1

/* $Id$ */

     2

     3

#include "../stdafx.h"

     4

#include "../macros.h"

     5

#include "../string.h"

     6

#include "../table/control_codes.h"

     7

#include "../helpers.hpp"

     8

#include <stdio.h>

     9

#include <string.h>

    10

#include <stdlib.h>

    11

#include <stdarg.h>

    12

    13

#if (!defined(WIN32) && !defined(WIN64)) || defined(__CYGWIN__)

    14

#include <unistd.h>

    15

#include <sys/types.h>

    16

#include <sys/stat.h>

    17

#endif

    18

    19

#if defined WIN32 || defined __WATCOMC__

    20

#include <direct.h>

    21

#endif /* WIN32 || __WATCOMC__ */

    22

    23

#ifdef __MORPHOS__

    24

#ifdef stderr

    25

#undef stderr

    26

#endif

    27

#define stderr stdout

    28

#endif /* __MORPHOS__ */

    29

    30

#ifdef __WATCOMC__

    31

	uint _map_log_x;     // an unpleasant hack required because Watcom is insisting on

    32

	uint _map_size_x;    // these variables being valid references in map.h

    33

	uint _map_size_y;

    34

	uint _map_tile_mask;

    35

	uint _map_size;

    36

#endif /* __WATCOMC__ */

    37

    38

/* Compiles a list of strings into a compiled string list */

    39

    40

typedef void (*ParseCmdProc)(char *buf, int value);

    41

    42

typedef struct LanguagePackHeader {

    43

	uint32 ident;

    44

	uint32 version;     // 32-bits of auto generated version info which is basically a hash of strings.h

    45

	char name[32];      // the international name of this language

    46

	char own_name[32];  // the localized name of this language

    47

	char isocode[16];   // the ISO code for the language (not country code)

    48

	uint16 offsets[32]; // the offsets

    49

	byte plural_form;   // plural form index

    50

	byte pad[3];        // pad header to be a multiple of 4

    51

} LanguagePackHeader;

    52

    53

typedef struct CmdStruct {

    54

	const char *cmd;

    55

	ParseCmdProc proc;

    56

	long value;

    57

	int8 consumes;

    58

	byte flags;

    59

} CmdStruct;

    60

    61

enum {

    62

	C_DONTCOUNT = 1,

    63

	C_CASE      = 2,

    64

};

    65

    66

    67

typedef struct Case {

    68

	int caseidx;

    69

	char *string;

    70

	struct Case *next;

    71

} Case;

    72

    73

static bool _masterlang;

    74

static bool _translated;

    75

static const char* _file = "(unknown file)";

    76

static int _cur_line;

    77

static int _errors, _warnings;

    78

    79

typedef struct LangString {

    80

	char *name;            // Name of the string

    81

	char *english;         // English text

    82

	char *translated;      // Translated text

    83

	uint16 hash_next;      // next hash entry

    84

	uint16 index;

    85

	int line;              // line of string in source-file

    86

	Case *english_case;    // cases for english

    87

	Case *translated_case; // cases for foreign

    88

} LangString;

    89

    90

static LangString *_strings[65536];

    91

    92

    93

#define HASH_SIZE 32767

    94

static uint16 _hash_head[HASH_SIZE];

    95

    96

static byte _put_buf[4096];

    97

static int _put_pos;

    98

static int _next_string_id;

    99

   100

static uint32 _hash;

   101

static char _lang_name[32], _lang_ownname[32], _lang_isocode[16];

   102

static byte _lang_pluralform;

   103

#define MAX_NUM_GENDER 8

   104

static char _genders[MAX_NUM_GENDER][8];

   105

static int _numgenders;

   106

   107

// contains the name of all cases.

   108

#define MAX_NUM_CASES 50

   109

static char _cases[MAX_NUM_CASES][16];

   110

static int _numcases;

   111

   112

// for each plural value, this is the number of plural forms.

   113

static const byte _plural_form_counts[] = { 2, 1, 2, 3, 3, 3, 3, 3, 4 };

   114

   115

static const char *_cur_ident;

   116

   117

typedef struct CmdPair {

   118

	const CmdStruct *a;

   119

	const char *v;

   120

} CmdPair;

   121

   122

typedef struct ParsedCommandStruct {

   123

	int np;

   124

	CmdPair pairs[32];

   125

	const CmdStruct *cmd[32]; // ordered by param #

   126

} ParsedCommandStruct;

   127

   128

// Used when generating some advanced commands.

   129

static ParsedCommandStruct _cur_pcs;

   130

static int _cur_argidx;

   131

   132

static uint HashStr(const char *s)

   133

{

   134

	uint hash = 0;

   135

	for (; *s != '\0'; s++) hash = ROL(hash, 3) ^ *s;

   136

	return hash % HASH_SIZE;

   137

}

   138

   139

static void HashAdd(const char *s, LangString *ls)

   140

{

   141

	uint hash = HashStr(s);

   142

	ls->hash_next = _hash_head[hash];

   143

	_hash_head[hash] = ls->index + 1;

   144

}

   145

   146

static LangString *HashFind(const char *s)

   147

{

   148

	int idx = _hash_head[HashStr(s)];

   149

   150

	while (--idx >= 0) {

   151

		LangString* ls = _strings[idx];

   152

   153

		if (strcmp(ls->name, s) == 0) return ls;

   154

		idx = ls->hash_next;

   155

	}

   156

	return NULL;

   157

}

   158

   159

#ifdef _MSC_VER

   160

# define LINE_NUM_FMT "(%d)"

   161

#else

   162

# define LINE_NUM_FMT ":%d"

   163

#endif

   164

   165

static void CDECL warning(const char *s, ...)

   166

{

   167

	char buf[1024];

   168

	va_list va;

   169

	va_start(va, s);

   170

	vsnprintf(buf, lengthof(buf), s, va);

   171

	va_end(va);

   172

	fprintf(stderr, "%s" LINE_NUM_FMT ": warning: %s\n", _file, _cur_line, buf);

   173

	_warnings++;

   174

}

   175

   176

void CDECL error(const char *s, ...)

   177

{

   178

	char buf[1024];

   179

	va_list va;

   180

	va_start(va, s);

   181

	vsnprintf(buf, lengthof(buf), s, va);

   182

	va_end(va);

   183

	fprintf(stderr, "%s" LINE_NUM_FMT ": error: %s\n", _file, _cur_line, buf);

   184

	_errors++;

   185

}

   186

   187

   188

static void NORETURN CDECL fatal(const char *s, ...)

   189

{

   190

	char buf[1024];

   191

	va_list va;

   192

	va_start(va, s);

   193

	vsnprintf(buf, lengthof(buf), s, va);

   194

	va_end(va);

   195

	fprintf(stderr, "%s" LINE_NUM_FMT ": FATAL: %s\n", _file, _cur_line, buf);

   196

	exit(1);

   197

}

   198

   199

static void PutByte(byte c)

   200

{

   201

	if (_put_pos == lengthof(_put_buf)) fatal("Put buffer too small");

   202

	_put_buf[_put_pos++] = c;

   203

}

   204

   205

   206

static void PutUtf8(uint32 value)

   207

{

   208

	if (value < 0x80) {

   209

		PutByte(value);

   210

	} else if (value < 0x800) {

   211

		PutByte(0xC0 + GB(value,  6, 5));

   212

		PutByte(0x80 + GB(value,  0, 6));

   213

	} else if (value < 0x10000) {

   214

		PutByte(0xE0 + GB(value, 12, 4));

   215

		PutByte(0x80 + GB(value,  6, 6));

   216

		PutByte(0x80 + GB(value,  0, 6));

   217

	} else if (value < 0x110000) {

   218

		PutByte(0xF0 + GB(value, 18, 3));

   219

		PutByte(0x80 + GB(value, 12, 6));

   220

		PutByte(0x80 + GB(value,  6, 6));

   221

		PutByte(0x80 + GB(value,  0, 6));

   222

	} else {

   223

		warning("Invalid unicode value U+0x%X", value);

   224

	}

   225

}

   226

   227

   228

size_t Utf8Validate(const char *s)

   229

{

   230

	uint32 c;

   231

   232

	if (!HASBIT(s[0], 7)) {

   233

		/* 1 byte */

   234

		return 1;

   235

	} else if (GB(s[0], 5, 3) == 6 && IsUtf8Part(s[1])) {

   236

		/* 2 bytes */

   237

		c = GB(s[0], 0, 5) << 6 | GB(s[1], 0, 6);

   238

		if (c >= 0x80) return 2;

   239

	} else if (GB(s[0], 4, 4) == 14 && IsUtf8Part(s[1]) && IsUtf8Part(s[2])) {

   240

		/* 3 bytes */

   241

		c = GB(s[0], 0, 4) << 12 | GB(s[1], 0, 6) << 6 | GB(s[2], 0, 6);

   242

		if (c >= 0x800) return 3;

   243

	} else if (GB(s[0], 3, 5) == 30 && IsUtf8Part(s[1]) && IsUtf8Part(s[2]) && IsUtf8Part(s[3])) {

   244

		/* 4 bytes */

   245

		c = GB(s[0], 0, 3) << 18 | GB(s[1], 0, 6) << 12 | GB(s[2], 0, 6) << 6 | GB(s[3], 0, 6);

   246

		if (c >= 0x10000 && c <= 0x10FFFF) return 4;

   247

	}

   248

   249

	return 0;

   250

}

   251

   252

   253

static void EmitSingleChar(char *buf, int value)

   254

{

   255

	if (*buf != '\0') warning("Ignoring trailing letters in command");

   256

	PutUtf8(value);

   257

}

   258

   259

   260

static void EmitSetX(char *buf, int value)

   261

{

   262

	char *err;

   263

	int x = strtol(buf, &err, 0);

   264

	if (*err != 0) fatal("SetX param invalid");

   265

	PutUtf8(SCC_SETX);

   266

	PutByte((byte)x);

   267

}

   268

   269

   270

static void EmitSetXY(char *buf, int value)

   271

{

   272

	char *err;

   273

	int x;

   274

	int y;

   275

   276

	x = strtol(buf, &err, 0);

   277

	if (*err != ' ') fatal("SetXY param invalid");

   278

	y = strtol(err + 1, &err, 0);

   279

	if (*err != 0) fatal("SetXY param invalid");

   280

   281

	PutUtf8(SCC_SETXY);

   282

	PutByte((byte)x);

   283

	PutByte((byte)y);

   284

}

   285

   286

// The plural specifier looks like

   287

// {NUM} {PLURAL -1 passenger passengers} then it picks either passenger/passengers depending on the count in NUM

   288

   289

// This is encoded like

   290

//  CommandByte <ARG#> <NUM> {Length of each string} {each string}

   291

   292

bool ParseRelNum(char **buf, int *value)

   293

{

   294

	const char* s = *buf;

   295

	char* end;

   296

	bool rel = false;

   297

	int v;

   298

   299

	while (*s == ' ' || *s == '\t') s++;

   300

	if (*s == '+') {

   301

		rel = true;

   302

		s++;

   303

	}

   304

	v = strtol(s, &end, 0);

   305

	if (end == s) return false;

   306

	if (rel || v < 0) {

   307

		*value += v;

   308

	} else {

   309

		*value = v;

   310

	}

   311

	*buf = end;

   312

	return true;

   313

}

   314

   315

// Parse out the next word, or NULL

   316

char *ParseWord(char **buf)

   317

{

   318

	char *s = *buf, *r;

   319

   320

	while (*s == ' ' || *s == '\t') s++;

   321

	if (*s == '\0') return NULL;

   322

   323

	if (*s == '"') {

   324

		r = ++s;

   325

		// parse until next " or NUL

   326

		for (;;) {

   327

			if (*s == '\0') break;

   328

			if (*s == '"') {

   329

				*s++ = '\0';

   330

				break;

   331

			}

   332

			s++;

   333

		}

   334

	} else {

   335

		// proceed until whitespace or NUL

   336

		r = s;

   337

		for (;;) {

   338

			if (*s == '\0') break;

   339

			if (*s == ' ' || *s == '\t') {

   340

				*s++ = '\0';

   341

				break;

   342

			}

   343

			s++;

   344

		}

   345

	}

   346

	*buf = s;

   347

	return r;

   348

}

   349

   350

// Forward declaration

   351

static int TranslateArgumentIdx(int arg);

   352

   353

static void EmitWordList(const char* const* words, uint nw)

   354

{

   355

	uint i;

   356

	uint j;

   357

   358

	PutByte(nw);

   359

	for (i = 0; i < nw; i++) PutByte(strlen(words[i]));

   360

	for (i = 0; i < nw; i++) {

   361

		for (j = 0; words[i][j] != '\0'; j++) PutByte(words[i][j]);

   362

	}

   363

}

   364

   365

static void EmitPlural(char *buf, int value)

   366

{

   367

	int argidx = _cur_argidx;

   368

	const char* words[5];

   369

	int nw = 0;

   370

   371

	// Parse out the number, if one exists. Otherwise default to prev arg.

   372

	if (!ParseRelNum(&buf, &argidx)) argidx--;

   373

   374

	// Parse each string

   375

	for (nw = 0; nw < 5; nw++) {

   376

		words[nw] = ParseWord(&buf);

   377

		if (words[nw] == NULL) break;

   378

	}

   379

   380

	if (nw == 0)

   381

		fatal("%s: No plural words", _cur_ident);

   382

   383

	if (_plural_form_counts[_lang_pluralform] != nw) {

   384

		if (_translated) {

   385

			fatal("%s: Invalid number of plural forms. Expecting %d, found %d.", _cur_ident,

   386

				_plural_form_counts[_lang_pluralform], nw);

   387

		} else {

   388

			warning("'%s' is untranslated. Tweaking english string to allow compilation for plural forms", _cur_ident);

   389

			if (nw > _plural_form_counts[_lang_pluralform]) {

   390

				nw = _plural_form_counts[_lang_pluralform];

   391

			} else {

   392

				for (; nw < _plural_form_counts[_lang_pluralform]; nw++) {

   393

					words[nw] = words[nw - 1];

   394

				}

   395

			}

   396

		}

   397

	}

   398

   399

	PutUtf8(SCC_PLURAL_LIST);

   400

	PutByte(TranslateArgumentIdx(argidx));

   401

	EmitWordList(words, nw);

   402

}

   403

   404

   405

static void EmitGender(char *buf, int value)

   406

{

   407

	int argidx = _cur_argidx;

   408

	uint nw;

   409

   410

	if (buf[0] == '=') {

   411

		buf++;

   412

   413

		// This is a {G=DER} command

   414

		for (nw = 0; ; nw++) {

   415

			if (nw >= 8) fatal("G argument '%s' invalid", buf);

   416

			if (strcmp(buf, _genders[nw]) == 0) break;

   417

		}

   418

		// now nw contains the gender index

   419

		PutUtf8(SCC_GENDER_INDEX);

   420

		PutByte(nw);

   421

	} else {

   422

		const char* words[8];

   423

   424

		// This is a {G 0 foo bar two} command.

   425

		// If no relative number exists, default to +0

   426

		if (!ParseRelNum(&buf, &argidx)) {}

   427

   428

		for (nw = 0; nw < 8; nw++) {

   429

			words[nw] = ParseWord(&buf);

   430

			if (words[nw] == NULL) break;

   431

		}

   432

		if (nw != _numgenders) fatal("Bad # of arguments for gender command");

   433

		PutUtf8(SCC_GENDER_LIST);

   434

		PutByte(TranslateArgumentIdx(argidx));

   435

		EmitWordList(words, nw);

   436

	}

   437

}

   438

   439

   440

static const CmdStruct _cmd_structs[] = {

   441

	// Update position

   442

	{"SETX",  EmitSetX,  SCC_SETX,  0, 0},

   443

	{"SETXY", EmitSetXY, SCC_SETXY, 0, 0},

   444

   445

	// Font size

   446

	{"TINYFONT", EmitSingleChar, SCC_TINYFONT, 0, 0},

   447

	{"BIGFONT",  EmitSingleChar, SCC_BIGFONT,  0, 0},

   448

   449

	// Colors

   450

	{"BLUE",    EmitSingleChar, SCC_BLUE,    0, 0},

   451

	{"SILVER",  EmitSingleChar, SCC_SILVER,  0, 0},

   452

	{"GOLD",    EmitSingleChar, SCC_GOLD,    0, 0},

   453

	{"RED",     EmitSingleChar, SCC_RED,     0, 0},

   454

	{"PURPLE",  EmitSingleChar, SCC_PURPLE,  0, 0},

   455

	{"LTBROWN", EmitSingleChar, SCC_LTBROWN, 0, 0},

   456

	{"ORANGE",  EmitSingleChar, SCC_ORANGE,  0, 0},

   457

	{"GREEN",   EmitSingleChar, SCC_GREEN,   0, 0},

   458

	{"YELLOW",  EmitSingleChar, SCC_YELLOW,  0, 0},

   459

	{"DKGREEN", EmitSingleChar, SCC_DKGREEN, 0, 0},

   460

	{"CREAM",   EmitSingleChar, SCC_CREAM,   0, 0},

   461

	{"BROWN",   EmitSingleChar, SCC_BROWN,   0, 0},

   462

	{"WHITE",   EmitSingleChar, SCC_WHITE,   0, 0},

   463

	{"LTBLUE",  EmitSingleChar, SCC_LTBLUE,  0, 0},

   464

	{"GRAY",    EmitSingleChar, SCC_GRAY,    0, 0},

   465

	{"DKBLUE",  EmitSingleChar, SCC_DKBLUE,  0, 0},

   466

	{"BLACK",   EmitSingleChar, SCC_BLACK,   0, 0},

   467

   468

	{"CURRCOMPACT",   EmitSingleChar, SCC_CURRENCY_COMPACT,    1, 0}, // compact currency (32 bits)

   469

	{"REV",           EmitSingleChar, SCC_REVISION,            0, 0}, // openttd revision string

   470

	{"SHORTCARGO",    EmitSingleChar, SCC_CARGO_SHORT,         2, 0}, // short cargo description, only ### tons, or ### litres

   471

	{"CURRCOMPACT64", EmitSingleChar, SCC_CURRENCY_COMPACT_64, 2, 0}, // compact currency 64 bits

   472

   473

	// These are special versions of {STRING1}

   474

	// The first string includes the second string.

   475

	{"COMPANY",    EmitSingleChar, SCC_STRING1, 1, 0},

   476

	{"PLAYERNAME", EmitSingleChar, SCC_STRING1, 1, 0},

   477

	{"VEHICLE",    EmitSingleChar, SCC_STRING1, 1, 0},

   478

   479

	{"STRING1", EmitSingleChar, SCC_STRING1, 1, C_CASE}, // included string that consumes ONE argument

   480

	{"STRING2", EmitSingleChar, SCC_STRING2, 2, C_CASE}, // included string that consumes TWO arguments

   481

	{"STRING3", EmitSingleChar, SCC_STRING3, 3, C_CASE}, // included string that consumes THREE arguments

   482

	{"STRING4", EmitSingleChar, SCC_STRING4, 4, C_CASE}, // included string that consumes FOUR arguments

   483

	{"STRING5", EmitSingleChar, SCC_STRING5, 5, C_CASE}, // included string that consumes FIVE arguments

   484

   485

	{"STATIONFEATURES", EmitSingleChar, SCC_STATION_FEATURES, 1, 0}, // station features string, icons of the features

   486

	{"INDUSTRY",        EmitSingleChar, SCC_INDUSTRY_NAME,    1, 0}, // industry, takes an industry #