Update to 14.0-beta1

This commit is contained in:
dP
2024-02-04 02:18:17 +05:30
parent 79037e2c65
commit 33ef333b57
1325 changed files with 138461 additions and 70983 deletions
+1
View File
@@ -10,6 +10,7 @@ if (NOT HOST_BINARY_DIR)
strgen_base.cpp
../core/alloc_func.cpp
../misc/getoptdata.cpp
../error.cpp
../string.cpp
)
add_definitions(-DSTRGEN)
+124 -240
View File
@@ -9,24 +9,18 @@
#include "../stdafx.h"
#include "../core/endian_func.hpp"
#include "../core/mem_func.hpp"
#include "../error_func.h"
#include "../string_func.h"
#include "../strings_type.h"
#include "../misc/getoptdata.h"
#include "../table/control_codes.h"
#include "../3rdparty/fmt/std.h"
#include "strgen.h"
#include <stdarg.h>
#include <exception>
#if !defined(_WIN32) || defined(__CYGWIN__)
#include <unistd.h>
#include <sys/stat.h>
#endif
#if defined(_WIN32) || defined(__WATCOMC__)
#include <direct.h>
#endif /* _WIN32 || __WATCOMC__ */
#include <filesystem>
#include <fstream>
#include "../table/strgen_tables.h"
@@ -34,68 +28,48 @@
#ifdef _MSC_VER
# define LINE_NUM_FMT(s) "%s (%d): warning: %s (" s ")\n"
# define LINE_NUM_FMT(s) "{} ({}): warning: {} (" s ")\n"
#else
# define LINE_NUM_FMT(s) "%s:%d: " s ": %s\n"
# define LINE_NUM_FMT(s) "{}:{}: " s ": {}\n"
#endif
void CDECL strgen_warning(const char *s, ...)
void StrgenWarningI(const std::string &msg)
{
char buf[1024];
va_list va;
va_start(va, s);
vseprintf(buf, lastof(buf), s, va);
va_end(va);
if (_show_todo > 0) {
fprintf(stderr, LINE_NUM_FMT("warning"), _file, _cur_line, buf);
fmt::print(stderr, LINE_NUM_FMT("warning"), _file, _cur_line, msg);
} else {
fprintf(stderr, LINE_NUM_FMT("info"), _file, _cur_line, buf);
fmt::print(stderr, LINE_NUM_FMT("info"), _file, _cur_line, msg);
}
_warnings++;
}
void CDECL strgen_error(const char *s, ...)
void StrgenErrorI(const std::string &msg)
{
char buf[1024];
va_list va;
va_start(va, s);
vseprintf(buf, lastof(buf), s, va);
va_end(va);
fprintf(stderr, LINE_NUM_FMT("error"), _file, _cur_line, buf);
fmt::print(stderr, LINE_NUM_FMT("error"), _file, _cur_line, msg);
_errors++;
}
void NORETURN CDECL strgen_fatal(const char *s, ...)
[[noreturn]] void StrgenFatalI(const std::string &msg)
{
char buf[1024];
va_list va;
va_start(va, s);
vseprintf(buf, lastof(buf), s, va);
va_end(va);
fprintf(stderr, LINE_NUM_FMT("FATAL"), _file, _cur_line, buf);
fmt::print(stderr, LINE_NUM_FMT("FATAL"), _file, _cur_line, msg);
#ifdef _MSC_VER
fprintf(stderr, LINE_NUM_FMT("warning"), _file, _cur_line, "language is not compiled");
fmt::print(stderr, LINE_NUM_FMT("warning"), _file, _cur_line, "language is not compiled");
#endif
throw std::exception();
}
void NORETURN CDECL error(const char *s, ...)
[[noreturn]] void FatalErrorI(const std::string &msg)
{
char buf[1024];
va_list va;
va_start(va, s);
vseprintf(buf, lastof(buf), s, va);
va_end(va);
fprintf(stderr, LINE_NUM_FMT("FATAL"), _file, _cur_line, buf);
fmt::print(stderr, LINE_NUM_FMT("FATAL"), _file, _cur_line, msg);
#ifdef _MSC_VER
fprintf(stderr, LINE_NUM_FMT("warning"), _file, _cur_line, "language is not compiled");
fmt::print(stderr, LINE_NUM_FMT("warning"), _file, _cur_line, "language is not compiled");
#endif
exit(2);
}
/** A reader that simply reads using fopen. */
struct FileStringReader : StringReader {
FILE *fh; ///< The file we are reading.
std::ifstream input_stream;
/**
* Create the reader.
@@ -104,22 +78,17 @@ struct FileStringReader : StringReader {
* @param master Are we reading the master file?
* @param translation Are we reading a translation?
*/
FileStringReader(StringData &data, const char *file, bool master, bool translation) :
StringReader(data, file, master, translation)
FileStringReader(StringData &data, const std::filesystem::path &file, bool master, bool translation) :
StringReader(data, file.generic_string(), master, translation)
{
this->fh = fopen(file, "rb");
if (this->fh == nullptr) error("Could not open %s", file);
this->input_stream.open(file, std::ifstream::binary);
}
/** Free/close the file. */
virtual ~FileStringReader()
std::optional<std::string> ReadLine() override
{
fclose(this->fh);
}
char *ReadLine(char *buffer, const char *last) override
{
return fgets(buffer, ClampToU16(last - buffer + 1), this->fh);
std::string result;
if (!std::getline(this->input_stream, result)) return std::nullopt;
return result;
}
void HandlePragma(char *str) override;
@@ -129,7 +98,7 @@ struct FileStringReader : StringReader {
this->StringReader::ParseFile();
if (StrEmpty(_lang.name) || StrEmpty(_lang.own_name) || StrEmpty(_lang.isocode)) {
error("Language must include ##name, ##ownname and ##isocode");
FatalError("Language must include ##name, ##ownname and ##isocode");
}
}
};
@@ -137,7 +106,7 @@ struct FileStringReader : StringReader {
void FileStringReader::HandlePragma(char *str)
{
if (!memcmp(str, "id ", 3)) {
this->data.next_string_id = strtoul(str + 3, nullptr, 0);
this->data.next_string_id = std::strtoul(str + 3, nullptr, 0);
} else if (!memcmp(str, "name ", 5)) {
strecpy(_lang.name, str + 5, lastof(_lang.name));
} else if (!memcmp(str, "ownname ", 8)) {
@@ -150,7 +119,7 @@ void FileStringReader::HandlePragma(char *str)
} else if (!memcmp(str + 8, "rtl", 3)) {
_lang.text_dir = TD_RTL;
} else {
error("Invalid textdir %s", str + 8);
FatalError("Invalid textdir {}", str + 8);
}
} else if (!memcmp(str, "digitsep ", 9)) {
str += 9;
@@ -163,39 +132,39 @@ void FileStringReader::HandlePragma(char *str)
strecpy(_lang.digit_decimal_separator, strcmp(str, "{NBSP}") == 0 ? NBSP : str, lastof(_lang.digit_decimal_separator));
} else if (!memcmp(str, "winlangid ", 10)) {
const char *buf = str + 10;
long langid = strtol(buf, nullptr, 16);
long langid = std::strtol(buf, nullptr, 16);
if (langid > (long)UINT16_MAX || langid < 0) {
error("Invalid winlangid %s", buf);
FatalError("Invalid winlangid {}", buf);
}
_lang.winlangid = (uint16)langid;
_lang.winlangid = (uint16_t)langid;
} else if (!memcmp(str, "grflangid ", 10)) {
const char *buf = str + 10;
long langid = strtol(buf, nullptr, 16);
long langid = std::strtol(buf, nullptr, 16);
if (langid >= 0x7F || langid < 0) {
error("Invalid grflangid %s", buf);
FatalError("Invalid grflangid {}", buf);
}
_lang.newgrflangid = (uint8)langid;
_lang.newgrflangid = (uint8_t)langid;
} else if (!memcmp(str, "gender ", 7)) {
if (this->master) error("Genders are not allowed in the base translation.");
if (this->master) FatalError("Genders are not allowed in the base translation.");
char *buf = str + 7;
for (;;) {
const char *s = ParseWord(&buf);
if (s == nullptr) break;
if (_lang.num_genders >= MAX_NUM_GENDERS) error("Too many genders, max %d", MAX_NUM_GENDERS);
if (_lang.num_genders >= MAX_NUM_GENDERS) FatalError("Too many genders, max {}", MAX_NUM_GENDERS);
strecpy(_lang.genders[_lang.num_genders], s, lastof(_lang.genders[_lang.num_genders]));
_lang.num_genders++;
}
} else if (!memcmp(str, "case ", 5)) {
if (this->master) error("Cases are not allowed in the base translation.");
if (this->master) FatalError("Cases are not allowed in the base translation.");
char *buf = str + 5;
for (;;) {
const char *s = ParseWord(&buf);
if (s == nullptr) break;
if (_lang.num_cases >= MAX_NUM_CASES) error("Too many cases, max %d", MAX_NUM_CASES);
if (_lang.num_cases >= MAX_NUM_CASES) FatalError("Too many cases, max {}", MAX_NUM_CASES);
strecpy(_lang.cases[_lang.num_cases], s, lastof(_lang.cases[_lang.num_cases]));
_lang.num_cases++;
}
@@ -204,108 +173,80 @@ void FileStringReader::HandlePragma(char *str)
}
}
bool CompareFiles(const char *n1, const char *n2)
bool CompareFiles(const std::filesystem::path &path1, const std::filesystem::path &path2)
{
FILE *f2 = fopen(n2, "rb");
if (f2 == nullptr) return false;
/* Check for equal size, but ignore the error code for cases when a file does not exist. */
std::error_code error_code;
if (std::filesystem::file_size(path1, error_code) != std::filesystem::file_size(path2, error_code)) return false;
FILE *f1 = fopen(n1, "rb");
if (f1 == nullptr) {
fclose(f2);
error("can't open %s", n1);
}
std::ifstream stream1(path1, std::ifstream::binary);
std::ifstream stream2(path2, std::ifstream::binary);
size_t l1, l2;
do {
char b1[4096];
char b2[4096];
l1 = fread(b1, 1, sizeof(b1), f1);
l2 = fread(b2, 1, sizeof(b2), f2);
if (l1 != l2 || memcmp(b1, b2, l1)) {
fclose(f2);
fclose(f1);
return false;
}
} while (l1 != 0);
fclose(f2);
fclose(f1);
return true;
return std::equal(std::istreambuf_iterator<char>(stream1.rdbuf()),
std::istreambuf_iterator<char>(),
std::istreambuf_iterator<char>(stream2.rdbuf()));
}
/** Base class for writing data to disk. */
struct FileWriter {
FILE *fh; ///< The file handle we're writing to.
const char *filename; ///< The file name we're writing to.
std::ofstream output_stream; ///< The stream to write all the output to.
const std::filesystem::path path; ///< The file name we're writing to.
/**
* Open a file to write to.
* @param filename The file to open.
* @param path The path to the file to open.
* @param openmode The openmode flags for opening the file.
*/
FileWriter(const char *filename)
FileWriter(const std::filesystem::path &path, std::ios_base::openmode openmode) : path(path)
{
this->filename = stredup(filename);
this->fh = fopen(this->filename, "wb");
if (this->fh == nullptr) {
error("Could not open %s", this->filename);
}
this->output_stream.open(path, openmode);
}
/** Finalise the writing. */
void Finalise()
{
fclose(this->fh);
this->fh = nullptr;
this->output_stream.close();
}
/** Make sure the file is closed. */
virtual ~FileWriter()
{
/* If we weren't closed an exception was thrown, so remove the temporary file. */
if (fh != nullptr) {
fclose(this->fh);
unlink(this->filename);
if (this->output_stream.is_open()) {
this->output_stream.close();
std::filesystem::remove(this->path);
}
free(this->filename);
}
};
struct HeaderFileWriter : HeaderWriter, FileWriter {
/** The real file name we eventually want to write to. */
const char *real_filename;
/** The real path we eventually want to write to. */
const std::filesystem::path real_path;
/** The previous string ID that was printed. */
int prev;
uint total_strings;
/**
* Open a file to write to.
* @param filename The file to open.
* @param path The path to the file to open.
*/
HeaderFileWriter(const char *filename) : FileWriter("tmp.xxx"),
real_filename(stredup(filename)), prev(0), total_strings(0)
HeaderFileWriter(const std::filesystem::path &path) : FileWriter("tmp.xxx", std::ofstream::out),
real_path(path), prev(0), total_strings(0)
{
fprintf(this->fh, "/* This file is automatically generated. Do not modify */\n\n");
fprintf(this->fh, "#ifndef TABLE_STRINGS_H\n");
fprintf(this->fh, "#define TABLE_STRINGS_H\n");
this->output_stream << "/* This file is automatically generated. Do not modify */\n\n";
this->output_stream << "#ifndef TABLE_STRINGS_H\n";
this->output_stream << "#define TABLE_STRINGS_H\n";
}
/** Free the filename. */
~HeaderFileWriter()
void WriteStringID(const std::string &name, int stringid) override
{
free(real_filename);
}
void WriteStringID(const char *name, int stringid)
{
if (prev + 1 != stringid) fprintf(this->fh, "\n");
fprintf(this->fh, "static const StringID %s = 0x%X;\n", name, stringid);
if (prev + 1 != stringid) this->output_stream << "\n";
fmt::print(this->output_stream, "static const StringID {} = 0x{:X};\n", name, stringid);
prev = stringid;
total_strings++;
}
void Finalise(const StringData &data)
void Finalise(const StringData &data) override
{
/* Find the plural form with the most amount of cases. */
int max_plural_forms = 0;
@@ -313,29 +254,31 @@ struct HeaderFileWriter : HeaderWriter, FileWriter {
max_plural_forms = std::max(max_plural_forms, _plural_forms[i].plural_count);
}
fprintf(this->fh,
fmt::print(this->output_stream,
"\n"
"static const uint LANGUAGE_PACK_VERSION = 0x%X;\n"
"static const uint LANGUAGE_MAX_PLURAL = %u;\n"
"static const uint LANGUAGE_MAX_PLURAL_FORMS = %d;\n"
"static const uint LANGUAGE_TOTAL_STRINGS = %u;\n"
"static const uint LANGUAGE_PACK_VERSION = 0x{:X};\n"
"static const uint LANGUAGE_MAX_PLURAL = {};\n"
"static const uint LANGUAGE_MAX_PLURAL_FORMS = {};\n"
"static const uint LANGUAGE_TOTAL_STRINGS = {};\n"
"\n",
(uint)data.Version(), (uint)lengthof(_plural_forms), max_plural_forms, total_strings
data.Version(), lengthof(_plural_forms), max_plural_forms, total_strings
);
fprintf(this->fh, "#endif /* TABLE_STRINGS_H */\n");
this->output_stream << "#endif /* TABLE_STRINGS_H */\n";
this->FileWriter::Finalise();
if (CompareFiles(this->filename, this->real_filename)) {
std::error_code error_code;
if (CompareFiles(this->path, this->real_path)) {
/* files are equal. tmp.xxx is not needed */
unlink(this->filename);
std::filesystem::remove(this->path, error_code); // Just ignore the error
} else {
/* else rename tmp.xxx into filename */
# if defined(_WIN32)
unlink(this->real_filename);
std::filesystem::remove(this->real_path, error_code); // Just ignore the error, file probably doesn't exist
# endif
if (rename(this->filename, this->real_filename) == -1) error("rename() failed");
std::filesystem::rename(this->path, this->real_path, error_code);
if (error_code) FatalError("rename({}, {}) failed: {}", this->path, this->real_path, error_code.message());
}
}
};
@@ -344,80 +287,31 @@ struct HeaderFileWriter : HeaderWriter, FileWriter {
struct LanguageFileWriter : LanguageWriter, FileWriter {
/**
* Open a file to write to.
* @param filename The file to open.
* @param path The path to the file to open.
*/
LanguageFileWriter(const char *filename) : FileWriter(filename)
LanguageFileWriter(const std::filesystem::path &path) : FileWriter(path, std::ofstream::binary | std::ofstream::out)
{
}
void WriteHeader(const LanguagePackHeader *header)
void WriteHeader(const LanguagePackHeader *header) override
{
this->Write((const byte *)header, sizeof(*header));
}
void Finalise()
void Finalise() override
{
if (fputc(0, this->fh) == EOF) {
error("Could not write to %s", this->filename);
}
this->output_stream.put(0);
this->FileWriter::Finalise();
}
void Write(const byte *buffer, size_t length)
void Write(const byte *buffer, size_t length) override
{
if (fwrite(buffer, sizeof(*buffer), length, this->fh) != length) {
error("Could not write to %s", this->filename);
}
this->output_stream.write((const char *)buffer, length);
}
};
/** Multi-OS mkdirectory function */
static inline void ottd_mkdir(const char *directory)
{
/* Ignore directory creation errors; they'll surface later on, and most
* of the time they are 'directory already exists' errors anyhow. */
#if defined(_WIN32) || defined(__WATCOMC__)
mkdir(directory);
#else
mkdir(directory, 0755);
#endif
}
/**
* Create a path consisting of an already existing path, a possible
* path separator and the filename. The separator is only appended if the path
* does not already end with a separator
*/
static inline char *mkpath(char *buf, const char *last, const char *path, const char *file)
{
strecpy(buf, path, last); // copy directory into buffer
char *p = strchr(buf, '\0'); // add path separator if necessary
if (p[-1] != PATHSEPCHAR && p != last) *p++ = PATHSEPCHAR;
strecpy(p, file, last); // concatenate filename at end of buffer
return buf;
}
#if defined(_WIN32)
/**
* On MingW, it is common that both / as \ are accepted in the
* params. To go with those flow, we rewrite all incoming /
* simply to \, so internally we can safely assume \, and do
* this for all Windows machines to keep identical behaviour,
* no matter what your compiler was.
*/
static inline char *replace_pathsep(char *s)
{
for (char *c = s; *c != '\0'; c++) if (*c == '/') *c = '\\';
return s;
}
#else
static inline char *replace_pathsep(char *s) { return s; }
#endif
/** Options of strgen. */
static const OptionData _opts[] = {
GETOPT_NOVAL( 'v', "--version"),
GETOPT_GENERAL('C', '\0', "-export-commands", ODF_NO_VALUE),
GETOPT_GENERAL('L', '\0', "-export-plurals", ODF_NO_VALUE),
GETOPT_GENERAL('P', '\0', "-export-pragmas", ODF_NO_VALUE),
@@ -432,9 +326,8 @@ static const OptionData _opts[] = {
int CDECL main(int argc, char *argv[])
{
char pathbuf[MAX_PATH];
const char *src_dir = ".";
const char *dest_dir = nullptr;
std::filesystem::path src_dir(".");
std::filesystem::path dest_dir;
GetOptData mgo(argc - 1, argv + 1, _opts);
for (;;) {
@@ -442,12 +335,8 @@ int CDECL main(int argc, char *argv[])
if (i == -1) break;
switch (i) {
case 'v':
puts("$Revision$");
return 0;
case 'C':
printf("args\tflags\tcommand\treplacement\n");
fmt::print("args\tflags\tcommand\treplacement\n");
for (const CmdStruct *cs = _cmd_structs; cs < endof(_cmd_structs); cs++) {
char flags;
if (cs->proc == EmitGender) {
@@ -459,22 +348,22 @@ int CDECL main(int argc, char *argv[])
} else {
flags = '0'; // Command needs no parameters
}
printf("%i\t%c\t\"%s\"\t\"%s\"\n", cs->consumes, flags, cs->cmd, strstr(cs->cmd, "STRING") ? "STRING" : cs->cmd);
fmt::print("{}\t{:c}\t\"{}\"\t\"{}\"\n", cs->consumes, flags, cs->cmd, strstr(cs->cmd, "STRING") ? "STRING" : cs->cmd);
}
return 0;
case 'L':
printf("count\tdescription\tnames\n");
fmt::print("count\tdescription\tnames\n");
for (const PluralForm *pf = _plural_forms; pf < endof(_plural_forms); pf++) {
printf("%i\t\"%s\"\t%s\n", pf->plural_count, pf->description, pf->names);
fmt::print("{}\t\"{}\"\t{}\n", pf->plural_count, pf->description, pf->names);
}
return 0;
case 'P':
printf("name\tflags\tdefault\tdescription\n");
for (size_t i = 0; i < lengthof(_pragmas); i++) {
printf("\"%s\"\t%s\t\"%s\"\t\"%s\"\n",
_pragmas[i][0], _pragmas[i][1], _pragmas[i][2], _pragmas[i][3]);
fmt::print("name\tflags\tdefault\tdescription\n");
for (size_t j = 0; j < lengthof(_pragmas); j++) {
fmt::print("\"{}\"\t{}\t\"{}\"\t\"{}\"\n",
_pragmas[j][0], _pragmas[j][1], _pragmas[j][2], _pragmas[j][3]);
}
return 0;
@@ -487,9 +376,8 @@ int CDECL main(int argc, char *argv[])
break;
case 'h':
puts(
"strgen - $Revision$\n"
" -v | --version print version information and exit\n"
fmt::print(
"strgen\n"
" -t | --todo replace any untranslated strings with '<TODO>'\n"
" -w | --warning print a warning for any untranslated strings\n"
" -h | -? | --help print this help message and exit\n"
@@ -500,25 +388,25 @@ int CDECL main(int argc, char *argv[])
" -export-pragmas export all pragmas and exit\n"
" Run without parameters and strgen will search for english.txt and parse it,\n"
" creating strings.h. Passing an argument, strgen will translate that language\n"
" file using english.txt as a reference and output <language>.lng."
" file using english.txt as a reference and output <language>.lng.\n"
);
return 0;
case 's':
src_dir = replace_pathsep(mgo.opt);
src_dir = mgo.opt;
break;
case 'd':
dest_dir = replace_pathsep(mgo.opt);
dest_dir = mgo.opt;
break;
case -2:
fprintf(stderr, "Invalid arguments\n");
fmt::print(stderr, "Invalid arguments\n");
return 0;
}
}
if (dest_dir == nullptr) dest_dir = src_dir; // if dest_dir is not specified, it equals src_dir
if (dest_dir.empty()) dest_dir = src_dir; // if dest_dir is not specified, it equals src_dir
try {
/* strgen has two modes of operation. If no (free) arguments are passed
@@ -526,57 +414,53 @@ int CDECL main(int argc, char *argv[])
* with a (free) parameter the program will translate that language to destination
* directory. As input english.txt is parsed from the source directory */
if (mgo.numleft == 0) {
mkpath(pathbuf, lastof(pathbuf), src_dir, "english.txt");
std::filesystem::path input_path = src_dir;
input_path /= "english.txt";
/* parse master file */
StringData data(TEXT_TAB_END);
FileStringReader master_reader(data, pathbuf, true, false);
FileStringReader master_reader(data, input_path, true, false);
master_reader.ParseFile();
if (_errors != 0) return 1;
/* write strings.h */
ottd_mkdir(dest_dir);
mkpath(pathbuf, lastof(pathbuf), dest_dir, "strings.h");
std::filesystem::path output_path = dest_dir;
std::filesystem::create_directories(dest_dir);
output_path /= "strings.h";
HeaderFileWriter writer(pathbuf);
HeaderFileWriter writer(output_path);
writer.WriteHeader(data);
writer.Finalise(data);
if (_errors != 0) return 1;
} else if (mgo.numleft >= 1) {
char *r;
mkpath(pathbuf, lastof(pathbuf), src_dir, "english.txt");
std::filesystem::path input_path = src_dir;
input_path /= "english.txt";
StringData data(TEXT_TAB_END);
/* parse master file and check if target file is correct */
FileStringReader master_reader(data, pathbuf, true, false);
FileStringReader master_reader(data, input_path, true, false);
master_reader.ParseFile();
for (int i = 0; i < mgo.numleft; i++) {
data.FreeTranslation();
const char *translation = replace_pathsep(mgo.argv[i]);
const char *file = strrchr(translation, PATHSEPCHAR);
FileStringReader translation_reader(data, translation, false, file == nullptr || strcmp(file + 1, "english.txt") != 0);
std::filesystem::path lang_file = mgo.argv[i];
FileStringReader translation_reader(data, lang_file, false, lang_file.filename() != "english.txt");
translation_reader.ParseFile(); // target file
if (_errors != 0) return 1;
/* get the targetfile, strip any directories and append to destination path */
r = strrchr(mgo.argv[i], PATHSEPCHAR);
mkpath(pathbuf, lastof(pathbuf), dest_dir, (r != nullptr) ? &r[1] : mgo.argv[i]);
std::filesystem::path output_file = dest_dir;
output_file /= lang_file.filename();
output_file.replace_extension("lng");
/* rename the .txt (input-extension) to .lng */
r = strrchr(pathbuf, '.');
if (r == nullptr || strcmp(r, ".txt") != 0) r = strchr(pathbuf, '\0');
strecpy(r, ".lng", lastof(pathbuf));
LanguageFileWriter writer(pathbuf);
LanguageFileWriter writer(output_file);
writer.WriteLang(data);
writer.Finalise();
/* if showing warnings, print a summary of the language */
if ((_show_todo & 2) != 0) {
fprintf(stdout, "%d warnings and %d errors for %s\n", _warnings, _errors, pathbuf);
fmt::print("{} warnings and {} errors for {}\n", _warnings, _errors, output_file);
}
}
}
+47 -33
View File
@@ -11,46 +11,44 @@
#define STRGEN_H
#include "../language.h"
#include "../3rdparty/fmt/format.h"
#include <unordered_map>
#include <array>
/** Container for the different cases of a string. */
struct Case {
int caseidx; ///< The index of the case.
char *string; ///< The translation of the case.
Case *next; ///< The next, chained, case.
int caseidx; ///< The index of the case.
std::string string; ///< The translation of the case.
Case(int caseidx, const char *string, Case *next);
~Case();
Case(int caseidx, const std::string &string);
};
/** Information about a single string. */
struct LangString {
char *name; ///< Name of the string.
char *english; ///< English text.
char *translated; ///< Translated text.
size_t hash_next; ///< Next hash entry.
size_t index; ///< The index in the language file.
int line; ///< Line of string in source-file.
Case *translated_case; ///< Cases of the translation.
std::string name; ///< Name of the string.
std::string english; ///< English text.
std::string translated; ///< Translated text.
size_t index; ///< The index in the language file.
int line; ///< Line of string in source-file.
std::vector<Case> translated_cases; ///< Cases of the translation.
LangString(const char *name, const char *english, size_t index, int line);
~LangString();
LangString(const std::string &name, const std::string &english, size_t index, int line);
void FreeTranslation();
};
/** Information about the currently known strings. */
struct StringData {
LangString **strings; ///< Array of all known strings.
size_t *hash_heads; ///< Hash table for the strings.
std::vector<std::unique_ptr<LangString>> strings; ///< List of all known strings.
std::unordered_map<std::string_view, LangString *> name_to_string; ///< Lookup table for the strings.
size_t tabs; ///< The number of 'tabs' of strings.
size_t max_strings; ///< The maximum number of strings.
size_t next_string_id;///< The next string ID to allocate.
StringData(size_t tabs);
~StringData();
void FreeTranslation();
uint HashStr(const char *s) const;
void Add(const char *s, LangString *ls);
LangString *Find(const char *s);
void Add(std::unique_ptr<LangString> ls);
LangString *Find(const std::string_view s);
uint VersionHashStr(uint hash, const char *s) const;
uint Version() const;
uint CountInUse(uint tab) const;
@@ -59,21 +57,19 @@ struct StringData {
/** Helper for reading strings. */
struct StringReader {
StringData &data; ///< The data to fill during reading.
const char *file; ///< The file we are reading.
const std::string file; ///< The file we are reading.
bool master; ///< Are we reading the master file?
bool translation; ///< Are we reading a translation, implies !master. However, the base translation will have this false.
StringReader(StringData &data, const char *file, bool master, bool translation);
virtual ~StringReader();
StringReader(StringData &data, const std::string &file, bool master, bool translation);
virtual ~StringReader() {}
void HandleString(char *str);
/**
* Read a single line from the source of strings.
* @param buffer The buffer to read the data in to.
* @param last The last element in the buffer.
* @return The buffer, or nullptr if at the end of the file.
* @return The line, or std::nullopt if at the end of the file.
*/
virtual char *ReadLine(char *buffer, const char *last) = 0;
virtual std::optional<std::string> ReadLine() = 0;
/**
* Handle the pragma of the file.
@@ -94,7 +90,7 @@ struct HeaderWriter {
* @param name The name of the string.
* @param stringid The ID of the string.
*/
virtual void WriteStringID(const char *name, int stringid) = 0;
virtual void WriteStringID(const std::string &name, int stringid) = 0;
/**
* Finalise writing the file.
@@ -103,7 +99,7 @@ struct HeaderWriter {
virtual void Finalise(const StringData &data) = 0;
/** Especially destroy the subclasses. */
virtual ~HeaderWriter() {};
virtual ~HeaderWriter() = default;
void WriteHeader(const StringData &data);
};
@@ -130,15 +126,33 @@ struct LanguageWriter {
virtual void Finalise() = 0;
/** Especially destroy the subclasses. */
virtual ~LanguageWriter() {}
virtual ~LanguageWriter() = default;
virtual void WriteLength(uint length);
virtual void WriteLang(const StringData &data);
};
void CDECL strgen_warning(const char *s, ...) WARN_FORMAT(1, 2);
void CDECL strgen_error(const char *s, ...) WARN_FORMAT(1, 2);
void NORETURN CDECL strgen_fatal(const char *s, ...) WARN_FORMAT(1, 2);
struct CmdStruct;
struct CmdPair {
const CmdStruct *cmd;
std::string param;
};
struct ParsedCommandStruct {
std::vector<CmdPair> non_consuming_commands;
std::array<const CmdStruct*, 32> consuming_commands{ nullptr }; // ordered by param #
};
const CmdStruct *TranslateCmdForCompare(const CmdStruct *a);
ParsedCommandStruct ExtractCommandString(const char *s, bool warnings);
void StrgenWarningI(const std::string &msg);
void StrgenErrorI(const std::string &msg);
[[noreturn]] void StrgenFatalI(const std::string &msg);
#define StrgenWarning(format_string, ...) StrgenWarningI(fmt::format(FMT_STRING(format_string), ## __VA_ARGS__))
#define StrgenError(format_string, ...) StrgenErrorI(fmt::format(FMT_STRING(format_string), ## __VA_ARGS__))
#define StrgenFatal(format_string, ...) StrgenFatalI(fmt::format(FMT_STRING(format_string), ## __VA_ARGS__))
char *ParseWord(char **buf);
extern const char *_file;
+135 -211
View File
@@ -8,7 +8,10 @@
/** @file strgen_base.cpp Tool to create computer readable (stand-alone) translation files. */
#include "../stdafx.h"
#include "../core/alloc_func.hpp"
#include "../core/endian_func.hpp"
#include "../core/mem_func.hpp"
#include "../error_func.h"
#include "../string_func.h"
#include "../table/control_codes.h"
@@ -35,20 +38,12 @@ static const CmdStruct *ParseCommandString(const char **str, char *param, int *a
* Create a new case.
* @param caseidx The index of the case.
* @param string The translation of the case.
* @param next The next chained case.
*/
Case::Case(int caseidx, const char *string, Case *next) :
caseidx(caseidx), string(stredup(string)), next(next)
Case::Case(int caseidx, const std::string &string) :
caseidx(caseidx), string(string)
{
}
/** Free everything we allocated. */
Case::~Case()
{
free(this->string);
delete this->next;
}
/**
* Create a new string.
* @param name The name of the string.
@@ -56,29 +51,16 @@ Case::~Case()
* @param index The index in the string table.
* @param line The line this string was found on.
*/
LangString::LangString(const char *name, const char *english, size_t index, int line) :
name(stredup(name)), english(stredup(english)), translated(nullptr),
hash_next(0), index(index), line(line), translated_case(nullptr)
LangString::LangString(const std::string &name, const std::string &english, size_t index, int line) :
name(name), english(english), index(index), line(line)
{
}
/** Free everything we allocated. */
LangString::~LangString()
{
free(this->name);
free(this->english);
free(this->translated);
delete this->translated_case;
}
/** Free all data related to the translation. */
void LangString::FreeTranslation()
{
free(this->translated);
this->translated = nullptr;
delete this->translated_case;
this->translated_case = nullptr;
this->translated.clear();
this->translated_cases.clear();
}
/**
@@ -87,52 +69,28 @@ void LangString::FreeTranslation()
*/
StringData::StringData(size_t tabs) : tabs(tabs), max_strings(tabs * TAB_SIZE)
{
this->strings = CallocT<LangString *>(max_strings);
this->hash_heads = CallocT<size_t>(max_strings);
this->strings.resize(max_strings);
this->next_string_id = 0;
}
/** Free everything we allocated. */
StringData::~StringData()
{
for (size_t i = 0; i < this->max_strings; i++) delete this->strings[i];
free(this->strings);
free(this->hash_heads);
}
/** Free all data related to the translation. */
void StringData::FreeTranslation()
{
for (size_t i = 0; i < this->max_strings; i++) {
LangString *ls = this->strings[i];
LangString *ls = this->strings[i].get();
if (ls != nullptr) ls->FreeTranslation();
}
}
/**
* Create a hash of the string for finding them back quickly.
* @param s The string to hash.
* @return The hashed string.
*/
uint StringData::HashStr(const char *s) const
{
uint hash = 0;
for (; *s != '\0'; s++) hash = ROL(hash, 3) ^ *s;
return hash % this->max_strings;
}
/**
* Add a newly created LangString.
* @param s The name of the string.
* @param ls The string to add.
*/
void StringData::Add(const char *s, LangString *ls)
void StringData::Add(std::unique_ptr<LangString> ls)
{
uint hash = this->HashStr(s);
ls->hash_next = this->hash_heads[hash];
/* Off-by-one for hash find. */
this->hash_heads[hash] = ls->index + 1;
this->strings[ls->index] = ls;
this->name_to_string[ls->name] = ls.get();
this->strings[ls->index].swap(ls);
}
/**
@@ -140,17 +98,12 @@ void StringData::Add(const char *s, LangString *ls)
* @param s The string name to search on.
* @return The LangString or nullptr if it is not known.
*/
LangString *StringData::Find(const char *s)
LangString *StringData::Find(const std::string_view s)
{
size_t idx = this->hash_heads[this->HashStr(s)];
auto it = this->name_to_string.find(s);
if (it == this->name_to_string.end()) return nullptr;
while (idx-- > 0) {
LangString *ls = this->strings[idx];
if (strcmp(ls->name, s) == 0) return ls;
idx = ls->hash_next;
}
return nullptr;
return it->second;
}
/**
@@ -162,7 +115,7 @@ LangString *StringData::Find(const char *s)
uint StringData::VersionHashStr(uint hash, const char *s) const
{
for (; *s != '\0'; s++) {
hash = ROL(hash, 3) ^ *s;
hash = std::rotl(hash, 3) ^ *s;
hash = (hash & 1 ? hash >> 1 ^ 0xDEADBEEF : hash >> 1);
}
return hash;
@@ -177,7 +130,7 @@ uint StringData::Version() const
uint hash = 0;
for (size_t i = 0; i < this->max_strings; i++) {
const LangString *ls = this->strings[i];
const LangString *ls = this->strings[i].get();
if (ls != nullptr) {
const CmdStruct *cs;
@@ -186,12 +139,12 @@ uint StringData::Version() const
int argno;
int casei;
s = ls->name;
s = ls->name.c_str();
hash ^= i * 0x717239;
hash = (hash & 1 ? hash >> 1 ^ 0xDEADBEEF : hash >> 1);
hash = this->VersionHashStr(hash, s + 1);
s = ls->english;
s = ls->english.c_str();
while ((cs = ParseCommandString(&s, buf, &argno, &casei)) != nullptr) {
if (cs->flags & C_DONTCOUNT) continue;
@@ -217,17 +170,6 @@ uint StringData::CountInUse(uint tab) const
static const char *_cur_ident;
struct CmdPair {
const CmdStruct *a;
const char *v;
};
struct ParsedCommandStruct {
uint np;
CmdPair pairs[32];
const CmdStruct *cmd[32]; // ordered by param #
};
/* Used when generating some advanced commands. */
static ParsedCommandStruct _cur_pcs;
static int _cur_argidx;
@@ -247,7 +189,7 @@ struct Buffer : std::vector<byte> {
* Add an Unicode character encoded in UTF-8 to the buffer.
* @param value The character to add.
*/
void AppendUtf8(uint32 value)
void AppendUtf8(uint32_t value)
{
if (value < 0x80) {
this->push_back(value);
@@ -264,14 +206,14 @@ struct Buffer : std::vector<byte> {
this->push_back(0x80 + GB(value, 6, 6));
this->push_back(0x80 + GB(value, 0, 6));
} else {
strgen_warning("Invalid unicode value U+0x%X", value);
StrgenWarning("Invalid unicode value U+0x{:X}", value);
}
}
};
size_t Utf8Validate(const char *s)
{
uint32 c;
uint32_t c;
if (!HasBit(s[0], 7)) {
/* 1 byte */
@@ -296,7 +238,7 @@ size_t Utf8Validate(const char *s)
void EmitSingleChar(Buffer *buffer, char *buf, int value)
{
if (*buf != '\0') strgen_warning("Ignoring trailing letters in command");
if (*buf != '\0') StrgenWarning("Ignoring trailing letters in command");
buffer->AppendUtf8(value);
}
@@ -318,7 +260,7 @@ bool ParseRelNum(char **buf, int *value, int *offset)
rel = true;
s++;
}
int v = strtol(s, &end, 0);
int v = std::strtol(s, &end, 0);
if (end == s) return false;
if (rel || v < 0) {
*value += v;
@@ -328,7 +270,7 @@ bool ParseRelNum(char **buf, int *value, int *offset)
if (offset != nullptr && *end == ':') {
/* Take the Nth within */
s = end + 1;
*offset = strtol(s, &end, 0);
*offset = std::strtol(s, &end, 0);
if (end == s) return false;
}
*buf = end;
@@ -373,7 +315,7 @@ char *ParseWord(char **buf)
/* Forward declaration */
static int TranslateArgumentIdx(int arg, int offset = 0);
static void EmitWordList(Buffer *buffer, const char * const *words, uint nw)
static void EmitWordList(Buffer *buffer, const std::vector<const char *> &words, uint nw)
{
buffer->AppendByte(nw);
for (uint i = 0; i < nw; i++) buffer->AppendByte((byte)strlen(words[i]) + 1);
@@ -383,22 +325,22 @@ static void EmitWordList(Buffer *buffer, const char * const *words, uint nw)
}
}
void EmitPlural(Buffer *buffer, char *buf, int value)
void EmitPlural(Buffer *buffer, char *buf, int)
{
int argidx = _cur_argidx;
int offset = -1;
int expected = _plural_forms[_lang.plural_form].plural_count;
const char **words = AllocaM(const char *, std::max(expected, MAX_PLURALS));
std::vector<const char *> words(std::max(expected, MAX_PLURALS), nullptr);
int nw = 0;
/* Parse out the number, if one exists. Otherwise default to prev arg. */
if (!ParseRelNum(&buf, &argidx, &offset)) argidx--;
const CmdStruct *cmd = _cur_pcs.cmd[argidx];
const CmdStruct *cmd = _cur_pcs.consuming_commands[argidx];
if (offset == -1) {
/* Use default offset */
if (cmd == nullptr || cmd->default_plural_offset < 0) {
strgen_fatal("Command '%s' has no (default) plural position", cmd == nullptr ? "<empty>" : cmd->cmd);
StrgenFatal("Command '{}' has no (default) plural position", cmd == nullptr ? "<empty>" : cmd->cmd);
}
offset = cmd->default_plural_offset;
}
@@ -410,15 +352,15 @@ void EmitPlural(Buffer *buffer, char *buf, int value)
}
if (nw == 0) {
strgen_fatal("%s: No plural words", _cur_ident);
StrgenFatal("{}: No plural words", _cur_ident);
}
if (expected != nw) {
if (_translated) {
strgen_fatal("%s: Invalid number of plural forms. Expecting %d, found %d.", _cur_ident,
StrgenFatal("{}: Invalid number of plural forms. Expecting {}, found {}.", _cur_ident,
expected, nw);
} else {
if ((_show_todo & 2) != 0) strgen_warning("'%s' is untranslated. Tweaking english string to allow compilation for plural forms", _cur_ident);
if ((_show_todo & 2) != 0) StrgenWarning("'{}' is untranslated. Tweaking english string to allow compilation for plural forms", _cur_ident);
if (nw > expected) {
nw = expected;
} else {
@@ -435,8 +377,7 @@ void EmitPlural(Buffer *buffer, char *buf, int value)
EmitWordList(buffer, words, nw);
}
void EmitGender(Buffer *buffer, char *buf, int value)
void EmitGender(Buffer *buffer, char *buf, int)
{
int argidx = _cur_argidx;
int offset = 0;
@@ -447,28 +388,28 @@ void EmitGender(Buffer *buffer, char *buf, int value)
/* This is a {G=DER} command */
nw = _lang.GetGenderIndex(buf);
if (nw >= MAX_NUM_GENDERS) strgen_fatal("G argument '%s' invalid", buf);
if (nw >= MAX_NUM_GENDERS) StrgenFatal("G argument '{}' invalid", buf);
/* now nw contains the gender index */
buffer->AppendUtf8(SCC_GENDER_INDEX);
buffer->AppendByte(nw);
} else {
const char *words[MAX_NUM_GENDERS];
std::vector<const char *> words(MAX_NUM_GENDERS, nullptr);
/* This is a {G 0 foo bar two} command.
* If no relative number exists, default to +0 */
ParseRelNum(&buf, &argidx, &offset);
const CmdStruct *cmd = _cur_pcs.cmd[argidx];
const CmdStruct *cmd = _cur_pcs.consuming_commands[argidx];
if (cmd == nullptr || (cmd->flags & C_GENDER) == 0) {
strgen_fatal("Command '%s' can't have a gender", cmd == nullptr ? "<empty>" : cmd->cmd);
StrgenFatal("Command '{}' can't have a gender", cmd == nullptr ? "<empty>" : cmd->cmd);
}
for (nw = 0; nw < MAX_NUM_GENDERS; nw++) {
words[nw] = ParseWord(&buf);
if (words[nw] == nullptr) break;
}
if (nw != _lang.num_genders) strgen_fatal("Bad # of arguments for gender command");
if (nw != _lang.num_genders) StrgenFatal("Bad # of arguments for gender command");
assert(IsInsideBS(cmd->value, SCC_CONTROL_START, UINT8_MAX));
buffer->AppendUtf8(SCC_GENDER_LIST);
@@ -493,8 +434,8 @@ static uint ResolveCaseName(const char *str, size_t len)
memcpy(case_str, str, len);
case_str[len] = '\0';
uint8 case_idx = _lang.GetCaseIndex(case_str);
if (case_idx >= MAX_NUM_CASES) strgen_fatal("Invalid case-name '%s'", case_str);
uint8_t case_idx = _lang.GetCaseIndex(case_str);
if (case_idx >= MAX_NUM_CASES) StrgenFatal("Invalid case-name '{}'", case_str);
return case_idx + 1;
}
@@ -518,8 +459,8 @@ static const CmdStruct *ParseCommandString(const char **str, char *param, int *a
if (*s >= '0' && *s <= '9') {
char *end;
*argno = strtoul(s, &end, 0);
if (*end != ':') strgen_fatal("missing arg #");
*argno = std::strtoul(s, &end, 0);
if (*end != ':') StrgenFatal("missing arg #");
s = end + 1;
}
@@ -531,7 +472,8 @@ static const CmdStruct *ParseCommandString(const char **str, char *param, int *a
const CmdStruct *cmd = FindCmd(start, s - start - 1);
if (cmd == nullptr) {
strgen_error("Undefined command '%.*s'", (int)(s - start - 1), start);
std::string command(start, s - start - 1);
StrgenError("Undefined command '{}'", command);
return nullptr;
}
@@ -539,7 +481,7 @@ static const CmdStruct *ParseCommandString(const char **str, char *param, int *a
const char *casep = s;
if (!(cmd->flags & C_CASE)) {
strgen_fatal("Command '%s' can't have a case", cmd->cmd);
StrgenFatal("Command '{}' can't have a case", cmd->cmd);
}
do {
@@ -549,7 +491,7 @@ static const CmdStruct *ParseCommandString(const char **str, char *param, int *a
}
if (c == '\0') {
strgen_error("Missing } from command '%s'", start);
StrgenError("Missing }} from command '{}'", start);
return nullptr;
}
@@ -562,10 +504,10 @@ static const CmdStruct *ParseCommandString(const char **str, char *param, int *a
c = *s++;
if (c == '}') break;
if (c == '\0') {
strgen_error("Missing } from command '%s'", start);
StrgenError("Missing }} from command '{}'", start);
return nullptr;
}
if (s - start == MAX_COMMAND_PARAM_SIZE) error("param command too long");
if (s - start == MAX_COMMAND_PARAM_SIZE) FatalError("param command too long");
*param++ = c;
}
}
@@ -583,25 +525,19 @@ static const CmdStruct *ParseCommandString(const char **str, char *param, int *a
* @param master Are we reading the master file?
* @param translation Are we reading a translation?
*/
StringReader::StringReader(StringData &data, const char *file, bool master, bool translation) :
data(data), file(stredup(file)), master(master), translation(translation)
StringReader::StringReader(StringData &data, const std::string &file, bool master, bool translation) :
data(data), file(file), master(master), translation(translation)
{
}
/** Make sure the right reader gets freed. */
StringReader::~StringReader()
{
free(file);
}
static void ExtractCommandString(ParsedCommandStruct *p, const char *s, bool warnings)
ParsedCommandStruct ExtractCommandString(const char *s, bool)
{
char param[MAX_COMMAND_PARAM_SIZE];
int argno;
int argidx = 0;
int casei;
memset(p, 0, sizeof(*p));
ParsedCommandStruct p;
for (;;) {
/* read until next command from a. */
@@ -610,25 +546,24 @@ static void ExtractCommandString(ParsedCommandStruct *p, const char *s, bool war
if (ar == nullptr) break;
/* Sanity checking */
if (argno != -1 && ar->consumes == 0) strgen_fatal("Non consumer param can't have a paramindex");
if (argno != -1 && ar->consumes == 0) StrgenFatal("Non consumer param can't have a paramindex");
if (ar->consumes) {
if (argno != -1) argidx = argno;
if (argidx < 0 || (uint)argidx >= lengthof(p->cmd)) strgen_fatal("invalid param idx %d", argidx);
if (p->cmd[argidx] != nullptr && p->cmd[argidx] != ar) strgen_fatal("duplicate param idx %d", argidx);
if (argidx < 0 || (uint)argidx >= p.consuming_commands.max_size()) StrgenFatal("invalid param idx {}", argidx);
if (p.consuming_commands[argidx] != nullptr && p.consuming_commands[argidx] != ar) StrgenFatal("duplicate param idx {}", argidx);
p->cmd[argidx++] = ar;
p.consuming_commands[argidx++] = ar;
} else if (!(ar->flags & C_DONTCOUNT)) { // Ignore some of them
if (p->np >= lengthof(p->pairs)) strgen_fatal("too many commands in string, max " PRINTF_SIZE, lengthof(p->pairs));
p->pairs[p->np].a = ar;
p->pairs[p->np].v = param[0] != '\0' ? stredup(param) : "";
p->np++;
p.non_consuming_commands.emplace_back(CmdPair{ar, param});
}
}
return p;
}
static const CmdStruct *TranslateCmdForCompare(const CmdStruct *a)
const CmdStruct *TranslateCmdForCompare(const CmdStruct *a)
{
if (a == nullptr) return nullptr;
@@ -647,7 +582,7 @@ static const CmdStruct *TranslateCmdForCompare(const CmdStruct *a)
}
static bool CheckCommandsMatch(char *a, char *b, const char *name)
static bool CheckCommandsMatch(const char *a, const char *b, const char *name)
{
/* If we're not translating, i.e. we're compiling the base language,
* it is pointless to do all these checks as it'll always be correct.
@@ -655,45 +590,42 @@ static bool CheckCommandsMatch(char *a, char *b, const char *name)
*/
if (!_translation) return true;
ParsedCommandStruct templ;
ParsedCommandStruct lang;
bool result = true;
ExtractCommandString(&templ, b, true);
ExtractCommandString(&lang, a, true);
ParsedCommandStruct templ = ExtractCommandString(b, true);
ParsedCommandStruct lang = ExtractCommandString(a, true);
/* For each string in templ, see if we find it in lang */
if (templ.np != lang.np) {
strgen_warning("%s: template string and language string have a different # of commands", name);
if (templ.non_consuming_commands.max_size() != lang.non_consuming_commands.max_size()) {
StrgenWarning("{}: template string and language string have a different # of commands", name);
result = false;
}
for (uint i = 0; i < templ.np; i++) {
for (auto &templ_nc : templ.non_consuming_commands) {
/* see if we find it in lang, and zero it out */
bool found = false;
for (uint j = 0; j < lang.np; j++) {
if (templ.pairs[i].a == lang.pairs[j].a &&
strcmp(templ.pairs[i].v, lang.pairs[j].v) == 0) {
for (auto &lang_nc : lang.non_consuming_commands) {
if (templ_nc.cmd == lang_nc.cmd && templ_nc.param == lang_nc.param) {
/* it was found in both. zero it out from lang so we don't find it again */
lang.pairs[j].a = nullptr;
lang_nc.cmd = nullptr;
found = true;
break;
}
}
if (!found) {
strgen_warning("%s: command '%s' exists in template file but not in language file", name, templ.pairs[i].a->cmd);
StrgenWarning("{}: command '{}' exists in template file but not in language file", name, templ_nc.cmd->cmd);
result = false;
}
}
/* if we reach here, all non consumer commands match up.
* Check if the non consumer commands match up also. */
for (uint i = 0; i < lengthof(templ.cmd); i++) {
if (TranslateCmdForCompare(templ.cmd[i]) != lang.cmd[i]) {
strgen_warning("%s: Param idx #%d '%s' doesn't match with template command '%s'", name, i,
lang.cmd[i] == nullptr ? "<empty>" : TranslateCmdForCompare(lang.cmd[i])->cmd,
templ.cmd[i] == nullptr ? "<empty>" : templ.cmd[i]->cmd);
for (uint i = 0; i < templ.consuming_commands.max_size(); i++) {
if (TranslateCmdForCompare(templ.consuming_commands[i]) != lang.consuming_commands[i]) {
StrgenWarning("{}: Param idx #{} '{}' doesn't match with template command '{}'", name, i,
lang.consuming_commands[i] == nullptr ? "<empty>" : TranslateCmdForCompare(lang.consuming_commands[i])->cmd,
templ.consuming_commands[i] == nullptr ? "<empty>" : templ.consuming_commands[i]->cmd);
result = false;
}
}
@@ -713,7 +645,7 @@ void StringReader::HandleString(char *str)
char *s = strchr(str, ':');
if (s == nullptr) {
strgen_error("Line has no ':' delimiter");
StrgenError("Line has no ':' delimiter");
return;
}
@@ -728,15 +660,15 @@ void StringReader::HandleString(char *str)
const char *tmp;
for (tmp = s; *tmp != '\0';) {
size_t len = Utf8Validate(tmp);
if (len == 0) strgen_fatal("Invalid UTF-8 sequence in '%s'", s);
if (len == 0) StrgenFatal("Invalid UTF-8 sequence in '{}'", s);
WChar c;
char32_t c;
Utf8Decode(&c, tmp);
if (c <= 0x001F || // ASCII control character range
c == 0x200B || // Zero width space
(c >= 0xE000 && c <= 0xF8FF) || // Private range
(c >= 0xFFF0 && c <= 0xFFFF)) { // Specials range
strgen_fatal("Unwanted UTF-8 character U+%04X in sequence '%s'", c, s);
StrgenFatal("Unwanted UTF-8 character U+{:04X} in sequence '{}'", (int)c, s);
}
tmp += len;
@@ -752,40 +684,40 @@ void StringReader::HandleString(char *str)
if (this->master) {
if (casep != nullptr) {
strgen_error("Cases in the base translation are not supported.");
StrgenError("Cases in the base translation are not supported.");
return;
}
if (ent != nullptr) {
strgen_error("String name '%s' is used multiple times", str);
StrgenError("String name '{}' is used multiple times", str);
return;
}
if (this->data.strings[this->data.next_string_id] != nullptr) {
strgen_error("String ID 0x" PRINTF_SIZEX " for '%s' already in use by '%s'", this->data.next_string_id, str, this->data.strings[this->data.next_string_id]->name);
StrgenError("String ID 0x{:X} for '{}' already in use by '{}'", this->data.next_string_id, str, this->data.strings[this->data.next_string_id]->name);
return;
}
/* Allocate a new LangString */
this->data.Add(str, new LangString(str, s, this->data.next_string_id++, _cur_line));
this->data.Add(std::make_unique<LangString>(str, s, this->data.next_string_id++, _cur_line));
} else {
if (ent == nullptr) {
strgen_warning("String name '%s' does not exist in master file", str);
StrgenWarning("String name '{}' does not exist in master file", str);
return;
}
if (ent->translated && casep == nullptr) {
strgen_error("String name '%s' is used multiple times", str);
if (!ent->translated.empty() && casep == nullptr) {
StrgenError("String name '{}' is used multiple times", str);
return;
}
/* make sure that the commands match */
if (!CheckCommandsMatch(s, ent->english, str)) return;
if (!CheckCommandsMatch(s, ent->english.c_str(), str)) return;
if (casep != nullptr) {
ent->translated_case = new Case(ResolveCaseName(casep, strlen(casep)), s, ent->translated_case);
ent->translated_cases.emplace_back(ResolveCaseName(casep, strlen(casep)), s);
} else {
ent->translated = stredup(s);
ent->translated = s;
/* If the string was translated, use the line from the
* translated language so errors in the translated file
* are properly referenced to. */
@@ -799,27 +731,24 @@ void StringReader::HandlePragma(char *str)
if (!memcmp(str, "plural ", 7)) {
_lang.plural_form = atoi(str + 7);
if (_lang.plural_form >= lengthof(_plural_forms)) {
strgen_fatal("Invalid pluralform %d", _lang.plural_form);
StrgenFatal("Invalid pluralform {}", _lang.plural_form);
}
} else {
strgen_fatal("unknown pragma '%s'", str);
StrgenFatal("unknown pragma '{}'", str);
}
}
static void rstrip(char *buf)
static void StripTrailingWhitespace(std::string &str)
{
size_t i = strlen(buf);
while (i > 0 && (buf[i - 1] == '\r' || buf[i - 1] == '\n' || buf[i - 1] == ' ')) i--;
buf[i] = '\0';
str.erase(str.find_last_not_of("\r\n ") + 1);
}
void StringReader::ParseFile()
{
char buf[2048];
_warnings = _errors = 0;
_translation = this->translation;
_file = this->file;
_file = this->file.c_str();
/* Abusing _show_todo to replace "warning" with "info" for translations. */
_show_todo &= 3;
@@ -832,14 +761,17 @@ void StringReader::ParseFile()
strecpy(_lang.digit_decimal_separator, ".", lastof(_lang.digit_decimal_separator));
_cur_line = 1;
while (this->data.next_string_id < this->data.max_strings && this->ReadLine(buf, lastof(buf)) != nullptr) {
rstrip(buf);
this->HandleString(buf);
while (this->data.next_string_id < this->data.max_strings) {
std::optional<std::string> line = this->ReadLine();
if (!line.has_value()) return;
StripTrailingWhitespace(line.value());
this->HandleString(line.value().data());
_cur_line++;
}
if (this->data.next_string_id == this->data.max_strings) {
strgen_error("Too many strings, maximum allowed is " PRINTF_SIZE, this->data.max_strings);
StrgenError("Too many strings, maximum allowed is {}", this->data.max_strings);
}
}
@@ -864,20 +796,20 @@ static int TranslateArgumentIdx(int argidx, int offset)
{
int sum;
if (argidx < 0 || (uint)argidx >= lengthof(_cur_pcs.cmd)) {
strgen_fatal("invalid argidx %d", argidx);
if (argidx < 0 || (uint)argidx >= _cur_pcs.consuming_commands.max_size()) {
StrgenFatal("invalid argidx {}", argidx);
}
const CmdStruct *cs = _cur_pcs.cmd[argidx];
const CmdStruct *cs = _cur_pcs.consuming_commands[argidx];
if (cs != nullptr && cs->consumes <= offset) {
strgen_fatal("invalid argidx offset %d:%d", argidx, offset);
StrgenFatal("invalid argidx offset {}:{}", argidx, offset);
}
if (_cur_pcs.cmd[argidx] == nullptr) {
strgen_fatal("no command for this argidx %d", argidx);
if (_cur_pcs.consuming_commands[argidx] == nullptr) {
StrgenFatal("no command for this argidx {}", argidx);
}
for (int i = sum = 0; i < argidx; i++) {
const CmdStruct *cs = _cur_pcs.cmd[i];
cs = _cur_pcs.consuming_commands[i];
sum += (cs != nullptr) ? cs->consumes : 1;
}
@@ -923,9 +855,9 @@ static void PutCommandString(Buffer *buffer, const char *str)
}
/* Output the one from the master string... it's always accurate. */
cs = _cur_pcs.cmd[_cur_argidx++];
cs = _cur_pcs.consuming_commands[_cur_argidx++];
if (cs == nullptr) {
strgen_fatal("%s: No argument exists at position %d", _cur_ident, _cur_argidx - 1);
StrgenFatal("{}: No argument exists at position {}", _cur_ident, _cur_argidx - 1);
}
}
@@ -942,7 +874,7 @@ void LanguageWriter::WriteLength(uint length)
char buffer[2];
int offs = 0;
if (length >= 0x4000) {
strgen_fatal("string too long");
StrgenFatal("string too long");
}
if (length >= 0xC0) {
@@ -958,16 +890,16 @@ void LanguageWriter::WriteLength(uint length)
*/
void LanguageWriter::WriteLang(const StringData &data)
{
uint *in_use = AllocaM(uint, data.tabs);
std::vector<uint> in_use;
for (size_t tab = 0; tab < data.tabs; tab++) {
uint n = data.CountInUse((uint)tab);
in_use[tab] = n;
in_use.push_back(n);
_lang.offsets[tab] = TO_LE16(n);
for (uint j = 0; j != in_use[tab]; j++) {
const LangString *ls = data.strings[(tab * TAB_SIZE) + j];
if (ls != nullptr && ls->translated == nullptr) _lang.missing++;
const LangString *ls = data.strings[(tab * TAB_SIZE) + j].get();
if (ls != nullptr && ls->translated.empty()) _lang.missing++;
}
}
@@ -981,9 +913,8 @@ void LanguageWriter::WriteLang(const StringData &data)
for (size_t tab = 0; tab < data.tabs; tab++) {
for (uint j = 0; j != in_use[tab]; j++) {
const LangString *ls = data.strings[(tab * TAB_SIZE) + j];
const Case *casep;
const char *cmdp;
const LangString *ls = data.strings[(tab * TAB_SIZE) + j].get();
const std::string *cmdp;
/* For undefined strings, just set that it's an empty string */
if (ls == nullptr) {
@@ -991,13 +922,13 @@ void LanguageWriter::WriteLang(const StringData &data)
continue;
}
_cur_ident = ls->name;
_cur_ident = ls->name.c_str();
_cur_line = ls->line;
/* Produce a message if a string doesn't have a translation. */
if (_show_todo > 0 && ls->translated == nullptr) {
if (_show_todo > 0 && ls->translated.empty()) {
if ((_show_todo & 2) != 0) {
strgen_warning("'%s' is untranslated", ls->name);
StrgenWarning("'{}' is untranslated", ls->name);
}
if ((_show_todo & 1) != 0) {
const char *s = "<TODO> ";
@@ -1006,40 +937,33 @@ void LanguageWriter::WriteLang(const StringData &data)
}
/* Extract the strings and stuff from the english command string */
ExtractCommandString(&_cur_pcs, ls->english, false);
_cur_pcs = ExtractCommandString(ls->english.c_str(), false);
if (ls->translated_case != nullptr || ls->translated != nullptr) {
casep = ls->translated_case;
cmdp = ls->translated;
if (!ls->translated_cases.empty() || !ls->translated.empty()) {
cmdp = &ls->translated;
} else {
casep = nullptr;
cmdp = ls->english;
cmdp = &ls->english;
}
_translated = cmdp != ls->english;
if (casep != nullptr) {
const Case *c;
uint num;
_translated = cmdp != &ls->english;
if (!ls->translated_cases.empty()) {
/* Need to output a case-switch.
* It has this format
* <0x9E> <NUM CASES> <CASE1> <LEN1> <STRING1> <CASE2> <LEN2> <STRING2> <CASE3> <LEN3> <STRING3> <STRINGDEFAULT>
* Each LEN is printed using 2 bytes in big endian order. */
buffer.AppendUtf8(SCC_SWITCH_CASE);
/* Count the number of cases */
for (num = 0, c = casep; c; c = c->next) num++;
buffer.AppendByte(num);
buffer.AppendByte((byte)ls->translated_cases.size());
/* Write each case */
for (c = casep; c != nullptr; c = c->next) {
buffer.AppendByte(c->caseidx);
for (const Case &c : ls->translated_cases) {
buffer.AppendByte(c.caseidx);
/* Make some space for the 16-bit length */
uint pos = (uint)buffer.size();
buffer.AppendByte(0);
buffer.AppendByte(0);
/* Write string */
PutCommandString(&buffer, c->string);
PutCommandString(&buffer, c.string.c_str());
buffer.AppendByte(0); // terminate with a zero
/* Fill in the length */
uint size = (uint)buffer.size() - (pos + 2);
@@ -1048,7 +972,7 @@ void LanguageWriter::WriteLang(const StringData &data)
}
}
if (cmdp != nullptr) PutCommandString(&buffer, cmdp);
if (!cmdp->empty()) PutCommandString(&buffer, cmdp->c_str());
this->WriteLength((uint)buffer.size());
this->Write(buffer.data(), buffer.size());