && (t[1] == '[')) {
			p1 = strchr(t+2, 'm');
			if (p1) {
				color_num.assign(t+2, p1-t-2);
				if (color_num == "30") {
					color = "#000000";
				} else if (color_num == "31") {
					color = "#FF0000";
				} else if (color_num == "32") {
					color = "#008000";
				} else if (color_num == "33") {
					color = "#FFFF00";
				} else if (color_num == "34") {
					color = "#0000FF";
				} else if (color_num == "35") {
					color = "#FF00FF";
				} else if (color_num == "36") {
					color = "#00FFFF";
				} else if (color_num == "37") {
					color = "#FFFFFF";
				} else {
					color = "#EEEEEE"; // No reason.
				}
				t = p1 +1 ;
				pango += "<span foreground=\"";
				pango += color;
				pango += "\">";
				p1 = strstr(t, "");
				if (p1) {
					tmp_str.assign(t, p1-t);
					pango += tmp_str;
					t = p1 +3;
					pango += "</span>";
				} else {
					pango += t;
					pango += "</span>";
					break;
				}
			} else {
				break;
			}
		} else {
			pango += *t;
			t++;
		}
	}
}

void common_terminal2pango_2(const char *t, std::string &pango)
{
	pango.clear();
	std::string prev_str;
	std::string prev_char;
	const char *p1;
	while (*t) {
		if (*t == 8) {
			if (g_str_has_prefix(t+1, prev_char.c_str())) {
				t++;
				pango += "<b>";
				pango += prev_str;
				pango += "</b>";
			} else if (prev_char == "_") {
				t++;
				p1 = g_utf8_next_char(t);
				char *mark = g_markup_escape_text(t, p1 -t);
				pango += "<u>";
				pango += mark;
				pango += "</u>";
				g_free(mark);
			}
			prev_str.clear();
		} else {
			pango += prev_str;
			switch (*t) {
				case '&':
					prev_str = "&amp;";
					break;
				case '<':
					prev_str = "&lt;";
					break;
				case '>':
					prev_str = "&gt;";
					break;
				case '\'':
					prev_str = "&apos;";
					break;
				case '"':
					prev_str = "&quot;";
					break;
				default:
					p1 = g_utf8_next_char(t);
					prev_str.assign(t, p1-t);
					break;
			}
		}
		p1 = g_utf8_next_char(t);
		prev_char.assign(t, p1-t);
		t = p1;
	}
	pango += prev_str;
}

/* based on g_mkstemp_full 
#if defined(_WIN32)
	flags is type of operation allowed parameter of _wsopen_s function
	mode is permission settings parameter of _wsopen_s function
#else
	for the meaning of flags and mode parameters see g_mkstemp_full function
#endif
*/
gint
stardict_mkstemp_full (gchar *tmpl,
	int    flags,
	int    mode)
{
	char *XXXXXX;
	int count, fd;
	static const char letters[] =
		"ABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789";
	static const int NLETTERS = sizeof (letters) - 1;
	glong value;
	//GTimeVal tv;
	gint64 tv_usec;
	gint64 tv_sec;
	static int counter = 0;

	g_return_val_if_fail (tmpl != NULL, -1);


	/* find the last occurrence of "XXXXXX" */
	XXXXXX = g_strrstr (tmpl, "XXXXXX");

	if (!XXXXXX || strncmp (XXXXXX, "XXXXXX", 6))
	{
		errno = EINVAL;
		return -1;
	}
#if defined(_WIN32)
	std::string tmpl_utf8;
	if(!file_name_to_utf8(tmpl, tmpl_utf8))
		return -1;
	std_win_string tmpl_win;
	if(!utf8_to_windows(tmpl_utf8, tmpl_win))
		return -1;
	size_t XXXXXXind = tmpl_win.rfind(TEXT("XXXXXX"));
	if(XXXXXXind == std_win_string::npos)
		return -1;
#endif

	/* Get some more or less random data.  */
	//g_get_current_time (&tv);
	//value = (tv.tv_usec ^ tv.tv_sec) + counter++;
	tv_usec = g_get_real_time();
	tv_sec = tv_usec / G_USEC_PER_SEC;
	value = (tv_usec ^ tv_sec) + counter++;

	for (count = 0; count < 100; value += 7777, ++count)
	{
		glong v = value;

		/* Fill in the random bits.  */
		XXXXXX[0] = letters[v % NLETTERS];
		v /= NLETTERS;
		XXXXXX[1] = letters[v % NLETTERS];
		v /= NLETTERS;
		XXXXXX[2] = letters[v % NLETTERS];
		v /= NLETTERS;
		XXXXXX[3] = letters[v % NLETTERS];
		v /= NLETTERS;
		XXXXXX[4] = letters[v % NLETTERS];
		v /= NLETTERS;
		XXXXXX[5] = letters[v % NLETTERS];

#if defined(_WIN32)
		tmpl_win[XXXXXXind + 0] = (TCHAR)XXXXXX[0];
		tmpl_win[XXXXXXind + 1] = (TCHAR)XXXXXX[1];
		tmpl_win[XXXXXXind + 2] = (TCHAR)XXXXXX[2];
		tmpl_win[XXXXXXind + 3] = (TCHAR)XXXXXX[3];
		tmpl_win[XXXXXXind + 4] = (TCHAR)XXXXXX[4];
		tmpl_win[XXXXXXind + 5] = (TCHAR)XXXXXX[5];
		errno_t err = _wsopen_s(&fd, tmpl_win.c_str(),
			flags | _O_CREAT | _O_EXCL,
			_SH_DENYWR, mode);
		if(err == EEXIST || err == EACCES)
			continue;
		if(err != 0)
			return -1;
		if(fd >= 0)
			return fd;
#else
		/* tmpl is in UTF-8 on Windows, thus use g_open() */
		fd = g_open (tmpl, flags | O_CREAT | O_EXCL, mode);

		if (fd >= 0)
			return fd;
		else if (errno != EEXIST)
			/* Any other error will apply also to other names we might
			*  try, and there are 2^32 or so of them, so give up now.
			*/
			return -1;
#endif
	}

	/* We got out of the loop because we ran out of combinations to try.  */
	errno = EEXIST;
	return -1;
}

/* based on g_mkstemp */
gint
stardict_mkstemp (gchar *tmpl)
{
#if defined(_WIN32)
	return stardict_mkstemp_full (tmpl, _O_RDWR | _O_BINARY, _S_IREAD | _S_IWRITE);
#else
	return stardict_mkstemp_full (tmpl, O_RDWR | O_BINARY, 0600);
#endif
}

/* based on g_file_open_tmp */
gint
stardict_file_open_tmp (const gchar  *tmpl,
				 gchar       **name_used,
				 GError      **error)
{
	int retval;
	const char *tmpdir;
	const char *sep;
	char *fulltemplate;
	const char *slash;

	if (tmpl == NULL)
		tmpl = ".XXXXXX";

	if ((slash = strchr (tmpl, G_DIR_SEPARATOR)) != NULL
#ifdef G_OS_WIN32
		|| (strchr (tmpl, '/') != NULL && (slash = "/"))
#endif
		)
	{
		gchar *display_tmpl = g_filename_display_name (tmpl);
		char c[2];
		c[0] = *slash;
		c[1] = '\0';

		g_set_error (error,
			G_FILE_ERROR,
			G_FILE_ERROR_FAILED,
			_("Template '%s' invalid, should not contain a '%s'"),
			display_tmpl, c);
		g_free (display_tmpl);

		return -1;
	}

	if (strstr (tmpl, "XXXXXX") == NULL)
	{
		gchar *display_tmpl = g_filename_display_name (tmpl);
		g_set_error (error,
			G_FILE_ERROR,
			G_FILE_ERROR_FAILED,
			_("Template '%s' doesn't contain XXXXXX"),
			display_tmpl);
		g_free (display_tmpl);
		return -1;
	}

	tmpdir = g_get_tmp_dir ();

	if (G_IS_DIR_SEPARATOR (tmpdir [strlen (tmpdir) - 1]))
		sep = "";
	else
		sep = G_DIR_SEPARATOR_S;

	fulltemplate = g_strconcat (tmpdir, sep, tmpl, NULL);

	retval = stardict_mkstemp (fulltemplate);

	if (retval == -1)
	{
		int save_errno = errno;
		gchar *display_fulltemplate = g_filename_display_name (fulltemplate);

		g_set_error (error,
			G_FILE_ERROR,
			g_file_error_from_errno (save_errno),
			_("Failed to create file '%s': %s"),
			display_fulltemplate, g_strerror (save_errno));
		g_free (display_fulltemplate);
		g_free (fulltemplate);
		return -1;
	}

	if (name_used)
		*name_used = fulltemplate;
	else
		g_free (fulltemplate);

	return retval;
}

query_t analyse_query(const char *s, std::string& res)
{
	if (!s || !*s) {
		res="";
		return qtSIMPLE;
	}
	if (*s=='/') {
		res=s+1;
		return qtFUZZY;
	}
	if (*s==':') {
		res=s+1;
		return qtREGEX;
	}

	if (*s=='|') {
		res=s+1;
		return qtFULLTEXT;
	}

	bool pattern=false;
	const char *p=s;
	res="";
	for (; *p; res+=*p, ++p) {
		if (*p=='\\') {
			++p;
			if (!*p)
				break;
			continue;
		}
		if (*p=='*' || *p=='?')
			pattern=true;
	}
	if (pattern)
		return qtPATTERN;

	return qtSIMPLE;
}

void stardict_input_escape(const char *text, std::string &res)
{
	res.clear();
	const char *p = text;
	if (*p == '/' || *p == '|' || *p == ':') {
		res = "\\";
		res += *p;
		p++;
	}
	while (*p) {
		if (*p == '\\' || *p == '*' || *p == '?') {
			res += '\\';
			res += *p;
		} else {
			res += *p;
		}
		p++;
	}
}

static const char* html_entrs[] =     { "lt;", "gt;", "amp;", "apos;", "quot;", 0 };
static const int html_entrs_len[] =   { 3,     3,     4,      5,       5,       0 };
static const char html_raw_entrs[] =  { '<',   '>',   '&',    '\'',    '\"',    0 };

static const char* html_tags[] =     { "br>", 0 };
static const int html_tags_len[] =   { 3,     0 };
static const char html_raw_tags[] =  { '\n',  0 };

void html_decode(const char *str, std::string& decoded)
{
	decoded.clear();
	decoded.reserve(strlen(str));
	int ind;

	const char* p = str;
	while (*p)
		if (*p == '&') {
			if(*(p+1) == '#') {
				const char *q = strchr(p+2, ';');
				if(q) {
					long code = atol(p+2);
					char buf[10]; // must be at least 6 bytes long
					buf[g_unichar_to_utf8(gunichar(code), buf)] = '\0';
					decoded += buf;
					p = q + 1;
				} else {
					g_debug("unknown entry %s", p);
					break;
				}
			} else {
				for(ind = 0; html_entrs[ind] != 0; ++ind)
					if(strncmp(p + 1, html_entrs[ind], html_entrs_len[ind]) == 0) {
						decoded += html_raw_entrs[ind];
						p += html_entrs_len[ind]+1;
						break;
					}
				if (html_entrs[ind] == 0) { // unrecognized sequence
					const char *q = strchr(p+1, ';');
					if(q) {
						++q;
						g_debug("unknown entry %s", std::string(p, q-p).c_str());
						p = q;
					} else {
						g_debug("unknown entry %s", p);
						break;
					}
				}
			}
		} else if(*p == '<') {
			for(ind = 0; html_tags[ind] != 0; ++ind)
				if(strncmp(p + 1, html_tags[ind], html_tags_len[ind]) == 0) {
					decoded += html_raw_tags[ind];
					p += html_tags_len[ind]+1;
					break;
				}
			if (html_tags[ind] == 0) { // unrecognized sequence
				const char *q = strchr(p+1, '>');
				if(q) {
					++q;
					g_debug("unknown tag %s", std::string(p, q-p).c_str());
					p = q;
				} else {
					g_debug("unknown tag %s", p);
					break;
				}
			}
		} else {
			decoded += *p++;
		}
}

/* extract first pure english word (or a sequence of words separated by spaces)
	src in utf-8,
	dst must point to a buffer of size enough to hold the src string */
void GetPureEnglishAlpha(char *dst, const char *src)
{
	while(*src && (uchar(*src) >= 128 || !isalpha(*src)))
		src++;
	while(*src && uchar(*src) < 128 && (isalpha(*src) || *src == ' '))
		*dst++ = *src++;
	*dst = '\0';
}

bool IsASCII(const char *str)
{
	// works with UTF-8 strings (bytes of a multi-byte character has the highest 
	// bit set)
	for(; *str; ++str)
		if(uchar(*str) >= 128)
			return false;
	return true;
}

/* returns pointer to the first non-space unichar in the string */
const char* skip_spaces(const char *str)
{
	while(*str && g_unichar_isspace(g_utf8_get_char(str)))
		str = g_utf8_next_char(str);
	return str;
}

/* copy src to dst converting adjacent unicode spaces into one ASCII space ' ' 
	return value - new '\0' character */
char* copy_normalize_spaces(char *dst, const char *src)
{
	while(*src) {
		if (g_unichar_isspace(g_utf8_get_char(src))) {
			*dst++=' ';
			src = g_utf8_next_char(src);
			while(g_unichar_isspace(g_utf8_get_char(src)))
				src = g_utf8_next_char(src);
		} else {
			g_utf8_strncpy(dst,src,1);
			src = g_utf8_next_char(src);
			dst = g_utf8_next_char(dst);
		}
	}
	*dst='\0';
	return dst;
}

/* copy src to dst converting adjacent unicode spaces into one ASCII space ' ',
	remove leading and trailing unicode spaces */
void copy_normalize_trim_spaces(char *dst, const char *src)
{
	src = skip_spaces(src);
	char* end = copy_normalize_spaces(dst, src);
	delete_trailing_spaces_ASCII(dst, end);
}

/* delete trailing ASCII spaces. 
begin - beginning of the string, end - '\0' character 
return value - new '\0' character */
char* delete_trailing_spaces_ASCII(const char *begin, char *end)
{
	while(begin < end && *(end - 1) == ' ')
		--end;
	*end = '\0';
	return end;
}

/* delete the last word separated by ASCII space
In other word we replace the last ASCII space with '\0'.
begin - beginning of the string, end - '\0' character 
return value - new '\0' character */
char* delete_trailing_word_ASCII(const char *begin, char *end)
{
	while(begin < end && *(end-1) != ' ')
		--end;
	if(begin < end)
		--end;
	*end = '\0';
	return end;
}

/* return value - new '\0' character */
char* delete_trailing_char(char *begin, char *end)
{
	char* p = g_utf8_find_prev_char(begin, end);
	if(p) {
		*p = '\0';
		return p;
	} else {
		*begin = '\0';
		return begin;
	}
}

/* like extract_word but does not copy the extracted word
instead assignes begin and end to the first and next to last chars in the source string */
void extract_word_in_place(const char **begin, const char **end, const char* src, 
	int BeginPos, gboolean (*is_splitter)(gunichar c))
{
	g_assert(begin);
	g_assert(end);
	g_assert(BeginPos >= 0);
	*begin = *end = NULL;
	const char* const pointer = src + BeginPos;
	const char *word_begin = NULL;
	while(true) {
		while(*src && is_splitter(g_utf8_get_char(src)))
			src = g_utf8_next_char(src);
		if(!*src)
			break;
		if(src <= pointer || !word_begin)
			word_begin = src;
		if(pointer <= src)
			break;
		while(*src && !is_splitter(g_utf8_get_char(src)))
			src = g_utf8_next_char(src);
		if(!*src)
			break;
	}
	if(!word_begin)
		return;
	src = word_begin;
	while(*src && !is_splitter(g_utf8_get_char(src))) {
		src = g_utf8_next_char(src);
	}
	*begin = word_begin;
	*end = src;
}

/* extract the word at position BeginPos
word separator chars are identified by is_splitter function
If BeginPos points to a splitter, get the word to the left of the point.
If there is no word to the left, get the word to the right of the point.
copy result into dst */
void extract_word(char *dst, const char* src, int BeginPos, gboolean (*is_splitter)(gunichar c))
{
	g_assert(BeginPos >= 0);
	const char *begin, *end;
	extract_word_in_place(&begin, &end, src, BeginPos, is_splitter);
	if(!begin || !end) {
		*dst = '\0';
		return;
	}
	const int num = end - begin;
	strncpy(dst, begin, num);
	dst[num] = '\0';
}

/* Extract a capitalized word
	the word must look like Word 
	where the first letter satisfies the is_first_letter function,
	the first letter must be followed by at least one letter satisfing the is_second_letter function */
void extract_capitalized_word_in_place(const char **begin, const char **end,
	const char* src, int BeginPos, 
	gboolean (*is_first_letter)(gunichar c), gboolean (*is_second_letter)(gunichar c))
{
	g_assert(begin);
	g_assert(end);
	g_assert(BeginPos >= 0);
	*begin = *end = NULL;
	const char* const pointer = src + BeginPos;
	const char *word_begin = NULL;
	while(true) {
		while(*src && !is_first_letter(g_utf8_get_char(src)))
			src = g_utf8_next_char(src);
		if(!*src)
			break;
		const char *cur_word_begin = src;
		src = g_utf8_next_char(src);
		if(!is_second_letter(g_utf8_get_char(src)))
			continue;
		if(cur_word_begin <= pointer || !word_begin)
			word_begin = cur_word_begin;
		if(pointer <= cur_word_begin)
			break;
		while(*src && is_second_letter(g_utf8_get_char(src)))
			src = g_utf8_next_char(src);
		if(!*src)
			break;
	}
	if(!word_begin)
		return;
	// skip the first letter
	src = g_utf8_next_char(word_begin);
	while(*src && is_second_letter(g_utf8_get_char(src))) {
		src = g_utf8_next_char(src);
	}
	*begin = word_begin;
	*end = src;
}

void extract_capitalized_word(char *dst, const char* src, int BeginPos, 
	gboolean (*is_first_letter)(gunichar c), gboolean (*is_second_letter)(gunichar c))
{
	g_assert(BeginPos >= 0);
	const char *begin, *end;
	extract_capitalized_word_in_place(&begin, &end, src, BeginPos, is_first_letter, is_second_letter);
	if(!begin || !end) {
		*dst = '\0';
		return;
	}
	const int num = end - begin;
	strncpy(dst, begin, num);
	dst[num] = '\0';
}

const char* find_first(const char* src, gboolean (*isfunc)(gunichar c))
{
	while(*src && !isfunc(g_utf8_get_char(src)))
		src = g_utf8_next_char(src);
	if(isfunc(g_utf8_get_char(src)))
		return src;
	else
		return NULL;
}

const char* find_first_not(const char* src, gboolean (*isfunc)(gunichar c))
{
	while(*src && isfunc(g_utf8_get_char(src)))
		src = g_utf8_next_char(src);
	if(isfunc(g_utf8_get_char(src)))
		return NULL;
	else
		return src;
}

gboolean is_space_or_punct(gunichar c)
{
	return g_unichar_isspace(c) || g_unichar_ispunct(c);
}

gboolean is_not_alpha(gunichar c)
{
	return !g_unichar_isalpha(c);
}

gboolean is_not_upper(gunichar c)
{
	return !g_unichar_isupper(c);
}

gboolean is_not_lower(gunichar c)
{
	return !g_unichar_islower(c);
}