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 "../table/control_codes.h"

    10

    11

#include <stdio.h>

    12

#include <string.h>

    13

#include <stdlib.h>

    14

#include <stdarg.h>

    15

    16

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

    17

#include <unistd.h>

    18

#include <sys/types.h>

    19

#include <sys/stat.h>

    20

#endif

    21

    22

#if defined WIN32 || defined __WATCOMC__

    23

#include <direct.h>

    24

#endif /* WIN32 || __WATCOMC__ */

    25

    26

#ifdef __MORPHOS__

    27

#ifdef stderr

    28

#undef stderr

    29

#endif

    30

#define stderr stdout

    31

#endif /* __MORPHOS__ */

    32

    33

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

    34

    35

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

    36

    37

struct LanguagePackHeader {

    38

	uint32 ident;

    39

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

    40

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

    41

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

    42

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

    43

	uint16 offsets[32]; // the offsets

    44

	byte plural_form;   // plural form index

    45

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

    46

};

    47

    48

struct CmdStruct {

    49

	const char *cmd;

    50

	ParseCmdProc proc;

    51

	long value;

    52

	int8 consumes;

    53

	byte flags;

    54

};

    55

    56

enum {

    57

	C_DONTCOUNT = 1,

    58

	C_CASE      = 2,

    59

};

    60

    61

    62

struct Case {

    63

	int caseidx;

    64

	char *string;

    65

	Case *next;

    66

};

    67

    68

static bool _masterlang;

    69

static bool _translated;

    70

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

    71

static int _cur_line;

    72

static int _errors, _warnings, _show_todo;

    73

    74

struct LangString {

    75

	char *name;            // Name of the string

    76

	char *english;         // English text

    77

	char *translated;      // Translated text

    78

	uint16 hash_next;      // next hash entry

    79

	uint16 index;

    80

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

    81

	Case *english_case;    // cases for english

    82

	Case *translated_case; // cases for foreign

    83

};

    84

    85

static LangString *_strings[65536];

    86

    87

    88

#define HASH_SIZE 32767

    89

static uint16 _hash_head[HASH_SIZE];

    90

    91

static byte _put_buf[4096];

    92

static int _put_pos;

    93

static int _next_string_id;

    94

    95

static uint32 _hash;

    96

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

    97

static byte _lang_pluralform;

    98

#define MAX_NUM_GENDER 8

    99

static char _genders[MAX_NUM_GENDER][8];

   100

static int _numgenders;

   101

   102

// contains the name of all cases.

   103

#define MAX_NUM_CASES 50

   104

static char _cases[MAX_NUM_CASES][16];

   105

static int _numcases;

   106

   107

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

   108

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

   109

   110

static const char *_cur_ident;

   111

   112

struct CmdPair {

   113

	const CmdStruct *a;

   114

	const char *v;

   115

};

   116

   117

struct ParsedCommandStruct {

   118

	int np;

   119

	CmdPair pairs[32];

   120

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

   121

};

   122

   123

// Used when generating some advanced commands.

   124

static ParsedCommandStruct _cur_pcs;

   125

static int _cur_argidx;

   126

   127

static uint HashStr(const char *s)

   128

{

   129

	uint hash = 0;

   130

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

   131

	return hash % HASH_SIZE;

   132

}

   133

   134

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

   135

{

   136

	uint hash = HashStr(s);

   137

	ls->hash_next = _hash_head[hash];

   138

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

   139

}

   140

   141

static LangString *HashFind(const char *s)

   142

{

   143

	int idx = _hash_head[HashStr(s)];

   144

   145

	while (--idx >= 0) {

   146

		LangString* ls = _strings[idx];

   147

   148

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

   149

		idx = ls->hash_next;

   150

	}

   151

	return NULL;

   152

}

   153

   154

#ifdef _MSC_VER

   155

# define LINE_NUM_FMT "(%d)"

   156

#else

   157

# define LINE_NUM_FMT ":%d"

   158

#endif

   159

   160

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

   161

{

   162

	char buf[1024];

   163

	va_list va;

   164

	va_start(va, s);

   165

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

   166

	va_end(va);

   167

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

   168

	_warnings++;

   169

}

   170

   171

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

   172

{

   173

	char buf[1024];

   174

	va_list va;

   175

	va_start(va, s);

   176

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

   177

	va_end(va);

   178

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

   179

	_errors++;

   180

}

   181

   182

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

   183

{

   184

	char buf[1024];

   185

	va_list va;

   186

	va_start(va, s);

   187

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

   188

	va_end(va);

   189

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

   190

	exit(1);

   191

}

   192

   193

static void PutByte(byte c)

   194

{

   195

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

   196

	_put_buf[_put_pos++] = c;

   197

}

   198

   199

   200

static void PutUtf8(uint32 value)

   201

{

   202

	if (value < 0x80) {

   203

		PutByte(value);

   204

	} else if (value < 0x800) {

   205

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

   206

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

   207

	} else if (value < 0x10000) {

   208

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

   209

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

   210

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

   211

	} else if (value < 0x110000) {

   212

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

   213

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

   214

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

   215

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

   216

	} else {

   217

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

   218

	}

   219

}

   220

   221

   222

size_t Utf8Validate(const char *s)

   223

{

   224

	uint32 c;

   225

   226

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

   227

		/* 1 byte */

   228

		return 1;

   229

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

   230

		/* 2 bytes */

   231

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

   232

		if (c >= 0x80) return 2;

   233

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

   234

		/* 3 bytes */

   235

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

   236

		if (c >= 0x800) return 3;

   237

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

   238

		/* 4 bytes */

   239

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

   240

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

   241

	}

   242

   243

	return 0;

   244

}

   245

   246

   247

static void EmitSingleChar(char *buf, int value)

   248

{

   249

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

   250

	PutUtf8(value);

   251

}

   252

   253

   254

static void EmitSetX(char *buf, int value)

   255

{

   256

	char *err;

   257

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

   258

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

   259

	PutUtf8(SCC_SETX);

   260

	PutByte((byte)x);

   261

}

   262

   263

   264

static void EmitSetXY(char *buf, int value)

   265

{

   266

	char *err;

   267

	int x;

   268

	int y;

   269

   270

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

   271

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

   272

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

   273

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

   274

   275

	PutUtf8(SCC_SETXY);

   276

	PutByte((byte)x);

   277

	PutByte((byte)y);

   278

}

   279

   280

// The plural specifier looks like

   281

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

   282

   283

// This is encoded like

   284

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

   285

   286

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

   287

{

   288

	const char* s = *buf;

   289

	char* end;

   290

	bool rel = false;

   291

	int v;

   292

   293

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

   294

	if (*s == '+') {

   295

		rel = true;

   296

		s++;

   297

	}

   298

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

   299

	if (end == s) return false;

   300

	if (rel || v < 0) {

   301

		*value += v;

   302

	} else {

   303

		*value = v;

   304

	}

   305

	*buf = end;

   306

	return true;

   307

}

   308

   309

// Parse out the next word, or NULL

   310

char *ParseWord(char **buf)

   311

{

   312

	char *s = *buf, *r;

   313

   314

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

   315

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

   316

   317

	if (*s == '"') {

   318

		r = ++s;

   319

		// parse until next " or NUL

   320

		for (;;) {

   321

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

   322

			if (*s == '"') {

   323

				*s++ = '\0';

   324

				break;

   325

			}

   326

			s++;

   327

		}

   328

	} else {

   329

		// proceed until whitespace or NUL

   330

		r = s;

   331

		for (;;) {

   332

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

   333

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

   334

				*s++ = '\0';

   335

				break;

   336

			}

   337

			s++;

   338

		}

   339

	}

   340

	*buf = s;

   341

	return r;

   342

}

   343

   344

// Forward declaration

   345

static int TranslateArgumentIdx(int arg);

   346

   347

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

   348

{

   349

	uint i;

   350

	uint j;

   351

   352

	PutByte(nw);

   353

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

   354

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

   355

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

   356

	}

   357

}

   358

   359

static void EmitPlural(char *buf, int value)

   360

{

   361

	int argidx = _cur_argidx;

   362

	const char* words[5];

   363

	int nw = 0;

   364

   365

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

   366

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

   367

   368

	// Parse each string

   369

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

   370

		words[nw] = ParseWord(&buf);

   371

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

   372

	}

   373

   374

	if (nw == 0)

   375

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

   376

   377

	if (_plural_form_counts[_lang_pluralform] != nw) {

   378

		if (_translated) {

   379

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

   380

				_plural_form_counts[_lang_pluralform], nw);

   381

		} else {

   382

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

   383

			if (nw > _plural_form_counts[_lang_pluralform]) {

   384

				nw = _plural_form_counts[_lang_pluralform];

   385

			} else {

   386

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

   387

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

   388

				}

   389

			}

   390

		}

   391

	}

   392

   393

	PutUtf8(SCC_PLURAL_LIST);

   394

	PutByte(TranslateArgumentIdx(argidx));

   395

	EmitWordList(words, nw);

   396

}

   397

   398

   399

static void EmitGender(char *buf, int value)

   400

{

   401

	int argidx = _cur_argidx;

   402

	uint nw;

   403

   404

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

   405

		buf++;

   406

   407

		// This is a {G=DER} command

   408

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

   409

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

   410

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

   411

		}

   412

		// now nw contains the gender index

   413

		PutUtf8(SCC_GENDER_INDEX);

   414

		PutByte(nw);

   415

	} else {

   416

		const char* words[8];

   417

   418

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

   419

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

   420

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

   421

   422

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

   423

			words[nw] = ParseWord(&buf);

   424

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

   425

		}

   426

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

   427

		PutUtf8(SCC_GENDER_LIST);

   428

		PutByte(TranslateArgumentIdx(argidx));

   429

		EmitWordList(words, nw);

   430

	}

   431

}

   432

   433

   434

static const CmdStruct _cmd_structs[] = {

   435

	// Update position

   436

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

   437

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

   438

   439

	// Font size

   440

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

   441

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

   442

   443

	// Colors

   444

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

   445

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

   446

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

   447

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

   448

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

   449

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

   450

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

   451

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

   452

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

   453

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

   454

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

   455

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

   456

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

   457

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

   458

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

   459

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

   460

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

   461

   462

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

   463

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

   464

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

   465

   466

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

   467

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

   468

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

   469

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

   470

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

   471

   472

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

   473

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

   474

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

   475

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

   476

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

   477

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

   478

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

   479

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

   480

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

   481

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

   482

   483

	{"P", EmitPlural, 0, 0, C_DONTCOUNT}, // plural specifier

   484

	{"G", EmitGender, 0, 0, C_DONTCOUNT}, // gender specifier

   485