Move revised ZRCola=>Unicode transliteration into the database
Signed-off-by: Simon Rozman <simon@rozman.si>
This commit is contained in:
parent
11924089d3
commit
f6b3b4568a
@ -729,27 +729,6 @@ bool ZRCola::DBSource::GetNormPerm(const winstd::com_obj<ADORecordset>& rs, std:
|
||||
}
|
||||
|
||||
|
||||
bool ZRCola::DBSource::SelectAllTranslations(com_obj<ADORecordset> &rs) const
|
||||
{
|
||||
// Create a new recordset.
|
||||
rs.free();
|
||||
wxCHECK(SUCCEEDED(::CoCreateInstance(CLSID_CADORecordset, NULL, CLSCTX_ALL, IID_IADORecordset, (LPVOID*)&rs)), false);
|
||||
|
||||
// Open it.
|
||||
if (FAILED(rs->Open(variant(
|
||||
L"SELECT [komb], [rang_komb], [Kano], [Kanoniziraj], [znak], [rang_znak] "
|
||||
L"FROM [VRS_ReplChar] "
|
||||
L"ORDER BY [rang_komb], LEN([komb]) DESC"), variant(m_db), adOpenStatic, adLockReadOnly, adCmdText)))
|
||||
{
|
||||
_ftprintf(stderr, wxT("%s: error ZCC0040: Error loading translations from database. Please make sure the file is ZRCola.zrc compatible.\n"), m_filename.c_str());
|
||||
LogErrors();
|
||||
return false;
|
||||
}
|
||||
|
||||
return true;
|
||||
}
|
||||
|
||||
|
||||
bool ZRCola::DBSource::SelectTranslations(com_obj<ADORecordset> &rs) const
|
||||
{
|
||||
// Create a new recordset.
|
||||
@ -1260,30 +1239,6 @@ bool ZRCola::DBSource::SelectCharacters(com_obj<ADORecordset>& rs) const
|
||||
}
|
||||
|
||||
|
||||
bool ZRCola::DBSource::SelectPUACharacters(com_obj<ADORecordset>& rs) const
|
||||
{
|
||||
// Create a new recordset.
|
||||
rs.free();
|
||||
wxCHECK(SUCCEEDED(::CoCreateInstance(CLSID_CADORecordset, NULL, CLSCTX_ALL, IID_IADORecordset, (LPVOID*)&rs)), false);
|
||||
|
||||
// Open it.
|
||||
if (FAILED(rs->Open(variant(
|
||||
L"SELECT [znak], [opis_en], [kat], [znak_v], [znak_m] "
|
||||
L"FROM [VRS_CharList] "
|
||||
L"WHERE "
|
||||
L"[znak]>='E000' AND [znak]<='F8FF' AND " // Private-Use-Area
|
||||
L"[aktiven]=1 " // Active characters only
|
||||
L"ORDER BY [znak]"), variant(m_db), adOpenStatic, adLockReadOnly, adCmdText)))
|
||||
{
|
||||
_ftprintf(stderr, wxT("%s: error ZCC0120: Error loading characters from database. Please make sure the file is ZRCola.zrc compatible.\n"), m_filename.c_str());
|
||||
LogErrors();
|
||||
return false;
|
||||
}
|
||||
|
||||
return true;
|
||||
}
|
||||
|
||||
|
||||
bool ZRCola::DBSource::GetCharacter(const com_obj<ADORecordset>& rs, character& chr) const
|
||||
{
|
||||
wxASSERT_MSG(rs, wxT("recordset is empty"));
|
||||
|
@ -131,12 +131,8 @@ namespace ZRCola {
|
||||
charseq src; ///< Source sequence
|
||||
std::string norm; ///< Normalization footprint
|
||||
charseq dst; ///< Destination sequence
|
||||
double score; ///< Score
|
||||
|
||||
inline translation() :
|
||||
set((int)ZRCOLA_TRANSEQID_DEFAULT),
|
||||
score(0)
|
||||
{}
|
||||
inline translation() : set((int)ZRCOLA_TRANSEQID_DEFAULT) {}
|
||||
};
|
||||
|
||||
|
||||
@ -625,17 +621,6 @@ namespace ZRCola {
|
||||
///
|
||||
bool GetNormPerm(const winstd::com_obj<ADORecordset>& rs, std::string& norm, normperm& np) const;
|
||||
|
||||
///
|
||||
/// Returns all character translations
|
||||
///
|
||||
/// \param[out] rs Recordset with results
|
||||
///
|
||||
/// \returns
|
||||
/// - true when query succeeds
|
||||
/// - false otherwise
|
||||
///
|
||||
bool SelectAllTranslations(winstd::com_obj<ADORecordset>& rs) const;
|
||||
|
||||
///
|
||||
/// Returns character translations
|
||||
///
|
||||
@ -820,17 +805,6 @@ namespace ZRCola {
|
||||
///
|
||||
bool SelectCharacters(winstd::com_obj<ADORecordset>& rs) const;
|
||||
|
||||
///
|
||||
/// Returns Private-Use-Area characters
|
||||
///
|
||||
/// \param[out] rs Recordset with results
|
||||
///
|
||||
/// \returns
|
||||
/// - true when query succeeds
|
||||
/// - false otherwise
|
||||
///
|
||||
bool SelectPUACharacters(winstd::com_obj<ADORecordset>& rs) const;
|
||||
|
||||
///
|
||||
/// Returns character data
|
||||
///
|
||||
|
@ -19,10 +19,6 @@
|
||||
|
||||
#include "pch.h"
|
||||
|
||||
#define FONT_MATCH_WIDTH 512 // must be a multiple of 8
|
||||
#define FONT_MATCH_HEIGHT 512
|
||||
#define FONT_MATCH_THRESHOLD 8e-2
|
||||
|
||||
using namespace std;
|
||||
using namespace stdex;
|
||||
using namespace winstd;
|
||||
@ -228,56 +224,6 @@ static inline set<ZRCola::DBSource::charseq> permutate_and_translate_inv(_In_ co
|
||||
}
|
||||
|
||||
|
||||
static bool contains_pua(_In_ const wstring &str)
|
||||
{
|
||||
for (auto p = str.c_str(), p_end = str.c_str() + str.size(); p < p_end; p++)
|
||||
if (L'\ue000' <= *p && *p <= L'\uf8ff')
|
||||
return true;
|
||||
return false;
|
||||
}
|
||||
|
||||
|
||||
static void replace_all(_Inout_ wstring &str, _In_ const wstring &from, _In_ const wstring &to)
|
||||
{
|
||||
size_t start_pos = 0;
|
||||
while ((start_pos = str.find(from, start_pos)) != wstring::npos) {
|
||||
str.replace(start_pos, from.length(), to);
|
||||
start_pos += to.length();
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
static double compare_bitmaps(
|
||||
_In_count_c_(FONT_MATCH_WIDTH * FONT_MATCH_HEIGHT / 8) const unsigned char *bits_orig,
|
||||
_In_count_c_(FONT_MATCH_WIDTH * FONT_MATCH_HEIGHT / 8) const unsigned char *bits)
|
||||
{
|
||||
#define B2(n) n, n + 1, n + 1, n + 2
|
||||
#define B4(n) B2(n), B2(n + 1), B2(n + 1), B2(n + 2)
|
||||
#define B6(n) B4(n), B4(n + 1), B4(n + 1), B4(n + 2)
|
||||
static const unsigned char number_of_bits[256] = { B6(0), B6(1), B6(1), B6(2) };
|
||||
#undef B2
|
||||
#undef B4
|
||||
#undef B6
|
||||
// Set divisors to 1 to prevent divide-by-zero.
|
||||
size_t b_orig = 1, b = 1, x = 0;
|
||||
for (size_t i = 0; i < FONT_MATCH_WIDTH * FONT_MATCH_HEIGHT / 8; ++i) {
|
||||
b_orig += number_of_bits[bits_orig[i]];
|
||||
b += number_of_bits[bits [i]];
|
||||
x += number_of_bits[bits_orig[i] ^ bits[i]];
|
||||
}
|
||||
return (double)x/b_orig * (double)x/b;
|
||||
}
|
||||
|
||||
|
||||
static string make_unicode(_In_ const wstring &str)
|
||||
{
|
||||
string out;
|
||||
for (size_t i = 0, n = str.length(); i < n; i++)
|
||||
out += string_printf(i ? "+%04X" : "%04X", str[i]);
|
||||
return out;
|
||||
}
|
||||
|
||||
|
||||
///
|
||||
/// Main function
|
||||
///
|
||||
@ -308,7 +254,6 @@ int _tmain(int argc, _TCHAR *argv[])
|
||||
{ wxCMD_LINE_PARAM , NULL, NULL , _("<Input file>" ), wxCMD_LINE_VAL_STRING, wxCMD_LINE_OPTION_MANDATORY },
|
||||
{ wxCMD_LINE_PARAM , NULL, NULL , _("<Output file>" ), wxCMD_LINE_VAL_STRING, wxCMD_LINE_OPTION_MANDATORY },
|
||||
{ wxCMD_LINE_OPTION, NULL, "pot-cat", _("Output POT catalog" ), wxCMD_LINE_VAL_STRING, wxCMD_LINE_PARAM_OPTIONAL },
|
||||
{ wxCMD_LINE_OPTION, NULL, "csv-rep", _("Output CSV report" ), wxCMD_LINE_VAL_STRING, wxCMD_LINE_PARAM_OPTIONAL },
|
||||
|
||||
{ wxCMD_LINE_NONE }
|
||||
};
|
||||
@ -355,10 +300,6 @@ int _tmain(int argc, _TCHAR *argv[])
|
||||
bool build_pot = parser.Found("pot-cat", &filenamePot);
|
||||
set<wstring> pot;
|
||||
|
||||
wxString filenameCsv;
|
||||
bool build_csv = parser.Found("csv-rep", &filenameCsv);
|
||||
vector<ZRCola::DBSource::translation> csv;
|
||||
|
||||
// Open file ID.
|
||||
streamoff dst_start = idrec::open<ZRCola::recordid_t, ZRCola::recordsize_t>(dst, ZRCOLA_DB_ID);
|
||||
|
||||
@ -395,19 +336,6 @@ int _tmain(int argc, _TCHAR *argv[])
|
||||
}
|
||||
|
||||
{
|
||||
// Build ZRCola Decomposed to ZRCola Composed translation set.
|
||||
ZRCola::DBSource::transet ts;
|
||||
ts.set = (int)ZRCOLA_TRANSEQID_DEFAULT;
|
||||
ts.src = L"ZRCola Decomposed";
|
||||
ts.dst = L"ZRCola Composed";
|
||||
if (build_pot) {
|
||||
pot.insert(ts.src);
|
||||
pot.insert(ts.dst);
|
||||
}
|
||||
|
||||
// Add translation set to index and data.
|
||||
db_transset << ts;
|
||||
|
||||
// Get translations.
|
||||
com_obj<ADORecordset> rs;
|
||||
if (src.SelectTranslations(rs)) {
|
||||
@ -487,249 +415,6 @@ int _tmain(int argc, _TCHAR *argv[])
|
||||
}
|
||||
}
|
||||
|
||||
{
|
||||
// Build ZRCola to Unicode translation set.
|
||||
ZRCola::DBSource::transet ts;
|
||||
ts.set = (int)ZRCOLA_TRANSEQID_UNICODE;
|
||||
ts.src = L"ZRCola Composed";
|
||||
ts.dst = L"Unicode";
|
||||
if (build_pot) {
|
||||
pot.insert(ts.src);
|
||||
pot.insert(ts.dst);
|
||||
}
|
||||
|
||||
// Add translation set to index and data.
|
||||
db_transset << ts;
|
||||
|
||||
// Get all translations.
|
||||
com_obj<ADORecordset> rs;
|
||||
if (src.SelectAllTranslations(rs)) {
|
||||
// Parse translations and build temporary database.
|
||||
vector<ZRCola::DBSource::translation> db_all, db_combining;
|
||||
for (; !ZRCola::DBSource::IsEOF(rs); rs->MoveNext()) {
|
||||
// Read translation from the database.
|
||||
ZRCola::DBSource::translation trans;
|
||||
if (src.GetTranslation(rs, trans)) {
|
||||
// Add translation to temporary databases.
|
||||
db_all.push_back(trans);
|
||||
if (!trans.src.str.empty() && trans.src.str[0] == L'\u203f') {
|
||||
trans.src.str.erase(0, 1);
|
||||
db_combining.push_back(trans);
|
||||
}
|
||||
} else
|
||||
has_errors = true;
|
||||
}
|
||||
|
||||
com_obj<ADORecordset> rs2;
|
||||
if (src.SelectPUACharacters(rs2)) {
|
||||
// Parse characters and build translations.
|
||||
static const LOGFONT
|
||||
lf_zrcola = {
|
||||
-FONT_MATCH_HEIGHT/2, 0,
|
||||
0, 0,
|
||||
FW_NORMAL,
|
||||
FALSE, FALSE, FALSE,
|
||||
ANSI_CHARSET,
|
||||
OUT_DEFAULT_PRECIS, CLIP_DEFAULT_PRECIS, DEFAULT_QUALITY,
|
||||
DEFAULT_PITCH | FF_DONTCARE,
|
||||
TEXT("ZRCola")
|
||||
},
|
||||
lf_times = {
|
||||
-FONT_MATCH_HEIGHT/2, 0,
|
||||
0, 0,
|
||||
FW_NORMAL,
|
||||
FALSE, FALSE, FALSE,
|
||||
ANSI_CHARSET,
|
||||
OUT_DEFAULT_PRECIS, CLIP_DEFAULT_PRECIS, DEFAULT_QUALITY,
|
||||
DEFAULT_PITCH | FF_DONTCARE,
|
||||
TEXT("Times New Roman")
|
||||
};
|
||||
gdi_handle<HFONT>
|
||||
fnt_zrcola(::CreateFontIndirect(&lf_zrcola)),
|
||||
fnt_times (::CreateFontIndirect(&lf_times ));
|
||||
gdi_handle<HBRUSH> brush_bg(::CreateSolidBrush(RGB(0x00, 0x00, 0x00)));
|
||||
gdi_handle<HBITMAP>
|
||||
bmp_orig(::CreateBitmap(FONT_MATCH_WIDTH, FONT_MATCH_HEIGHT, 1, 1, NULL)),
|
||||
bmp_comb(::CreateBitmap(FONT_MATCH_WIDTH, FONT_MATCH_HEIGHT, 1, 1, NULL)),
|
||||
bmp_pre (::CreateBitmap(FONT_MATCH_WIDTH, FONT_MATCH_HEIGHT, 1, 1, NULL));
|
||||
dc
|
||||
dc_orig(::CreateCompatibleDC(NULL)),
|
||||
dc_comb(::CreateCompatibleDC(NULL)),
|
||||
dc_pre (::CreateCompatibleDC(NULL));
|
||||
SetBkColor(dc_orig, RGB(0x00, 0x00, 0x00));
|
||||
SetBkColor(dc_comb, RGB(0x00, 0x00, 0x00));
|
||||
SetBkColor(dc_pre , RGB(0x00, 0x00, 0x00));
|
||||
SetBkMode (dc_orig, TRANSPARENT);
|
||||
SetBkMode (dc_comb, TRANSPARENT);
|
||||
SetBkMode (dc_pre , TRANSPARENT);
|
||||
SetTextColor(dc_orig, RGB(0xff, 0xff, 0xff));
|
||||
SetTextColor(dc_comb, RGB(0xff, 0xff, 0xff));
|
||||
SetTextColor(dc_pre , RGB(0xff, 0xff, 0xff));
|
||||
SetTextAlign(dc_orig, TA_BASELINE | TA_CENTER | TA_NOUPDATECP);
|
||||
SetTextAlign(dc_comb, TA_BASELINE | TA_CENTER | TA_NOUPDATECP);
|
||||
SetTextAlign(dc_pre , TA_BASELINE | TA_CENTER | TA_NOUPDATECP);
|
||||
dc_selector
|
||||
selector_font_orig(dc_orig, fnt_zrcola),
|
||||
selector_font_comb(dc_comb, fnt_times ),
|
||||
selector_font_pre (dc_pre , fnt_times );
|
||||
struct {
|
||||
BITMAPINFOHEADER bmiHeader;
|
||||
RGBQUAD bmiColors[2];
|
||||
} bmi =
|
||||
{
|
||||
{
|
||||
sizeof(BITMAPINFOHEADER),
|
||||
FONT_MATCH_WIDTH,
|
||||
FONT_MATCH_HEIGHT,
|
||||
1,
|
||||
1,
|
||||
BI_RGB,
|
||||
0,
|
||||
3780, 3780,
|
||||
2, 0
|
||||
},
|
||||
{
|
||||
{ 0x00, 0x00, 0x00 },
|
||||
{ 0xff, 0xff, 0xff },
|
||||
}
|
||||
};
|
||||
vector<unsigned char>
|
||||
bits_orig(FONT_MATCH_WIDTH * FONT_MATCH_HEIGHT / 8),
|
||||
bits_comb(FONT_MATCH_WIDTH * FONT_MATCH_HEIGHT / 8),
|
||||
bits_pre (FONT_MATCH_WIDTH * FONT_MATCH_HEIGHT / 8);
|
||||
map<wstring, map<wstring, pair<double, int>>> trans;
|
||||
|
||||
auto process_permutation = [&] (const wstring &comp_orig, const wstring &decomp_orig) {
|
||||
// Replace ZRCola decomposition with Unicode combining characters wherever possible.
|
||||
wstring decomp = decomp_orig;
|
||||
for (auto i = db_combining.cbegin(), i_end = db_combining.cend(); i != i_end; ++i)
|
||||
replace_all(decomp, i->src.str, i->dst.str);
|
||||
wstring comp = decomp;
|
||||
for (auto i = db_all.cbegin(), i_end = db_all.cend(); i != i_end; ++i)
|
||||
replace_all(comp, i->src.str, i->dst.str);
|
||||
// Check if we got anything useful.
|
||||
if (comp_orig == comp ||
|
||||
contains_pua(comp))
|
||||
return;
|
||||
// Do the Unicode normalization.
|
||||
wstring comp_pre;
|
||||
if (comp.length() > 2) {
|
||||
NormalizeString(NormalizationC, comp.c_str(), 2, comp_pre);
|
||||
comp_pre += comp.c_str() + 2;
|
||||
} else
|
||||
NormalizeString(NormalizationC, comp, comp_pre);
|
||||
{
|
||||
// Paint original character and Unicode precomposed/combining one.
|
||||
dc_selector
|
||||
selector_bmp_orig(dc_orig, bmp_orig),
|
||||
selector_bmp_comb(dc_comb, bmp_comb),
|
||||
selector_bmp_pre (dc_pre , bmp_pre );
|
||||
static const RECT bounds = { 0, 0, FONT_MATCH_WIDTH, FONT_MATCH_HEIGHT };
|
||||
FillRect(dc_orig, &bounds, brush_bg);
|
||||
FillRect(dc_comb, &bounds, brush_bg);
|
||||
FillRect(dc_pre , &bounds, brush_bg);
|
||||
TextOutW(dc_orig, FONT_MATCH_WIDTH/2, FONT_MATCH_HEIGHT*5/8, comp_orig.c_str(), comp_orig.length());
|
||||
TextOutW(dc_comb, FONT_MATCH_WIDTH/2, FONT_MATCH_HEIGHT*5/8, comp .c_str(), comp .length());
|
||||
TextOutW(dc_pre , FONT_MATCH_WIDTH/2, FONT_MATCH_HEIGHT*5/8, comp_pre .c_str(), comp_pre .length());
|
||||
}
|
||||
// Compare bitmaps.
|
||||
if (!GetDIBits(dc_orig, bmp_orig, 0, FONT_MATCH_HEIGHT, bits_orig.data(), (BITMAPINFO*)&bmi, DIB_PAL_COLORS) ||
|
||||
!GetDIBits(dc_comb, bmp_comb, 0, FONT_MATCH_HEIGHT, bits_comb.data(), (BITMAPINFO*)&bmi, DIB_PAL_COLORS) ||
|
||||
!GetDIBits(dc_pre , bmp_pre , 0, FONT_MATCH_HEIGHT, bits_pre .data(), (BITMAPINFO*)&bmi, DIB_PAL_COLORS))
|
||||
return;
|
||||
double
|
||||
score_comb = compare_bitmaps(bits_orig.data(), bits_comb.data()),
|
||||
score_pre = compare_bitmaps(bits_orig.data(), bits_pre .data());
|
||||
// Add results to a temporary database.
|
||||
auto hit = trans.find(comp_orig);
|
||||
if (hit != trans.end()) {
|
||||
if (build_csv || score_pre <= FONT_MATCH_THRESHOLD) {
|
||||
if (hit->second.find(comp_pre) == hit->second.end())
|
||||
hit->second.insert(make_pair(comp_pre, make_pair(score_pre, 1)));
|
||||
} if ((build_csv || score_comb <= FONT_MATCH_THRESHOLD) && comp_pre != comp) {
|
||||
if (hit->second.find(comp) == hit->second.end())
|
||||
hit->second.insert(make_pair(comp, make_pair(score_comb, 100)));
|
||||
}
|
||||
} else {
|
||||
map<wstring, pair<double, int>> v;
|
||||
if (build_csv || score_pre <= FONT_MATCH_THRESHOLD)
|
||||
v.insert(make_pair(comp_pre, make_pair(score_pre, 1)));
|
||||
if ((build_csv || score_comb <= FONT_MATCH_THRESHOLD) && comp_pre != comp)
|
||||
v.insert(make_pair(comp, make_pair(score_comb, 100)));
|
||||
if (!v.empty())
|
||||
trans.insert(make_pair(comp_orig, std::move(v)));
|
||||
}
|
||||
};
|
||||
|
||||
for (; !ZRCola::DBSource::IsEOF(rs2); rs2->MoveNext()) {
|
||||
// Read character from the database.
|
||||
ZRCola::DBSource::character chr;
|
||||
if (src.GetCharacter(rs2, chr)) {
|
||||
for (auto t = db_all.cbegin(), t_end = db_all.cend(); t != t_end; ++t) {
|
||||
if (t->dst.str != chr.first)
|
||||
continue;
|
||||
|
||||
// Process primary permutation.
|
||||
process_permutation(chr.first, t->src.str);
|
||||
|
||||
// Secondary permutation(s).
|
||||
auto const hit_np = db_np.find(t->norm);
|
||||
if (hit_np != db_np.end()) {
|
||||
for (auto perm = hit_np->second.cbegin(), perm_end = hit_np->second.cend(); perm != perm_end; ++perm) {
|
||||
// Prepare permutated string.
|
||||
translation_db::mapped_type::key_type str_perm;
|
||||
for (auto idx = perm->cbegin(), idx_end = perm->cend(); idx != idx_end; ++idx)
|
||||
str_perm += t->src.str[*idx];
|
||||
|
||||
// Process secondary permutation.
|
||||
process_permutation(chr.first, str_perm);
|
||||
}
|
||||
}
|
||||
}
|
||||
} else
|
||||
has_errors = true;
|
||||
}
|
||||
|
||||
// Preallocate memory.
|
||||
size_t reserve = db_trans.idxSrc.size() + trans.size()*2;
|
||||
db_trans.idxSrc.reserve(reserve);
|
||||
db_trans.idxDst.reserve(reserve);
|
||||
db_trans.data .reserve(reserve*5);
|
||||
if (build_csv)
|
||||
csv.reserve(trans.size()*2);
|
||||
|
||||
ZRCola::DBSource::translation t;
|
||||
t.set = (int)ZRCOLA_TRANSEQID_UNICODE;
|
||||
t.dst.rank = 1;
|
||||
vector<pair<double, pair<wstring, int>>> results;
|
||||
for (auto i = trans.cbegin(), i_end = trans.cend(); i != i_end; ++i) {
|
||||
// Sort results by score.
|
||||
results.clear();
|
||||
results.reserve(i->second.size());
|
||||
for (auto j = i->second.cbegin(), j_end = i->second.cend(); j != j_end; ++j)
|
||||
results.push_back(make_pair(j->second.first, make_pair(j->first, j->second.second)));
|
||||
sort(results.begin(), results.end(), [] (pair<double, pair<wstring, int>> const& a, pair<double, pair<wstring, int>> const& b) { return a.first < b.first; });
|
||||
int rank_comb = 0, rank_pre = 0;
|
||||
for (auto j = results.cbegin(), j_end = results.cend(); j != j_end; ++j) {
|
||||
t.src.str = i->first;
|
||||
t.src.rank = j->second.second + (j->second.second >= 100 ? rank_comb++ : rank_pre++);
|
||||
t.dst.str = j->second.first;
|
||||
t.score = j->first;
|
||||
db_trans << t;
|
||||
if (build_csv)
|
||||
csv.push_back(t);
|
||||
}
|
||||
}
|
||||
} else {
|
||||
_ftprintf(stderr, wxT("%s: error ZCC0016: Error getting characters from database. Please make sure the file is ZRCola.zrc compatible.\n"), (LPCTSTR)filenameIn.c_str());
|
||||
has_errors = true;
|
||||
}
|
||||
} else {
|
||||
_ftprintf(stderr, wxT("%s: error ZCC0003: Error getting translations from database. Please make sure the file is ZRCola.zrc compatible.\n"), (LPCTSTR)filenameIn.c_str());
|
||||
has_errors = true;
|
||||
}
|
||||
}
|
||||
|
||||
{
|
||||
// Get translation sets.
|
||||
com_obj<ADORecordset> rs;
|
||||
@ -737,8 +422,8 @@ int _tmain(int argc, _TCHAR *argv[])
|
||||
size_t count = src.GetRecordsetCount(rs);
|
||||
if (count < 0xffffffff) { // 4G check (-1 is reserved for error condition)
|
||||
// Preallocate memory.
|
||||
db_transset.idxTranSet.reserve((count+2));
|
||||
db_transset.data .reserve((count+2)*4);
|
||||
db_transset.idxTranSet.reserve((count+1));
|
||||
db_transset.data .reserve((count+1)*4);
|
||||
|
||||
// Parse translation sets and build index and data.
|
||||
for (; !ZRCola::DBSource::IsEOF(rs); rs->MoveNext()) {
|
||||
@ -1280,42 +965,6 @@ int _tmain(int argc, _TCHAR *argv[])
|
||||
}
|
||||
}
|
||||
|
||||
if (!has_errors && build_csv) {
|
||||
fstream dst_csv((LPCTSTR)filenameCsv, ios_base::out | ios_base::trunc);
|
||||
if (dst_csv.good()) {
|
||||
dst_csv
|
||||
<< "\xef\xbb\xbf" // UTF-8 BOM
|
||||
<< "\"znak\";"
|
||||
<< "\"znakZRCola\";"
|
||||
<< "\"znakRank\";"
|
||||
<< "\"komb\";"
|
||||
<< "\"kombZRCola\";"
|
||||
<< "\"kombRank\";"
|
||||
<< "\"razlika\"" << endl;
|
||||
wstring_convert<codecvt_utf8<wchar_t>> conv;
|
||||
for (auto i = csv.cbegin(), i_end = csv.cend(); i != i_end; ++i) {
|
||||
dst_csv
|
||||
<< "\"" << make_unicode(i->src.str) << "\";"
|
||||
<< "\"" << conv.to_bytes(i->src.str) << "\";"
|
||||
<< i->src.rank << ";"
|
||||
<< "\"" << make_unicode(i->dst.str) << "\";"
|
||||
<< "\"" << conv.to_bytes(i->dst.str) << "\";"
|
||||
<< i->dst.rank << ";"
|
||||
<< i->score << endl;
|
||||
}
|
||||
|
||||
if (dst_csv.fail()) {
|
||||
_ftprintf(stderr, wxT("%s: error ZCC0013: Writing to CSV report failed.\n"), (LPCTSTR)filenameOut.c_str());
|
||||
has_errors = true;
|
||||
}
|
||||
|
||||
dst_csv.close();
|
||||
} else {
|
||||
_ftprintf(stderr, wxT("%s: error ZCC0012: Error opening CSV report.\n"), filenameOut.fn_str());
|
||||
has_errors = true;
|
||||
}
|
||||
}
|
||||
|
||||
if (has_errors) {
|
||||
dst.close();
|
||||
wxRemoveFile(filenameOut);
|
||||
|
@ -37,7 +37,6 @@
|
||||
#include <stdex/idrec.h>
|
||||
|
||||
#include <WinStd/Common.h>
|
||||
#include <WinStd/GDI.h>
|
||||
|
||||
#include <initguid.h> // GUID helper to prevent LNK2001 errors (unresolved external symbol IID_IADO...)
|
||||
#pragma warning(push)
|
||||
@ -46,7 +45,6 @@
|
||||
#include <adoid.h>
|
||||
#pragma warning(pop)
|
||||
|
||||
#include <memory.h>
|
||||
#include <process.h>
|
||||
#include <tchar.h>
|
||||
|
||||
@ -58,5 +56,4 @@
|
||||
#include <fstream>
|
||||
#include <memory>
|
||||
#include <set>
|
||||
#include <utility>
|
||||
#include <vector>
|
||||
|
@ -47,7 +47,7 @@
|
||||
///
|
||||
/// ZRCola to Unicode translation sequence ID
|
||||
///
|
||||
#define ZRCOLA_TRANSEQID_UNICODE ((ZRCola::transeqid_t)0xfffe)
|
||||
#define ZRCOLA_TRANSEQID_UNICODE ((ZRCola::transeqid_t)31)
|
||||
|
||||
|
||||
namespace ZRCola {
|
||||
|
File diff suppressed because it is too large
Load Diff
Binary file not shown.
@ -694,7 +694,7 @@ msgstr ""
|
||||
msgid "Ukrainian"
|
||||
msgstr ""
|
||||
|
||||
msgid "Unicode"
|
||||
msgid "Unicode Combining"
|
||||
msgstr ""
|
||||
|
||||
msgid "Units"
|
||||
@ -703,7 +703,7 @@ msgstr ""
|
||||
msgid "ZRCola Composed"
|
||||
msgstr ""
|
||||
|
||||
msgid "ZRCola Decomposed"
|
||||
msgid "ZRCola Composed » Unicode Combining"
|
||||
msgstr ""
|
||||
|
||||
msgid "l+j l|j"
|
||||
|
@ -700,8 +700,8 @@ msgstr ""
|
||||
msgid "Ukrainian"
|
||||
msgstr ""
|
||||
|
||||
msgid "Unicode"
|
||||
msgstr "Unicode"
|
||||
msgid "Unicode Combining"
|
||||
msgstr ""
|
||||
|
||||
msgid "Units"
|
||||
msgstr ""
|
||||
@ -709,7 +709,7 @@ msgstr ""
|
||||
msgid "ZRCola Composed"
|
||||
msgstr ""
|
||||
|
||||
msgid "ZRCola Decomposed"
|
||||
msgid "ZRCola Composed » Unicode Combining"
|
||||
msgstr ""
|
||||
|
||||
msgid "l+j l|j"
|
||||
@ -720,3 +720,6 @@ msgstr ""
|
||||
|
||||
msgid "small Case"
|
||||
msgstr ""
|
||||
|
||||
#~ msgid "Unicode"
|
||||
#~ msgstr "Unicode"
|
||||
|
@ -702,8 +702,8 @@ msgstr "турецкий"
|
||||
msgid "Ukrainian"
|
||||
msgstr "украинский"
|
||||
|
||||
msgid "Unicode"
|
||||
msgstr "Юникод"
|
||||
msgid "Unicode Combining"
|
||||
msgstr ""
|
||||
|
||||
msgid "Units"
|
||||
msgstr "Единицы"
|
||||
@ -711,8 +711,8 @@ msgstr "Единицы"
|
||||
msgid "ZRCola Composed"
|
||||
msgstr "ZRCola составленное"
|
||||
|
||||
msgid "ZRCola Decomposed"
|
||||
msgstr "ZRCola разобранное"
|
||||
msgid "ZRCola Composed » Unicode Combining"
|
||||
msgstr ""
|
||||
|
||||
msgid "l+j l|j"
|
||||
msgstr "l+j l|j"
|
||||
@ -722,3 +722,9 @@ msgstr "без + |"
|
||||
|
||||
msgid "small Case"
|
||||
msgstr "строчние буквы"
|
||||
|
||||
#~ msgid "Unicode"
|
||||
#~ msgstr "Юникод"
|
||||
|
||||
#~ msgid "ZRCola Decomposed"
|
||||
#~ msgstr "ZRCola разобранное"
|
||||
|
@ -702,17 +702,17 @@ msgstr "turščina"
|
||||
msgid "Ukrainian"
|
||||
msgstr "ukrajinščina"
|
||||
|
||||
msgid "Unicode"
|
||||
msgstr "Unicode"
|
||||
msgid "Unicode Combining"
|
||||
msgstr "sestavljivi Unicode"
|
||||
|
||||
msgid "Units"
|
||||
msgstr "Enote"
|
||||
|
||||
msgid "ZRCola Composed"
|
||||
msgstr "ZRCola sestavljeno"
|
||||
msgstr "ZRCola-sestavljeno"
|
||||
|
||||
msgid "ZRCola Decomposed"
|
||||
msgstr "ZRCola razstavljeno"
|
||||
msgid "ZRCola Composed » Unicode Combining"
|
||||
msgstr "ZRCola-sestavljeno » sestavljivi Unicode"
|
||||
|
||||
msgid "l+j l|j"
|
||||
msgstr "l+j l|j"
|
||||
@ -722,3 +722,9 @@ msgstr "brez + |"
|
||||
|
||||
msgid "small Case"
|
||||
msgstr "male črke"
|
||||
|
||||
#~ msgid "Unicode"
|
||||
#~ msgstr "Unicode"
|
||||
|
||||
#~ msgid "ZRCola Decomposed"
|
||||
#~ msgstr "ZRCola razstavljeno"
|
||||
|
Loading…
x
Reference in New Issue
Block a user