fs2open.github.com/code/localization/localize.cpp

/*
 * Copyright (C) Volition, Inc. 1999.  All rights reserved.
 *
 * All source code herein is the property of Volition, Inc. You may not sell 
 * or otherwise commercially exploit the source or things you created based on the 
 * source.
 *
*/




#include <ctype.h>
#include "localization/localize.h"
#include "parse/parselo.h"
#include "osapi/osregistry.h"
#include "parse/encrypt.h"
#include "playerman/player.h"
#include "cfile/cfile.h"




// ------------------------------------------------------------------------------------------------------------
// LOCALIZE DEFINES/VARS
//

// general language/localization data ---------------------

// current language
int Lcl_current_lang = LCL_ENGLISH;

// language info table
typedef struct lang_info {
	char lang_name[LCL_LANG_NAME_LEN + 1];				// literal name of the language
	char lang_ext[LCL_LANG_NAME_LEN + 1];				// for adding to names on disk access
} lang_info;

lang_info Lcl_languages[LCL_NUM_LANGUAGES] = {
	{ "English",		"" },			// english
	{ "German",			"gr" },			// german
	{ "French",			"fr" },			// french
	{ "Polish",			"pl" },			// polish
};

#define DEFAULT_LANGUAGE							"English"

// following is the offset where special characters start in our font
#define LCL_SPECIAL_CHARS_FR	164
#define LCL_SPECIAL_CHARS_GR	164
#define LCL_SPECIAL_CHARS_PL	127
#define LCL_SPECIAL_CHARS		127
int Lcl_special_chars;

// use these to replace *_BUILD, values before
// only 1 will be active at a time
int Lcl_fr = 0;
int Lcl_gr = 0;
int Lcl_pl = 0;
int Lcl_english = 1;


// executable string localization data --------------------

// XSTR_SIZE is the total count of unique XSTR index numbers.  An index is used to
// reference an entry in strings.tbl.  This is used for translation of strings from
// the english version (in the code) to a foreign version (in the table).  Thus, if you
// add a new string to the code, you must assign it a new index.  Use the number below for
// that index and increase the number below by one.
#define XSTR_SIZE	1572


// struct to allow for strings.tbl-determined x offset
// offset is 0 for english, by default
typedef struct {
	char *str;
	int  offset_x;				// string offset in 640
	int  offset_x_hi;			// string offset in 1024
} lcl_xstr;

//char *Xstr_table[XSTR_SIZE];
lcl_xstr Xstr_table[XSTR_SIZE];
int Xstr_inited = 0;


// table/mission externalization stuff --------------------

#define TABLE_STRING_FILENAME						"tstrings.tbl"
// filename of the file to use when localizing table strings
char *Lcl_ext_filename = NULL;
CFILE *Lcl_ext_file = NULL;

// for scanning/parsing tstrings.tbl (from ExStr)
#define PARSE_TEXT_STRING_LEN			PARSE_BUF_SIZE
#define PARSE_ID_STRING_LEN			128
#define TS_SCANNING						0				// scanning for a line of text
#define TS_ID_STRING						1				// reading in an id string
#define TS_OPEN_QUOTE					2				// looking for an open quote
#define TS_STRING							3				// reading in the text string itself
int Ts_current_state = 0;
char Ts_text[PARSE_TEXT_STRING_LEN];				// string we're currently working with
char Ts_id_text[PARSE_ID_STRING_LEN];				// id string we're currently working with
int Ts_text_size;
int Ts_id_text_size;

// file pointers for optimized string lookups
// some example times for FreeSpace2 startup with granularities (mostly .tbl files, ~500 strings in the table file, many looked up more than once)
// granularity 20			:		13 secs
// granularity 10			:		11 secs
// granularity 5			:		9 secs
// granularity 2			:		7-8 secs
#define LCL_GRANULARITY					1				// how many strings between each pointer (lower granularities should give faster lookup times)
#define LCL_MAX_POINTERS				4500			// max # of pointers
#define LCL_MAX_STRINGS					(LCL_GRANULARITY * LCL_MAX_POINTERS)
int Lcl_pointers[LCL_MAX_POINTERS];
int Lcl_pointer_count = 0;


// ------------------------------------------------------------------------------------------------------------
// LOCALIZE FORWARD DECLARATIONS
//

// associate table file externalization with the specified input file
void lcl_ext_associate(char *filename);

// given a valid XSTR() tag piece of text, extract the string portion, return it in out, nonzero on success
int lcl_ext_get_text(char *xstr, char *out);

// given a valid XSTR() tag piece of text, extract the id# portion, return the value in out, nonzero on success
int lcl_ext_get_id(char *xstr, int *out);

// given a valid XSTR() id#, lookup the string in tstrings.tbl, filling in out if found, nonzero on success
int lcl_ext_lookup(char *out, int id);

// if the char is a valid char for a signed integer value string
int lcl_is_valid_numeric_char(char c);

// sub-parse function for individual lines of tstrings.tbl (from Exstr)
// returns : integer with the low bits having the following values :
// 0 on fail, 1 on success, 2 if found a matching id/string pair, 3 if end of language has been found
// for cases 1 and 2 : the high bit (1<<31) will be set if the parser detected the beginning of a new string id on this line
// so be sure to mask this value out to get the low portion of the return value
//
int lcl_ext_lookup_sub(char *text, char *out, int id);

// initialize the pointer array into tstrings.tbl (call from lcl_ext_open() ONLY)
void lcl_ext_setup_pointers();


// ------------------------------------------------------------------------------------------------------------
// LOCALIZE FUNCTIONS
//

// initialize localization, if no language is passed - use the language specified in the registry
void lcl_init(int lang_init)
{
	atexit(lcl_xstr_close);

	char lang_string[128];
	char *ret;
	int lang, idx;

	// initialize encryption
	encrypt_init();

	// read the language from the registry
	if(lang_init < 0){
		memset(lang_string, 0, 128);
		// default to DEFAULT_LANGUAGE (which should be English so we don't have to put German text 
		// in tstrings in the #default section)
		ret = os_config_read_string(NULL, "Language", DEFAULT_LANGUAGE);

		if(ret == NULL){
			Int3();
			strcpy_s(lang_string, DEFAULT_LANGUAGE);
		} else {
			strcpy_s(lang_string, ret);
		}

		// look it up
		lang = -1;
		for(idx=0; idx<LCL_NUM_LANGUAGES; idx++){
			if(!stricmp(Lcl_languages[idx].lang_name, lang_string)){
				lang = idx;
				break;
			}
		}
		if(lang < 0){
			lang = 0;
		}	
	} else {
		Assert((lang_init >= 0) && (lang_init < LCL_NUM_LANGUAGES));
		lang = lang_init;
	}

	// language markers
	Lcl_pointer_count = 0;

	// associate the table string file
	lcl_ext_associate(TABLE_STRING_FILENAME);		

	// set the language (this function takes care of setting up file pointers)
	lcl_set_language(lang);		
}

// added 2.2.99 by NeilK to take care of fs2 launcher memory leaks
// shutdown localization
void lcl_close()
{
	// if the filename exists, free it up
	if(Lcl_ext_filename != NULL){
		vm_free(Lcl_ext_filename);
	}
}

// determine what language we're running in, see LCL_* defines above
int lcl_get_language()
{
	return Lcl_current_lang;
}

void parse_stringstbl(char *filename)
{
	char chr, buf[4096];
	char language_tag[512];
	int i, z, index;
	char *p_offset = NULL;
	int offset_lo = 0, offset_hi = 0;

	// make sure localization is NOT running
	lcl_ext_close();

	read_file_text(filename, CF_TYPE_TABLES);
	reset_parse();

	// move down to the proper section		
	memset(language_tag, 0, sizeof(language_tag));
	strcpy_s(language_tag, "#");
	strcat_s(language_tag, Lcl_languages[Lcl_current_lang].lang_name);

	if ( skip_to_string(language_tag) != 1 )
		Error(LOCATION, "%s is corrupt", filename);

	// parse all the strings in this section of the table
	while ( !check_for_string("#") ) {
		int num_offsets_on_this_line = 0;

		stuff_int(&index);
		stuff_string(buf, F_NAME, sizeof(buf));

		if (Lcl_pl)
			lcl_fix_polish(buf);

		i = strlen(buf);

		while (i--) {
			if ( !isspace(buf[i]) )
				break;
		}

		// trim unneccesary end of string
		if (i >= 0) {
			// Assert(buf[i] == '"');
			if (buf[i] != '"') {
				// probably an offset on this entry

				// drop down a null terminator (prolly unnecessary)
				buf[i+1] = 0;

				// back up over the potential offset
				while ( !is_white_space(buf[i]) )
					i--;

				// now back up over intervening spaces
				while ( is_white_space(buf[i]) )
					i--;

				num_offsets_on_this_line = 1;

				if (buf[i] != '"') {
					// could have a 2nd offset value (one for 640, one for 1024)
					// so back up again
					while ( !is_white_space(buf[i]) )
						i--;

					// now back up over intervening spaces
					while ( is_white_space(buf[i]) )
						i--;

					num_offsets_on_this_line = 2;
				}

				p_offset = &buf[i+1];			// get ptr to string section with offset in it

				if (buf[i] != '"')
					Error(LOCATION, "%s is corrupt", filename);		// now its an error
			}

			buf[i] = 0;
		}

		// copy string into buf
		z = 0;
		for (i = 1; buf[i]; i++) {
			chr = buf[i];

			if (chr == '\\') {
				chr = buf[++i];

				if (chr == 'n')
					chr = '\n';
				else if (chr == 'r')
					chr = '\r';
			}

			buf[z++] = chr;
		}

		// null terminator on buf
		buf[z] = 0;

		// write into Xstr_table
		if (index >= 0 && index < XSTR_SIZE) {
			if ( Parsing_modular_table && (Xstr_table[index].str != NULL) ) {
				vm_free(Xstr_table[index].str);
				Xstr_table[index].str = NULL;
			}

			if (Xstr_table[index].str != NULL)
				Warning(LOCATION, "Strings table index %d used more than once", index);

			Xstr_table[index].str = vm_strdup(buf);
		}

		// read offset information, assume 0 if nonexistant
		if (p_offset != NULL) {
			if (sscanf(p_offset, "%d%d", &offset_lo, &offset_hi) < num_offsets_on_this_line) {
				// whatever is in the file ain't a proper offset
				Error(LOCATION, "%s is corrupt", filename);
			}
		}

		Xstr_table[index].offset_x = offset_lo;

		if (num_offsets_on_this_line == 2)
			Xstr_table[index].offset_x_hi = offset_hi;
		else
			Xstr_table[index].offset_x_hi = offset_lo;

		// clear out our vars
		p_offset = NULL;
		offset_lo = 0;
		offset_hi = 0;
		num_offsets_on_this_line = 0;
	}
}

// initialize the xstr table
void lcl_xstr_init()
{
	int i, rval;

	for (i = 0; i < XSTR_SIZE; i++)
		Xstr_table[i].str = NULL;

	if ( (rval = setjmp(parse_abort)) != 0 )
		mprintf(("TABLES: Unable to parse '%s'!  Error code = %i.\n", "strings.tbl", rval));
	else
		parse_stringstbl("strings.tbl");

	parse_modular_table(NOX("*-lcl.tbm"), parse_stringstbl);

	Xstr_inited = 1;
}


// free Xstr table
void lcl_xstr_close()
{
	for (int i=0; i<XSTR_SIZE; i++){
		if (Xstr_table[i].str != NULL) {
			vm_free(Xstr_table[i].str);
			Xstr_table[i].str = NULL;
		}
	}
	vm_free(Lcl_ext_filename);
}


// set our current language
void lcl_set_language(int lang)
{
	Lcl_current_lang = lang;

	nprintf(("General", "Setting language to %s\n", Lcl_languages[lang].lang_name));

	// flag the proper language as being active
	Lcl_fr = 0;
	Lcl_gr = 0;
	Lcl_pl = 0;
	Lcl_english = 0;
	switch(lang){
	case LCL_ENGLISH:
		Lcl_english = 1;		
		Lcl_special_chars = LCL_SPECIAL_CHARS;
		break;
	case LCL_FRENCH:
		Lcl_fr = 1;
		Lcl_special_chars = LCL_SPECIAL_CHARS_FR;
		break;
	case LCL_GERMAN:
		Lcl_gr = 1;
		Lcl_special_chars = LCL_SPECIAL_CHARS_GR;
		break;
	case LCL_POLISH:
		Lcl_pl = 1;
		Lcl_special_chars = LCL_SPECIAL_CHARS_PL;
		break;
	}

	// set to 0, so lcl_ext_open() knows to reset file pointers
	Lcl_pointer_count = 0;

	// reset file pointers to the proper language-section
	if(Lcl_current_lang != LCL_DEFAULT_LANGUAGE){
		lcl_ext_setup_pointers();
	}
}

// maybe add on an appropriate subdirectory when opening a localized file
void lcl_add_dir(char *current_path)
{
	char last_char;
	int path_len;

	// if the disk extension is 0 length, don't add enything
	if (strlen(Lcl_languages[Lcl_current_lang].lang_ext) <= 0) {
		return;
	}

	// get the length of the string so far
	path_len = strlen(current_path);
	if (path_len <= 0) {
		return;
	}
	
	// get the current last char
	last_char = current_path[path_len - 1];

	// if the last char is a slash, just copy in the disk extension
	if (last_char == DIR_SEPARATOR_CHAR) {
		strcat(current_path, Lcl_languages[Lcl_current_lang].lang_ext);
		strcat(current_path, DIR_SEPARATOR_STR);
	} 
	// otherwise add a slash, then copy in the disk extension
	else {
		strcat(current_path, DIR_SEPARATOR_STR);
		strcat(current_path, Lcl_languages[Lcl_current_lang].lang_ext);
	}
}

// maybe add localized directory to full path with file name when opening a localized file
int lcl_add_dir_to_path_with_filename(char *current_path, uint path_max)
{
	// if the disk extension is 0 length, don't add enything
	if (strlen(Lcl_languages[Lcl_current_lang].lang_ext) <= 0) {
		return 1;
	}

	int str_size = path_max + 1;

	char *temp = new char[str_size];

	// find position of last slash and copy rest of filename (not counting slash) to temp
	// mark end of current path with '\0', so strcat will work
	char *last_slash = strrchr(current_path, DIR_SEPARATOR_CHAR);
	if (last_slash == NULL) {
		strncpy(temp, current_path, str_size);
		current_path[0] = '\0';
	} else {
		strncpy(temp, last_slash+1, str_size);
		last_slash[1] = '\0';
	}

	// add extension
	strcat_s(current_path, path_max, Lcl_languages[Lcl_current_lang].lang_ext);
	strcat_s(current_path, path_max, DIR_SEPARATOR_STR );

	// copy rest of filename from temp
	strcat_s(current_path, path_max, temp);

	delete [] temp;
	return 1;
}


// externalization of table/mission files ----------------------- 

// open the externalization file for use during parsing (call before parsing a given file)
void lcl_ext_open()
{
	// if the file is already open, do nothing
	Assert(Lcl_ext_file == NULL);	

	// if we're running in the default language, do nothing
	if(Lcl_current_lang == LCL_DEFAULT_LANGUAGE){
		return;
	}

	// otherwise open the file
	Lcl_ext_file = cfopen(Lcl_ext_filename, "rt");
	if(Lcl_ext_file == NULL){
		return;
	}		
}

// close the externalization file (call after parsing a given file)
void lcl_ext_close()
{
	// if the file is not open, do nothing
	if(Lcl_ext_file == NULL){
		return;
	}

	// if we're running in the default language, do nothing
	if(Lcl_current_lang == LCL_DEFAULT_LANGUAGE){
		return;
	}
		
	// otherwise close it
	cfclose(Lcl_ext_file);
	Lcl_ext_file = NULL;
}

void lcl_replace_stuff(char *text, unsigned int max_len)
{
	if (Fred_running)
		return;

	Assert(text);	// Goober5000

	// delegate to SCP_string for the replacements
	SCP_string temp_text = text;
	lcl_replace_stuff(temp_text);

	// fill up the original string
	int len = temp_text.copy(text, max_len);
	text[len] = 0;
}

// Goober5000 - replace stuff in the string, e.g. $callsign with player's callsign
// now will also replace $rank with rank, e.g. "Lieutenant"
// now will also replace $quote with double quotation marks
// now will also replace $semicolon with semicolon mark
void lcl_replace_stuff(SCP_string &text)
{
	if (Fred_running)
		return;

	if (Player != NULL)
	{
		replace_all(text, "$callsign", Player->callsign);
		replace_all(text, "$rank", Ranks[Player->stats.rank].name);
	}
	replace_all(text, "$quote", "\"");
	replace_all(text, "$semicolon", ";");
}

void lcl_fred_replace_stuff(char *text, unsigned int max_len)
{
	if (!Fred_running)
		return;

	Assert(text);	// Goober5000

	// delegate to SCP_string for the replacements
	SCP_string temp_text = text;
	lcl_fred_replace_stuff(temp_text);

	// fill up the original string
	int len = temp_text.copy(text, max_len);
	text[len] = 0;
}

void lcl_fred_replace_stuff(SCP_string &text)
{
	if (!Fred_running)
		return;

	replace_all(text, "\"", "$quote");
	replace_all(text, ";", "$semicolon");
}

// get the localized version of the string. if none exists, return the original string
// valid input to this function includes :
// "this is some text"
// XSTR("wheeee", -1)
// XSTR("whee", 20)
// and these should cover all the externalized string cases
// fills in id if non-NULL. a value of -2 indicates it is not an external string
void lcl_ext_localize_sub(char *in, char *out, int max_len, int *id)
{			
	char first_four[5];
	char text_str[PARSE_BUF_SIZE]="";
	char lookup_str[PARSE_BUF_SIZE]="";
	int str_id;	
	int str_len;	

	Assert(in);
	Assert(out);

	// default (non-external string) value
	if(id != NULL){
		*id = -2;
	}	

	str_len = strlen(in);

	// if the string is < 9 chars, it can't be an XSTR("",) tag, so just copy it
	if(str_len < 9){
		if (str_len > max_len)
			error_display(0, "Token too long: [%s].  Length = %i.  Max is %i.\n", in, str_len, max_len);

		strncpy(out, in, max_len);

		if (id != NULL)
			*id = -2;

		return;
	}

	// otherwise, check to see if it's an XSTR() tag
	memset(first_four, 0, 5);
	strncpy(first_four, in, 4);
	if(stricmp(first_four, "XSTR")){
		// NOT an XSTR() tag
		if (str_len > max_len)
			error_display(0, "Token too long: [%s].  Length = %i.  Max is %i.\n", in, str_len, max_len);

		strncpy(out, in, max_len);

		if (id != NULL)
			*id = -2;

		return;
	}

	// at this point we _know_ its an XSTR() tag, so split off the strings and id sections		
	if(!lcl_ext_get_text(in, text_str)){
		Int3();
		strncpy(out, in, max_len);
		if(id != NULL){
			*id = -1;
		}
		return;
	}
	if(!lcl_ext_get_id(in, &str_id)){
		if ( strlen(in) > (uint)max_len )
			error_display(0, "Token too long: [%s].  Length = %i.  Max is %i.\n", in, strlen(in), max_len);

		strncpy(out, in, max_len);

		if (id != NULL)
			*id = -1;

		return;
	}
	
	// if the localization file is not open, or we're running in the default language, return the original string
	if ( (Lcl_ext_file == NULL) || (str_id < 0) || (Lcl_current_lang == LCL_DEFAULT_LANGUAGE) ) {
		if ( strlen(text_str) > (uint)max_len )
			error_display(0, "Token too long: [%s].  Length = %i.  Max is %i.\n", text_str, strlen(text_str), max_len);

		strncpy(out, text_str, max_len);

		if (id != NULL)
			*id = str_id;

		return;
	}		

	// attempt to find the string
	if(lcl_ext_lookup(lookup_str, str_id)){
		// copy to the outgoing string
		if ( strlen(lookup_str) > (uint)max_len )
			error_display(0, "Token too long: [%s].  Length = %i.  Max is %i.\n", lookup_str, strlen(lookup_str), max_len);

		strncpy(out, lookup_str, max_len);
	}
	// otherwise use what we have - probably should Int3() or assert here
	else {
		if ( strlen(text_str) > (uint)max_len )
			error_display(0, "Token too long: [%s].  Length = %i.  Max is %i.\n", text_str, strlen(text_str), max_len);

		strncpy(out, text_str, max_len);
	}	

	// set the id #
	if(id != NULL){
		*id = str_id;
	}
}

// Goober5000 - wrapper for lcl_ext_localize_sub; used because lcl_replace_stuff has to
// be called *after* the translation is done, and the original function returned in so
// many places that it would be messy to call lcl_replace_stuff everywhere
void lcl_ext_localize(char *in, char *out, int max_len, int *id)
{
	// do XSTR translation
	lcl_ext_localize_sub(in, out, max_len, id);

	// do translation of $callsign, $rank, etc.
	lcl_replace_stuff(out, max_len);
}

// translate the specified string based upon the current language
char *XSTR(char *str, int index)
{
	if(!Xstr_inited)
	{
		Int3();
		return str;
	}

	// perform a lookup
	if (index >= 0 && index < XSTR_SIZE)
	{
		// return translation of string
		if (Xstr_table[index].str)
			return Xstr_table[index].str;
	}

	// can't translate; return original english string
	return str;
}

// retrieve the offset for a localized string
int lcl_get_xstr_offset(int index, int res)
{
	if (res == GR_640) {
		return Xstr_table[index].offset_x;
	} else {
		return Xstr_table[index].offset_x_hi;
	}
}


// ------------------------------------------------------------------------------------------------------------
// LOCALIZE FORWARD DEFINITIONS
//

// associate table file externalization with the specified input file
void lcl_ext_associate(char *filename)
{
	// if the filename already exists, free it up
	if(Lcl_ext_filename != NULL){
		vm_free(Lcl_ext_filename);
	}

	// set the new filename
	Lcl_ext_filename = vm_strdup(filename);
}

// given a valid XSTR() tag piece of text, extract the string portion, return it in out, nonzero on success
int lcl_ext_get_text(char *xstr, char *out)
{
	int str_start, str_end;
	int str_len;
	char *p, *p2;

	Assert(xstr != NULL);
	Assert(out != NULL);
	str_len = strlen(xstr);
	
	// this is some crazy wack-ass code.
	// look for the open quote
	str_start = str_end = 0;
	p = strstr(xstr, "\"");
	if(p == NULL){
		error_display(0, "Error parsing XSTR() tag %s\n", xstr);		
		return 0;
	} else {
		str_start = p - xstr + 1;		
	}
	// make sure we're not about to walk past the end of the string
	if((p - xstr) >= str_len){
		error_display(0, "Error parsing XSTR() tag %s\n", xstr);		
		return 0;
	}

	// look for the close quote
	p2 = strstr(p+1, "\"");
	if(p2 == NULL){
		error_display(0, "Error parsing XSTR() tag %s\n", xstr);		
		return 0;
	} else {
		str_end = p2 - xstr;
	}

	// now that we know the boundaries of the actual string in the XSTR() tag. copy it
	memcpy(out, xstr + str_start, str_end - str_start);	

	if (Lcl_pl)
		lcl_fix_polish(out);

	// success
	return 1;
}

// given a valid XSTR() tag piece of text, extract the id# portion, return the value in out, nonzero on success
int lcl_ext_get_id(char *xstr, int *out)
{
	char *p, *pnext;
	int str_len;

	Assert(xstr != NULL);
	Assert(out != NULL);
	
	str_len = strlen(xstr);

	// find the first quote
	p = strstr(xstr, "\"");
	if(p == NULL){
		error_display(0, "Error parsing id# in XSTR() tag %s\n", xstr);
		return 0;
	}
	// make sure we're not about to walk off the end of the string
	if((p - xstr) >= str_len){
		error_display(0, "Error parsing id# in XSTR() tag %s\n", xstr);
		return 0;
	}
	p++;

	// continue searching until we find the close quote
	while(1){
		pnext = strstr(p, "\"");
		if(pnext == NULL){
			error_display(0, "Error parsing id# in XSTR() tag %s\n", xstr);
			return 0;
		}

		// if the previous char is a \, we know its not the "end-of-string" quote
		if(*(pnext - 1) != '\\'){
			p = pnext;
			break;
		}

		// continue
		p = pnext;
	}

	// search until we find a ,	
	pnext = strstr(p, ",");
	if(pnext == NULL){
		error_display(0, "Error parsing id# in XSTR() tag %s\n", xstr);
		return 0;
	}
	// make sure we're not about to walk off the end of the string
	if((pnext - xstr) >= str_len){
		error_display(0, "Error parsing id# in XSTR() tag %s\n", xstr);
		return 0;
	}
	pnext++;
	
	// now get the id string
	p = pnext;
	pnext = strtok(p, ")");
	if(pnext == NULL){
		error_display(0, "Error parsing id# in XSTR() tag %s\n", xstr);
		return 0;
	}

	// get the value and we're done
	*out = atoi(pnext);

	// success
	return 1;
}

// given a valid XSTR() id#, lookup the string in tstrings.tbl, filling in out if found, nonzero on success
int lcl_ext_lookup(char *out, int id)
{
	char text[1024];
	int ret;
	int pointer;
	
	Assert(Lcl_pointer_count >= 0);
	Assert(Lcl_pointers[0] >= 0);
	Assert(Lcl_pointers[Lcl_pointer_count - 1] >= 0);
	Assert(Lcl_ext_file != NULL);
	Assert(id >= 0);

	// seek to the closest pointer <= the id# we're looking for
	pointer = id / LCL_GRANULARITY;
	cfseek(Lcl_ext_file, Lcl_pointers[pointer], CF_SEEK_SET);

	// reset parsing vars and go to town
	Ts_current_state = TS_SCANNING;
	Ts_id_text_size = 0;
//	Ts_text_size;
	memset(Ts_text, 0, PARSE_TEXT_STRING_LEN);
	memset(Ts_id_text, 0, PARSE_ID_STRING_LEN);
	while((cftell(Lcl_ext_file) < Lcl_pointers[Lcl_pointer_count - 1]) && cfgets(text, 1024, Lcl_ext_file)){
		ret = lcl_ext_lookup_sub(text, out, id);
			
		// run the line parse function		
		switch(ret & 0x0fffffff){
		// error
		case 0 :
			Int3();			// should never get here - it means the string doens't exist in the table!!
			return 0;

		// success parsing the line - please continue
		case 1 :
			break;

		// found a matching string/id pair
		case 2 :			
			// success
			if (Lcl_gr) {
				// this is because tstrings.tbl reads in as ANSI for some reason
				// opening tstrings with "rb" mode didnt seem to help, so its now still "rt" like before
				lcl_fix_umlauts(out, LCL_TO_ASCII);
			}
			return 1;

		// end of language found
		case 3 :
			Int3();			// should never get here - it means the string doens't exist in the table!!
			return 0;		
		}
	}
	
	Int3();			// should never get here - it means the string doens't exist in the table!!
	return 0;
}

// sub-parse function for individual lines of tstrings.tbl (from Exstr)
// returns : integer with the low bits having the following values :
// 0 on fail, 1 on success, 2 if found a matching id/string pair, 3 if end of language has been found
// for cases 1 and 2 : the high bit (1<<31) will be set if the parser detected the beginning of a new string id on this line
//
int lcl_ext_lookup_sub(char *text, char *out, int id)
{
	char *front;			// front of the line
	char *p;					// current ptr
	int len = strlen(text);
	int count;	
	char text_copy[1024];	
	char *tok;
	int found_new_string_id = 0;

	front = text;
	p = text;
	count = 0;			
	while(count < len){
		// do something useful
		switch(Ts_current_state){		
		// scanning for a line of text
		case TS_SCANNING:
			// if the first word is #end, we're done with the file altogether
			strcpy_s(text_copy, text);
			tok = strtok(text_copy, " \n");
			if((tok != NULL) && !stricmp(tok, "#end")){
				return 3;
			}
			// if its a commented line, skip it
			else if((text[0] == ';') || (text[0] == ' ') || (text[0] == '\n')){
				return 1;
			}
			// otherwise we should have an ID #, so stuff it and move to the proper state
			else {
				if(lcl_is_valid_numeric_char(*p)){
					memset(Ts_id_text, 0, PARSE_ID_STRING_LEN);
					Ts_id_text_size = 0;
					Ts_id_text[Ts_id_text_size++] = *p;
					Ts_current_state = TS_ID_STRING;

					found_new_string_id = 1;
				}
				// error
				else {
					Int3();
					return 0;
				}
			}
			break;

		// scanning in an id string
		case TS_ID_STRING:
			// if we have another valid char
			if(lcl_is_valid_numeric_char(*p)){
				Ts_id_text[Ts_id_text_size++] = *p;
			}
			// if we found a comma, our id# is finished, look for the open quote
			else if(*p == ','){
				Ts_current_state = TS_OPEN_QUOTE;
			} else {
				Int3();
				return 0;
			}
			break;

		case TS_OPEN_QUOTE:
			// valid space or an open quote
			if((*p == ' ') || (*p == '\"')){
				if(*p == '\"'){
					memset(Ts_text, 0, PARSE_TEXT_STRING_LEN);
					Ts_text_size = 0;
					Ts_current_state = TS_STRING;
				}
			} else {
				Int3();
				return 0;
			}
			break;

		case TS_STRING:
			// if we have an end quote, we need to look for a comma
			if((*p == '\"') /*&& (Ts_text_size > 0)*/ && (Ts_text[Ts_text_size - 1] != '\\')){
				// we're now done - we have a string
				Ts_current_state = TS_SCANNING;

				// if the id#'s match, copy the string and return "string found"
				if((atoi(Ts_id_text) == id) && (out != NULL)){
					strcpy(out, Ts_text);

					return found_new_string_id ? (1<<1) | (1<<31) : (1<<1);					
				}
				
				// otherwise, just continue parsing				
				return found_new_string_id ? (1<<0) | (1<<31) : (1<<0);
			} 
			// otherwise add to the string
			else {
				Ts_text[Ts_text_size++] = *p;
			}										
			break;
		}		

		// if we have a newline, return success, we're done with this line
		if(*p == '\n'){
			return found_new_string_id ? (1<<0) | (1<<31) : (1<<0);
		}

		// next char in the line
		p++;
		count++;
	}	

	// success
	return found_new_string_id ? (1<<0) | (1<<31) : (1<<0);
}

// if the char is a valid char for a signed integer value
int lcl_is_valid_numeric_char(char c)
{
	return ( (c == '-') || (c == '0') || (c == '1') || (c == '2') || (c == '3') || (c == '4') ||
				(c == '5') || (c == '6') || (c == '7') || (c == '8') || (c == '9') ) ? 1 : 0;
}

// initialize the pointer array into tstrings.tbl (call from lcl_ext_open() ONLY)
void lcl_ext_setup_pointers()
{
	char language_string[128];
	char line[1024];
	char *tok;	
	int string_count;
	int ret;
	int found_start = 0;

	// open the localization file
	lcl_ext_open();
	if(Lcl_ext_file == NULL){
		error_display(0, "Error opening externalization file! File likely does not exist or could not be found");
		return;
	}

	// seek to the currently active language
	memset(language_string, 0, 128);
	strcpy_s(language_string, "#");
	if(!stricmp(DEFAULT_LANGUAGE, Lcl_languages[Lcl_current_lang].lang_name)){
		strcat_s(language_string, "default");
	} else {
		strcat_s(language_string, Lcl_languages[Lcl_current_lang].lang_name);
	}
	memset(line, 0, 1024);

	// reset seek variables and begin		
	Lcl_pointer_count = 0;
	while(cfgets(line, 1024, Lcl_ext_file)){
		tok = strtok(line, " \n");
		if(tok == NULL){
			continue;			
		}
		
		// if the language matches, we're good to start parsing strings
		if(!stricmp(language_string, tok)){
			found_start = 1;			
			break;
		}		
	}

	// if we didn't find the language specified, error
	if(found_start <= 0){
		error_display(0, "Could not find specified langauge in tstrings.tbl!\n");
		lcl_ext_close();
		return;
	}

	string_count = 0;	
	while(cfgets(line, 1024, Lcl_ext_file)){
		ret = lcl_ext_lookup_sub(line, NULL, -1);

		// do stuff
		switch(ret & 0x0fffffff){
		// error
		case 0 :
			lcl_ext_close();
			return;		

		// end of language found
		case 3 :
			// mark one final pointer
			Lcl_pointers[Lcl_pointer_count++] = cftell(Lcl_ext_file) - strlen(line) - 1;
			lcl_ext_close();
			return;
		}

		// the only other case we care about is the beginning of a new id#
		if(ret & (1<<31)){		
			if((string_count % LCL_GRANULARITY) == 0){
				// mark the pointer down
				Lcl_pointers[Lcl_pointer_count++] = cftell(Lcl_ext_file) - strlen(line) - 1;

				// if we're out of pointer slots
				if(Lcl_pointer_count >= LCL_MAX_POINTERS){
					error_display(0, "Out of pointer for tstrings.tbl lookup. Please increment LCL_MAX_POINTERS in localize.cpp");
					lcl_ext_close();
					return;
				}
			}
			// increment string count
			string_count++;			
		}
	}

	// should never get here. we should always be exiting through case 3 (end of language section) of the above switch
	// statement
	Int3();
	lcl_ext_close();
}

void lcl_get_language_name(char *lang_name)
{
	Assert(LCL_NUM_LANGUAGES == 3);

	strcpy(lang_name, Lcl_languages[Lcl_current_lang].lang_name);
}

// converts german umlauted chars from ASCII to ANSI
// so they appear in the launcher
// how friggin lame is this
// pass in a null terminated string, foo!
// returns ptr to string you sent in
char* lcl_fix_umlauts(char *str, int which_way)
{
	int i=0;

	if (which_way == LCL_TO_ANSI) {
		// moving to ANSI charset
		// run thru string and perform appropriate conversions
		while (str[i] != '\0') {
			switch (str[i]) {
			case '\x81':
				// lower umlaut u
				str[i] = '\xFC';
				break;
			case '\x84':
				// lower umlaut a
				str[i] = '\xE4';
				break;
			case '\x94':
				// lower umlaut o
				str[i] = '\xF6';
				break;
			case '\x9A':
				// upper umlaut u
				str[i] = '\xDC';
				break;
			case '\x8E':
				// upper umlaut a
				str[i] = '\xC4';
				break;
			case '\x99':
				// upper umlaut o
				str[i] = '\xD6';
				break;
			case '\xE1':
				// beta-lookin thing that means "ss"
				str[i] = '\xDF';
				break;
			}

			i++;
		}
	} else {
		// moving to ASCII charset
		// run thru string and perform appropriate conversions
		while (str[i] != '\0') {
			switch (str[i]) {
			case '\xFC':
				// lower umlaut u
				str[i] = '\x81';
				break;
			case '\xE4':
				// lower umlaut a
				str[i] = '\x84';
				break;
			case '\xF6':
				// lower umlaut o
				str[i] = '\x94';
				break;
			case '\xDC':
				// upper umlaut u
				str[i] = '\x9A';
				break;
			case '\xC4':
				// upper umlaut a
				str[i] = '\x8E';
				break;
			case '\xD6':
				// upper umlaut o
				str[i] = '\x99';
				break;
			case '\xDF':
				// beta-lookin thing that means "ss"
				str[i] = '\xE1';
				break;
			}

			i++;
		}
	}

	return str;
}

// convert some of the polish characters
void lcl_fix_polish(char *str)
{
	for (; *str; str++) {
		if(*str == '\xA2')
			*str = '\xF3';
		else if(*str == '\x88')
			*str = '\xEA';
	}
}

// ------------------------------------------------------------------
// lcl_translate_wep_name()
//
// For displaying weapon names in german version
// since we can't actually just change them outright.
//
void lcl_translate_wep_name(char *name)
{
	if (!strcmp(name, "Morning Star")) {	
		strcpy(name, "Morgenstern");
	} else if (!strcmp(name, "MorningStar")) {
		strcpy(name, "Morgenstern D");
	} else if (!strcmp(name, "UD-8 Kayser")) {
		strcpy(name, "Kayserstrahl");
	} else if (!strcmp(name, "UD-D Kayser")) {
		strcpy(name, "Kayserstrahl");
	}
}

// ------------------------------------------------------------------
// lcl_translate_brief_icon_name()
//
// For displaying ship names in german version
// since we can't actually just change them outright.
//
void lcl_translate_brief_icon_name(char *name)
{
	char *pos;
	char buf[128];

	if (!stricmp(name, "Subspace Portal")) {	
		strcpy(name, "Subraum Portal");

	} else if (!stricmp(name, "Alpha Wing")) {
		strcpy(name, "Alpha");

	} else if (!stricmp(name, "Beta Wing")) {
		strcpy(name, "Beta");

	} else if (!stricmp(name, "Zeta Wing")) {
		strcpy(name, "Zeta");

	} else if (!stricmp(name, "Capella Node")) {
		strcpy(name, "Capella");

	} else if (!stricmp(name, "Hostile")) {
		strcpy(name, "Gegner");

	} else if (!stricmp(name, "Hostile Craft")) {
		strcpy(name, "Gegner");

	} else if (!stricmp(name, "Rebel Wing")) {
		strcpy(name, "Rebellen");

	} else if (!stricmp(name, "Rebel Fleet")) {
		strcpy(name, "Rebellenflotte");

	} else if (!stricmp(name, "Sentry Gun")) {
		strcpy(name, "Gesch\x81tz");

	} else if (!stricmp(name, "Cargo")) {
		strcpy(name, "Fracht");

	} else if (!stricmp(name, "Knossos Device")) {
		strcpy(name, "Knossosger\x84t");
	
	} else if (!stricmp(name, "Support")) {
		strcpy(name, "Versorger");

	} else if (!stricmp(name, "Unknown")) {
		strcpy(name, "Unbekannt");

	} else if (!stricmp(name, "Instructor")) {
		strcpy(name, "Ausbilder");
	
	} else if (!stricmp(name, "Jump Node")) {
		strcpy(name, "Sprungknoten");

	} else if (!stricmp(name, "Escort")) {
		strcpy(name, "Geleitschutz");

	} else if (!stricmp(name, "Asteroid Field")) {
		strcpy(name, "Asteroidenfeld");

	} else if (!stricmp(name, "Enif Station")) {
		strcpy(name, "Station Enif");

	} else if (!stricmp(name, "Rally Point")) {
		strcpy(name, "Sammelpunkt");

	} else if ((pos = strstr(name, "Transport")) != NULL) {
		pos += 9;		// strlen of "transport"
		strcpy_s(buf, "Transporter");
		strcat_s(buf, pos);
		strcpy(name, buf);

	} else if ((pos = strstr(name, "Jump Node")) != NULL) {
		pos += 9;		// strlen of "jump node"
		strcpy_s(buf, "Sprungknoten");
		strcat_s(buf, pos);
		strcpy(name, buf);
	
	} else if (!stricmp(name, "Orion under repair")) {
		strcpy(name, "Orion wird repariert");

	// SOTY-specific ones below!
	
	} else if (!stricmp(name, "Wayfarer Station")) {
		strcpy(name, "Station Wayfarer");
	} else if (!stricmp(name, "Enemy")) {
		strcpy(name, "Gegner");
	} else if (!stricmp(name, "Supply Depot")) {
		strcpy(name, "Nachschubdepot");
	} else if (!stricmp(name, "Fighter Escort")) {
		strcpy(name, "Jagdschutz");
	} else if (!stricmp(name, "Shivans")) {
		strcpy(name, "Shivaner");
	} else if (!stricmp(name, "NTF Base of Operations")) {
		strcpy(name, "NTF-Operationsbasis");
	} else if (!stricmp(name, "NTF Bombers")) {
		strcpy(name, "NTF-Bomber");
	} else if (!stricmp(name, "NTF Fighters")) {
		strcpy(name, "NTF-J\x84ger");
	} else if (!stricmp(name, "Sentry")) {
		strcpy(name, "Sperrgesch\x81tz");
	} else if (!stricmp(name, "Cargo Containers")) {
		strcpy(name, "Frachtbeh\x84lter");
	} else if (!stricmp(name, "NTF Reinforcements")) {
		strcpy(name, "NTF-Verst\x84rkungen");
	} else if (!stricmp(name, "NTF Base")) {
		strcpy(name, "NTF-St\x81tzpunkt");
	} else if (!stricmp(name, "Refugee Convoy")) {
		strcpy(name, "Fl\x81""chtlingskonvoi");
	} else if (!stricmp(name, "Food Convoy")) {
		strcpy(name, "Nachschubkonvoi");
	} else if (!stricmp(name, "Governor's Shuttle")) {
		strcpy(name, "F\x84hre des Gouverneurs");
	} else if (!stricmp(name, "GTVA Patrol")) {
		strcpy(name, "GTVA-Patrouille");
	} else if (!stricmp(name, "Escort fighters")) {
		strcpy(name, "Geleitschutz");
	} else if (!stricmp(name, "Nagada Outpost")) {
		strcpy(name, "Nagada-Aussenposten");
	} else if (!stricmp(name, "Fighters")) {
		strcpy(name, "J\x84ger");
	} else if (!stricmp(name, "Bombers")) {
		strcpy(name, "Bomber");
	} else if (!stricmp(name, "Enemy Destroyers")) {
		strcpy(name, "Feindliche Zerst\x94rer");
	} else if (!stricmp(name, "Ross 128 System")) {
		strcpy(name, "System Ross 128");
	} else if (!stricmp(name, "Knossos Station")) {
		strcpy(name, "Knossos-Station");
	} else if (!stricmp(name, "Transporters")) {
		strcpy(name, "Transporter");
	} else if (!stricmp(name, "Pirates?")) {
		strcpy(name, "Piraten?");
	} else if (!stricmp(name, "Escorts")) {
		strcpy(name, "Geleitschutz");
	} else if (!stricmp(name, "Shivan Fighters")) {
		strcpy(name, "J\x84ger");
	} else if (!stricmp(name, "Shivan Territory")) {
		strcpy(name, "Shivaner");
	}
}

// ------------------------------------------------------------------
// lcl_translate_brief_icon_name_pl()
//
// For displaying ship names in polish version
// since we can't actually just change them outright.
//
void lcl_translate_brief_icon_name_pl(char *name)
{
char *pos;
char buf[128];
  	 
	if (!stricmp(name, "Subspace Portal")) {
		strcpy(name, "Portal podprz.");

	} else if (!stricmp(name, "Alpha Wing")) {
		strcpy(name, "Alfa");

	} else if (!stricmp(name, "Beta Wing")) {
		strcpy(name, "Beta");

	} else if (!stricmp(name, "Zeta Wing")) {
		strcpy(name, "Zeta");

	} else if (!stricmp(name, "Capella Node")) {
		strcpy(name, "Capella");

	} else if (!stricmp(name, "Hostile")) {
		strcpy(name, "Wr\xF3g");

	} else if (!stricmp(name, "Hostile Craft")) {
		strcpy(name, "Wr\xF3g");

	} else if (!stricmp(name, "Rebel Wing")) {
		strcpy(name, "Rebelianci");

	} else if (!stricmp(name, "Rebel Fleet")) {
		strcpy(name, "Flota Rebelii");

	} else if (!stricmp(name, "Sentry Gun")) {
		strcpy(name, "Dzia\xB3o str.");

	} else if (!stricmp(name, "Cargo")) {
		strcpy(name, "\xA3\x61\x64unek");

	} else if (!stricmp(name, "Knossos Device")) {
		strcpy(name, "Urz. Knossos");

	} else if (!stricmp(name, "Support")) {
		strcpy(name, "Wsparcie");

	} else if (!stricmp(name, "Unknown")) {
		strcpy(name, "Nieznany");

	} else if (!stricmp(name, "Instructor")) {
		strcpy(name, "Instruktor");

	} else if (!stricmp(name, "Jump Node")) {
		strcpy(name, "W\xEAze\xB3 skokowy");

	} else if (!stricmp(name, "Escort")) {
		strcpy(name, "Eskorta");

	} else if (!stricmp(name, "Asteroid Field")) {
		strcpy(name, "Pole asteroid");

	} else if (!stricmp(name, "Enif Station")) {
		strcpy(name, "Stacja Enif");

	} else if (!stricmp(name, "Rally Point")) {
		strcpy(name, "Pkt zborny");

	} else if ((pos = strstr(name, "Transport")) != NULL) {
		pos += 9;		// strlen of "transport"
		strcpy_s(buf, "Transporter");
		strcat_s(buf, pos);
		strcpy(name, buf);

	} else if ((pos = strstr(name, "Jump Node")) != NULL) {
		pos += 9;		// strlen of "jump node"
		strcpy_s(buf, "W\xEAze\xB3 skokowy");
		strcat_s(buf, pos);
		strcpy(name, buf);

	} else if (!stricmp(name, "Orion under repair")) {
		strcpy(name, "Naprawiany Orion");
	}
}

// ------------------------------------------------------------------
// lcl_translate_ship_name()
//
// For displaying ship names in german version in the briefing
// since we can't actually just change them outright.
//
void lcl_translate_ship_name(char *name)
{
	if (!strcmp(name, "GTDR Amazon Advanced")) {	
		strcpy(name, "GTDR Amazon VII");
	} 
}

// ------------------------------------------------------------------
// lcl_translate_targetbox_name()
//
// For displaying ship names in german version in the targetbox
// since we can't actually just change them outright.
//
void lcl_translate_targetbox_name(char *name)
{
	char *pos;
	char buf[128];
	
	if ((pos = strstr(name, "Sentry")) != NULL) {
		pos += 6;		// strlen of "sentry"
		strcpy_s(buf, "Sperrgesch\x81tz");
		strcat_s(buf, pos);
		strcpy(name, buf);

	} else if ((pos = strstr(name, "Support")) != NULL) {
		pos += 7;		// strlen of "support"
		strcpy_s(buf, "Versorger");
		strcat_s(buf, pos);
		strcpy(name, buf);

	} else if ((pos = strstr(name, "Unknown")) != NULL) {
		pos += 7;		// strlen of "unknown"
		strcpy_s(buf, "Unbekannt");
		strcat_s(buf, pos);
		strcpy(name, buf);

	} else if ((pos = strstr(name, "Drone")) != NULL) {
		pos += 5;		// strlen of "drone"
		strcpy_s(buf, "Drohne");
		strcat_s(buf, pos);
		strcpy(name, buf);

	} else if ((pos = strstr(name, "Jump Node")) != NULL) {
		pos += 9;		// strlen of "jump node"
		strcpy_s(buf, "Sprungknoten");
		strcat_s(buf, pos);
		strcpy(name, buf);

	} else if (!stricmp(name, "Instructor")) {
		strcpy(name, "Ausbilder");

	} else if (!stricmp(name, "NTF Vessel")) {
		strcpy(name, "NTF-Schiff");

	} else if (!stricmp(name, "Enif Station")) {
		strcpy(name, "Station Enif");
	}
}

// ------------------------------------------------------------------
// lcl_translate_targetbox_name_pl()
//
// For displaying ship names in polish version in the targetbox
// since we can't actually just change them outright.
//
void lcl_translate_targetbox_name_pl(char *name)
{
	char *pos;
	char buf[128];

	if ((pos = strstr(name, "Sentry")) != NULL) {
		pos += 6;		// strlen of "sentry"
		strcpy_s(buf, "Stra\xBFnik");
		strcat_s(buf, pos);
		strcpy(name, buf);

	} else if ((pos = strstr(name, "Support")) != NULL) {
		pos += 7;		// strlen of "support"
		strcpy_s(buf, "Wsparcie");
		strcat_s(buf, pos);
		strcpy(name, buf);

	} else if ((pos = strstr(name, "Unknown")) != NULL) {
		pos += 7;		// strlen of "unknown"
		strcpy_s(buf, "Nieznany");
		strcat_s(buf, pos);
		strcpy(name, buf);

	} else if ((pos = strstr(name, "Drone")) != NULL) {
		pos += 5;		// strlen of "drone"
		strcpy_s(buf, "Sonda");
		strcat_s(buf, pos);
		strcpy(name, buf);

	} else if ((pos = strstr(name, "Jump Node")) != NULL) {
		pos += 9;		// strlen of "jump node"
		strcpy_s(buf, "W\xEAze\xB3 skokowy");
		strcat_s(buf, pos);
		strcpy(name, buf);

	} else if (!stricmp(name, "Instructor")) {
		strcpy(name, "Instruktor");

	} else if (!stricmp(name, "NTF Vessel")) {
		strcpy(name, "Okr\xEAt NTF");

	} else if (!stricmp(name, "Enif Station")) {
		strcpy(name, "Stacja Enif");
	}
}