void normalizeInput(SWKey &key, std::string &text) {
int utf8State = detectUTF8(text.c_str());
if (normalize) {
// Don't need to normalize text that is ASCII
// But assume other non-UTF-8 text is Latin1 (cp1252) and convert it to UTF-8
if (!utf8State) {
cout << "Warning: " << key.getText() << ": Converting to UTF-8 (" << text << ")" << endl;
converter.processText(text, (SWKey *)2); // note the hack of 2 to mimic a real key. TODO: remove all hacks
converted++;
// Prepare for double check. This probably can be removed.
// But for now we are running the check again.
// This is to determine whether we need to normalize output of the conversion.
utf8State = detectUTF8(text.c_str());
}
// Double check. This probably can be removed.
if (!utf8State) {
cout << "Error: " << key.getText() << ": Converting to UTF-8 (" << text << ")" << endl;
}
if (utf8State > 0) {
std::string before = text;
normalizer->processText(text, (SWKey *)2); // note the hack of 2 to mimic a real key. TODO: remove all hacks
if (before != text) {
normalized++;
}
}
}
}
void ListKey::add(const SWKey &ikey) {
if (++arraycnt > arraymax) {
array = (SWKey **) ((array) ? realloc(array, (arraycnt + 32) * sizeof(SWKey *)) : calloc(arraycnt + 32, sizeof(SWKey *)));
arraymax = arraycnt + 32;
}
array[arraycnt-1] = ikey.clone();
setToElement(arraycnt-1);
}
void writeEntry(SWKey &key, std::string &text) {
#ifdef DEBUG
cout << "(" << entryCount << ") " << key.getText() << endl;
#endif
module->setKey(key);
normalizeInput(key, text);
module->setEntry(text.c_str());
}
void ListKey::setText(const char *ikey) {
// at least try to set the current element to this text
for (arraypos = 0; arraypos < arraycnt; arraypos++) {
SWKey *key = array[arraypos];
if (key) {
if (key->isTraversable() && key->isBoundSet()) {
key->setText(ikey);
if (!key->popError())
break;
}
else {
if (!strcmp(key->getText(), ikey))
break;
}
}
}
if (arraypos >= arraycnt) {
error = 1;
arraypos = arraycnt-1;
}
SWKey::setText(ikey);
}
const QString CInfoDisplay::decodeCrossReference( const QString& data ) {
Q_ASSERT(!data.isEmpty());
if (data.isEmpty()) {
return QString("<div class=\"crossrefinfo\"><h3>%1</h3></div>")
.arg(tr("Cross references"));
}
// qWarning("setting crossref %s", data.latin1());
CSwordBackend::DisplayOptions dispOpts;
dispOpts.lineBreaks = false;
dispOpts.verseNumbers = true;
CSwordBackend::FilterOptions filterOpts;
filterOpts.headings = false;
filterOpts.strongNumbers = false;
filterOpts.morphTags = false;
filterOpts.lemmas = false;
filterOpts.footnotes = false;
filterOpts.scriptureReferences = false;
CrossRefRendering renderer(dispOpts, filterOpts);
CTextRendering::KeyTree tree;
// const bool isBible = true;
CSwordModuleInfo* module = CBTConfig::get
(CBTConfig::standardBible);
//a prefixed module gives the module to look into
QRegExp re("^[^ ]+:");
// re.setMinimal(true);
int pos = re.indexIn(data);
if (pos != -1) {
pos += re.matchedLength()-1;
}
if (pos > 0) {
const QString moduleName = data.left(pos);
// qWarning("found module %s", moduleName.latin1());
module = CPointers::backend()->findModuleByName(moduleName);
if (!module) {
module = CBTConfig::get
(CBTConfig::standardBible);
}
// Q_ASSERT(module);
}
//Q_ASSERT(module); //why? the existense of the module is tested later
CTextRendering::KeyTreeItem::Settings settings (
false,
CTextRendering::KeyTreeItem::Settings::CompleteShort
);
if (module && (module->type() == CSwordModuleInfo::Bible)) {
VerseKey vk;
sword::ListKey refs = vk.ParseVerseList((const char*)data.mid((pos == -1) ? 0 : pos+1).toUtf8(), "Gen 1:1", true);
for (int i = 0; i < refs.Count(); ++i) {
SWKey* key = refs.getElement(i);
Q_ASSERT(key);
VerseKey* vk = dynamic_cast<VerseKey*>(key);
CTextRendering::KeyTreeItem* itm = (CTextRendering::KeyTreeItem*)0; //explicit conversion for MS VS
if (vk && vk->isBoundSet()) { //render a range of keys
itm = new CTextRendering::KeyTreeItem(
QString::fromUtf8(vk->LowerBound().getText()),
QString::fromUtf8(vk->UpperBound().getText()),
module,
settings
);
}
else {
itm = new CTextRendering::KeyTreeItem(
QString::fromUtf8(key->getText()),
QString::fromUtf8(key->getText()),
module,
settings
);
}
Q_ASSERT(itm);
tree.append( itm );
}
}
else if (module) {
CTextRendering::KeyTreeItem* itm = new CTextRendering::KeyTreeItem(
data.mid((pos == -1) ? 0 : pos+1),
module,
settings
);
tree.append( itm );
}
// qWarning("rendered the tree: %s", renderer.renderKeyTree(tree).latin1());
//spanns containing rtl text need dir=rtl on their parent tag to be aligned properly
QString lang = "en"; // default english
if (module)
lang = module->language()->abbrev();
return QString("<div class=\"crossrefinfo\" lang=\"%1\"><h3>%2</h3><div class=\"para\" dir=\"%3\">%4</div></div>")
.arg(lang)
.arg(tr("Cross references"))
.arg(module ? ((module->textDirection() == CSwordModuleInfo::LeftToRight) ? "ltr" : "rtl") : "")
.arg(renderer.renderKeyTree(tree));
}
int main(int argc, char **argv)
{
BlockType iType = BOOKBLOCKS;
int compType = 1;
string cipherKey = "";
std::unique_ptr<SWCompress> compressor;
int compLevel = 0;
if ((argc < 3) || (argc > 7)) {
errorOutHelp(argv[0]);
}
if (argc > 3) {
switch (std::atoi(argv[3])) {
case 2: iType = VERSEBLOCKS; break;
case 3: iType = CHAPTERBLOCKS; break;
case 4: iType = BOOKBLOCKS; break;
default:
std::cerr << "Argument 3 must be one of <2|3|4>\n";
errorOutHelp(*argv);
}
if (argc > 4) {
compType = std::atoi(argv[4]);
if (argc > 5) {
compLevel = std::atoi(argv[5]);
if (argc > 6) {
cipherKey = argv[6];
}
}
}
}
if ((iType < 2) || (compType < 1) || (compType > 4) || compLevel < 0 || compLevel > 9 || (!std::strcmp(argv[1], "-h")) || (!std::strcmp(argv[1], "--help")) || (!std::strcmp(argv[1], "/?")) || (!std::strcmp(argv[1], "-?")) || (!std::strcmp(argv[1], "-help"))) {
errorOutHelp(argv[0]);
}
SWMgr mgr;
auto const it = mgr.modules().find(argv[1]);
if (it == mgr.modules().end()) {
fprintf(stderr, "error: %s: couldn't find module: %s\n", argv[0], argv[1]);
std::exit(-2);
}
SWModule & inModule = *it->second;
// Try to initialize a default set of datafiles and indicies at our
// datapath location passed to us from the user.
#define BIBLE 1
#define LEX 2
#define COM 3
int modType = 0;
if (inModule.getType() == "Biblical Texts") {
modType = BIBLE;
} else if (inModule.getType() == "Lexicons / Dictionaries") {
modType = LEX;
} else if (inModule.getType() == "Commentaries") {
modType = COM;
}
switch (compType) { // these are deleted by zText
case 1: compressor = std::make_unique<LZSSCompress>(); break;
case 2: compressor = std::make_unique<ZipCompress>(); break;
case 3: compressor = std::make_unique<Bzip2Compress>(); break;
case 4: compressor = std::make_unique<XzCompress>(); break;
}
if (compressor && compLevel > 0) {
compressor->setLevel(compLevel);
}
int result = 0;
switch (modType) {
case BIBLE:
case COM: {
SWKey *k = inModule.getKey();
VerseKey *vk = dynamic_cast<VerseKey *>(k);
result = zText::createModule(argv[2], iType, vk->getVersificationSystem().c_str());
break;
}
case LEX:
result = zLD::createModule(argv[2]);
break;
}
if (result) {
fprintf(stderr, "error: %s: couldn't create module at path: %s\n", argv[0], argv[2]);
std::exit(-3);
}
std::unique_ptr<SWModule> outModule;
switch (modType) {
case BIBLE:
case COM: {
SWKey *k = inModule.getKey();
VerseKey *vk = dynamic_cast<VerseKey *>(k);
outModule = std::make_unique<zText>(argv[2], nullptr, nullptr, iType, std::move(compressor),
ENC_UNKNOWN, DIRECTION_LTR, FMT_UNKNOWN, nullptr,
vk->getVersificationSystem().c_str()); // open our datapath with our RawText driver.
static_cast<VerseKey *>(inModule.getKey())->setIntros(true);
break;
}
case LEX:
outModule = std::make_unique<zLD>(argv[2], nullptr, nullptr, iType, std::move(compressor)); // open our datapath with our RawText driver.
break;
}
if (!cipherKey.empty())
outModule->addRawFilter(
std::make_shared<CipherFilter>(cipherKey.c_str()));
string lastBuffer = "Something that would never be first module entry";
SWKey bufferKey;
SWKey lastBufferKey;
auto const outModuleKey(outModule->createKey());
VerseKey *vkey = dynamic_cast<VerseKey *>(outModuleKey.get());
outModuleKey->setPersist(true);
if (vkey) {
vkey->setIntros(true);
vkey->setAutoNormalize(false);
}
outModule->setKey(*outModuleKey);
inModule.setSkipConsecutiveLinks(false);
inModule.positionToTop();
while (!inModule.popError()) {
bufferKey.positionFrom(*inModule.getKey());
// pseudo-check for link. Will get most common links.
if ((lastBuffer == inModule.getRawEntry()) &&(lastBuffer.length() > 0)) {
outModuleKey->positionFrom(bufferKey);
outModule->linkEntry(lastBufferKey); // link to last key
cout << "Adding [" << bufferKey.getText() << "] link to: [" << lastBufferKey.getText() << "]\n";
}
else {
lastBuffer = inModule.getRawEntry();
lastBufferKey.setText(inModule.getKeyText());
if (lastBuffer.length() > 0) {
cout << "Adding [" << bufferKey.getText() << "] new text.\n";
outModuleKey->positionFrom(bufferKey);
// outModule->getRawEntry(); // snap
// outModule->setKey(bufferKey);
outModule->setEntry(lastBuffer.c_str()); // save new text;
}
else {
cout << "Skipping [" << bufferKey.getText() << "] no entry in Module.\n";
}
}
inModule.increment();
}
}
const char *ListKey::getShortText() const {
int pos = arraypos;
SWKey *key = (pos >= arraycnt || !arraycnt) ? 0:array[pos];
return (key) ? key->getShortText() : keytext;
}
