1

/* $Id$ */

     2

     3

/** @file strgen.cpp Tool to create computer readable (stand-alone) translation files. */

     4

     5

#include "../stdafx.h"

     6

#include "../core/alloc_func.hpp"

     7

#include "../core/endian_func.hpp"

     8

#include "../string_func.h"

     9

#include "../strings_type.h"

    10

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

    11

    12

#include <stdio.h>

    13

#include <string.h>

    14

#include <stdlib.h>

    15

#include <stdarg.h>

    16

    17

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

    18

#include <unistd.h>

    19

#include <sys/types.h>

    20

#include <sys/stat.h>

    21

#endif

    22

    23

#if defined WIN32 || defined __WATCOMC__

    24

#include <direct.h>

    25

#endif /* WIN32 || __WATCOMC__ */

    26

    27

#ifdef __MORPHOS__

    28

#ifdef stderr

    29

#undef stderr

    30

#endif

    31

#define stderr stdout

    32

#endif /* __MORPHOS__ */

    33

    34

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

    35

    36

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

    37

    38

struct LanguagePackHeader {

    39

	uint32 ident;       // 32-bits identifier

    40

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

    41

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

    42

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

    43

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

    44

	uint16 offsets[32]; // the offsets

    45

	byte plural_form;   // plural form index

    46

	byte text_dir;      // default direction of the text

    47

	byte pad[2];        // pad header to be a multiple of 4

    48

};

    49

    50

struct CmdStruct {

    51

	const char *cmd;

    52

	ParseCmdProc proc;

    53

	long value;

    54

	int8 consumes;

    55

	byte flags;

    56

};

    57

    58

enum {

    59

	C_DONTCOUNT = 1,

    60

	C_CASE      = 2,

    61

};

    62

    63

    64

struct Case {

    65

	int caseidx;

    66

	char *string;

    67

	Case *next;

    68

};

    69

    70

static bool _masterlang;

    71

static bool _translated;

    72

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

    73

static int _cur_line;

    74

static int _errors, _warnings, _show_todo;

    75

    76

struct LangString {

    77

	char *name;            // Name of the string

    78

	char *english;         // English text

    79

	char *translated;      // Translated text

    80

	uint16 hash_next;      // next hash entry

    81

	uint16 index;

    82

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

    83

	Case *english_case;    // cases for english

    84

	Case *translated_case; // cases for foreign

    85

};

    86

    87

static LangString *_strings[65536];

    88

    89

    90

#define HASH_SIZE 32767

    91

static uint16 _hash_head[HASH_SIZE];

    92

    93

static byte _put_buf[4096];

    94

static int _put_pos;

    95

static int _next_string_id;

    96

    97

static uint32 _hash;

    98

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

    99

static byte _lang_pluralform;

   100

static byte _lang_textdir;

   101

#define MAX_NUM_GENDER 8

   103

static int _numgenders;

   104

   105

// contains the name of all cases.

   106

#define MAX_NUM_CASES 50

   107

static char _cases[MAX_NUM_CASES][16];

   108

static int _numcases;

   109

   110

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

   111

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

   112

   113

static const char *_cur_ident;

   114

   115

struct CmdPair {

   116

	const CmdStruct *a;

   117

	const char *v;

   118

};

   119

   120

struct ParsedCommandStruct {

   121

	int np;

   122

	CmdPair pairs[32];

   123

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

   124

};

   125

   126

// Used when generating some advanced commands.

   127

static ParsedCommandStruct _cur_pcs;

   128

static int _cur_argidx;

   129

   130

static uint HashStr(const char *s)

   131

{

   132

	uint hash = 0;

   133

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

   134

	return hash % HASH_SIZE;

   135

}

   136

   137

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

   138

{

   139

	uint hash = HashStr(s);

   140

	ls->hash_next = _hash_head[hash];

   141

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

   142

}

   143

   144

static LangString *HashFind(const char *s)

   145

{

   146

	int idx = _hash_head[HashStr(s)];

   147

   148

	while (--idx >= 0) {

   149

		LangString* ls = _strings[idx];

   150

   151

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

   152

		idx = ls->hash_next;

   153

	}

   154

	return NULL;

   155

}

   156

   157

#ifdef _MSC_VER

   158

# define LINE_NUM_FMT "(%d)"

   159

#else

   160

# define LINE_NUM_FMT ":%d"

   161

#endif

   162

   163

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

   164

{

   165

	char buf[1024];

   166

	va_list va;

   167

	va_start(va, s);

   168

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

   169

	va_end(va);

   170

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

   171

	_warnings++;

   172

}

   173

   174

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

   175

{

   176

	char buf[1024];

   177

	va_list va;

   178

	va_start(va, s);

   179

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

   180

	va_end(va);

   181

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

   182

	_errors++;

   183

}

   184

   185

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

   186

{

   187

	char buf[1024];

   188

	va_list va;

   189

	va_start(va, s);

   190

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

   191

	va_end(va);

   192

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

   193

	exit(1);

   194

}

   195

   196

static void PutByte(byte c)

   197

{

   198

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

   199

	_put_buf[_put_pos++] = c;

   200

}

   201

   202

   203

static void PutUtf8(uint32 value)

   204

{

   205

	if (value < 0x80) {

   206

		PutByte(value);

   207

	} else if (value < 0x800) {

   208

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

   209

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

   210

	} else if (value < 0x10000) {

   211

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

   212

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

   213

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

   214

	} else if (value < 0x110000) {

   215

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

   216

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

   217

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

   218

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

   219

	} else {

   220

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

   221

	}

   222

}

   223

   224

   225

size_t Utf8Validate(const char *s)

   226

{

   227

	uint32 c;

   228

   229

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

   230

		/* 1 byte */

   231

		return 1;

   232

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

   233

		/* 2 bytes */

   234

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

   235

		if (c >= 0x80) return 2;

   236

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

   237

		/* 3 bytes */

   238

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

   239

		if (c >= 0x800) return 3;

   240

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

   241

		/* 4 bytes */

   242

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

   243

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

   244

	}

   245

   246

	return 0;

   247

}

   248

   249

   250

static void EmitSingleChar(char *buf, int value)

   251

{

   252

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

   253

	PutUtf8(value);

   254

}

   255

   256

   257

static void EmitSetX(char *buf, int value)

   258

{

   259

	char *err;

   260

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

   261

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

   262

	PutUtf8(SCC_SETX);

   263

	PutByte((byte)x);

   264

}

   265

   266

   267

static void EmitSetXY(char *buf, int value)

   268

{

   269

	char *err;

   270

	int x;

   271

	int y;

   272

   273

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

   274

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

   275

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

   276

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

   277

   278

	PutUtf8(SCC_SETXY);

   279

	PutByte((byte)x);

   280

	PutByte((byte)y);

   281

}

   282

   283

// The plural specifier looks like

   284

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

   285

   286

// This is encoded like

   287

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

   288

   289

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

   290

{

   291

	const char* s = *buf;

   292

	char* end;

   293

	bool rel = false;

   294

	int v;

   295

   296

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

   297

	if (*s == '+') {

   298

		rel = true;

   299

		s++;

   300

	}

   301

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

   302

	if (end == s) return false;

   303

	if (rel || v < 0) {

   304

		*value += v;

   305

	} else {

   306

		*value = v;

   307

	}

   308

	*buf = end;

   309

	return true;

   310

}

   311

   312

// Parse out the next word, or NULL

   313

char *ParseWord(char **buf)

   314

{

   315

	char *s = *buf, *r;

   316

   317

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

   318

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

   319

   320

	if (*s == '"') {

   321

		r = ++s;

   322

		// parse until next " or NUL

   323

		for (;;) {

   324

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

   325

			if (*s == '"') {

   326

				*s++ = '\0';

   327

				break;

   328

			}

   329

			s++;

   330

		}

   331

	} else {

   332

		// proceed until whitespace or NUL

   333

		r = s;

   334

		for (;;) {

   335

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

   336

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

   337

				*s++ = '\0';

   338

				break;

   339

			}

   340

			s++;

   341

		}

   342

	}

   343

	*buf = s;

   344

	return r;

   345

}

   346

   347

// Forward declaration

   348

static int TranslateArgumentIdx(int arg);

   349

   350

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

   351

{

   352

	uint i;

   353

	uint j;

   354

   355

	PutByte(nw);

   356

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

   357

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

   358

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

   359

	}

   360

}

   361

   362

static void EmitPlural(char *buf, int value)

   363

{

   364

	int argidx = _cur_argidx;

   365

	const char* words[5];

   366

	int nw = 0;

   367

   368

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

   369

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

   370

   371

	// Parse each string

   372

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

   373

		words[nw] = ParseWord(&buf);

   374

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

   375

	}

   376

   377

	if (nw == 0)

   378

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

   379

   380

	if (_plural_form_counts[_lang_pluralform] != nw) {

   381

		if (_translated) {

   382

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

   383

				_plural_form_counts[_lang_pluralform], nw);

   384

		} else {

   385

			if ((_show_todo & 2) != 0) strgen_warning("'%s' is untranslated. Tweaking english string to allow compilation for plural forms", _cur_ident);

   386

			if (nw > _plural_form_counts[_lang_pluralform]) {

   387

				nw = _plural_form_counts[_lang_pluralform];

   388

			} else {

   389

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

   390

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

   391

				}

   392

			}

   393

		}

   394

	}

   395

   396

	PutUtf8(SCC_PLURAL_LIST);

   397

	PutByte(TranslateArgumentIdx(argidx));

   398

	EmitWordList(words, nw);

   399

}

   400

   401

   402

static void EmitGender(char *buf, int value)

   403

{

   404

	int argidx = _cur_argidx;

   405

	uint nw;

   406

   407

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

   408

		buf++;

   409

   410

		// This is a {G=DER} command

   411

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

   412

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

   413

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

   414

		}

   415

		// now nw contains the gender index

   416

		PutUtf8(SCC_GENDER_INDEX);

   417

		PutByte(nw);

   418

	} else {

   419

		const char* words[8];

   420

   421

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

   422

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

   423

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

   424

   425

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

   426

			words[nw] = ParseWord(&buf);

   427

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

   428

		}

   429

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

   430

		PutUtf8(SCC_GENDER_LIST);

   431

		PutByte(TranslateArgumentIdx(argidx));

   432

		EmitWordList(words, nw);

   433

	}

   434

}

   435

   436

   437

static const CmdStruct _cmd_structs[] = {

   438

	// Update position

   439

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

   440

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

   441

   442

	// Font size

   443

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

   444

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

   445

   446

	// Colors

   447

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

   448

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

   449

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

   450

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

   451

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

   452

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

   453

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

   454

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

   455

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

   456

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

   457

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

   458

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

   459

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

   460

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

   461

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

   462

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

   463

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

   464

   465

	{"CURRCOMPACT",   EmitSingleChar, SCC_CURRENCY_COMPACT,    1, 0}, // compact currency

   466

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

   467

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

   468

   469

	{"STRING1", EmitSingleChar, SCC_STRING1, 2, C_CASE}, // included string that consumes the string id and ONE argument

   470

	{"STRING2", EmitSingleChar, SCC_STRING2, 3, C_CASE}, // included string that consumes the string id and TWO arguments

   471

	{"STRING3", EmitSingleChar, SCC_STRING3, 4, C_CASE}, // included string that consumes the string id and THREE arguments

   472

	{"STRING4", EmitSingleChar, SCC_STRING4, 5, C_CASE}, // included string that consumes the string id and FOUR arguments

   473

	{"STRING5", EmitSingleChar, SCC_STRING5, 6, C_CASE}, // included string that consumes the string id and FIVE arguments

   474

   475

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

   476

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

   477

	{"CARGO",           EmitSingleChar, SCC_CARGO,            2, 0},

   478

	{"POWER",           EmitSingleChar, SCC_POWER,            1, 0},

   479

	{"VOLUME",          EmitSingleChar, SCC_VOLUME,           1, 0},

   480

	{"VOLUME_S",        EmitSingleChar, SCC_VOLUME_SHORT,     1, 0},

   481

	{"WEIGHT",          EmitSingleChar, SCC_WEIGHT,           1, 0},

   482

	{"WEIGHT_S",        EmitSingleChar, SCC_WEIGHT_SHORT,     1, 0},

   483

	{"FORCE",           EmitSingleChar, SCC_FORCE,            1, 0},

   484

	{"VELOCITY",        EmitSingleChar, SCC_VELOCITY,         1, 0},