void TreeKeyIdx::setText(const char *ikey) {
char *buf = 0;
stdstr(&buf, ikey);
SWBuf leaf = strtok(buf, "/");
leaf.trim();
root();
while ((leaf.size()) && (!popError())) {
bool ok, inChild = false;
error = KEYERR_OUTOFBOUNDS;
for (ok = firstChild(); ok; ok = nextSibling()) {
inChild = true;
if (leaf == getLocalName()) {
error = 0;
break;
}
}
leaf = strtok(0, "/");
leaf.trim();
if (!ok) {
if (inChild) { // if we didn't find a matching child node, default to first child
parent();
firstChild();
}
error = KEYERR_OUTOFBOUNDS;
}
}
if (leaf.size())
error = KEYERR_OUTOFBOUNDS;
delete [] buf;
unsnappedKeyText = ikey;
positionChanged();
}
void TreeKey::assureKeyPath(const char *keyBuffer) {
if (!keyBuffer) {
keyBuffer = unsnappedKeyText;
//assert we have something to do before setting root
if (!*keyBuffer)
return;
}
char *keybuf = 0;
stdstr(&keybuf, keyBuffer);
root();
// TODO: change to NOT use strtok. strtok is dangerous.
SWBuf tok = strtok(keybuf, "/");
tok.trim();
while (tok.size()) {
bool foundkey = false;
if (hasChildren()) {
firstChild();
if (tok == getLocalName()) {
foundkey = true;
}
else {
while (nextSibling()) {
if (getLocalName()) {
if (tok == getLocalName()) {
foundkey = true;
break;
}
}
}
}
if (!foundkey) {
append();
setLocalName(tok);
save();
}
}
else {
appendChild();
setLocalName(tok);
save();
}
#ifdef DEBUG
// std::cout << getLocalName() << " : " << tok << std::endl;
#endif
tok = strtok(0, "/");
tok.trim();
}
delete [] keybuf;
}
void assurePath(TreeKeyIdx *treeKey) {
char buf[1023];
std::cout << "Enter Path: ";
fgets(buf, 1000, stdin);
SWBuf path = buf;
treeKey->assureKeyPath(path.trim());
}
void gotoPath(TreeKeyIdx *treeKey) {
char buf[1023];
std::cout << "Enter Path: ";
fgets(buf, 1000, stdin);
SWBuf path = buf;
(*treeKey) = path.trim();
}
void setLocalName(TreeKeyIdx *treeKey) {
char buf[1023];
std::cout << "Enter New Node Name: ";
fgets(buf, 1000, stdin);
SWBuf name = buf;
treeKey->setLocalName(name.trim());
treeKey->save();
}
void appendSibbling(TreeKeyIdx *treeKey) {
if (treeKey->getOffset()) {
char buf[1023];
std::cout << "Enter New Sibbling Name: ";
fgets(buf, 1000, stdin);
SWBuf name = buf;
treeKey->append();
treeKey->setLocalName(name.trim());
treeKey->save();
}
else std::cout << "Can't add sibling to root node\n";
}
int main(int argc, char **argv) {
greekAccentsFilter.setOptionValue("Off"); // off = accents off
parseParams(argc, argv);
// Let's see if we can open our input file
FileDesc *fd = FileMgr::getSystemFileMgr()->open(inFile, FileMgr::RDONLY);
if (fd->getFd() < 0) {
fprintf(stderr, "error: %s: couldn't open input file: %s \n", argv[0], inFile.c_str());
exit(-2);
}
RawGenBook *book;
// Do some initialization stuff
if (!augMod) {
RawGenBook::createModule(outPath);
}
book = new RawGenBook(outPath);
SWBuf lineBuffer;
SWBuf keyBuffer;
SWBuf entBuffer;
bool more = true;
do {
more = FileMgr::getLine(fd, lineBuffer)!=0;
if (lineBuffer.startsWith("$$$")) {
if ((keyBuffer.size()) && (entBuffer.size())) {
writeEntry(book, keyBuffer, entBuffer);
}
keyBuffer = lineBuffer;
keyBuffer << 3;
keyBuffer.trim();
entBuffer.size(0);
}
else {
if (keyBuffer.size()) {
entBuffer += lineBuffer;
entBuffer += "\n";
}
}
} while (more);
if ((keyBuffer.size()) && (entBuffer.size())) {
writeEntry(book, keyBuffer, entBuffer);
}
delete book;
FileMgr::getSystemFileMgr()->close(fd);
return 0;
}
char OSISGlosses::processText(SWBuf &text, const SWKey *key, const SWModule *module) {
SWBuf token;
bool intoken = false;
const SWBuf orig = text;
const char * from = orig.c_str();
if (!option) {
for (text = ""; *from; ++from) {
if (*from == '<') {
intoken = true;
token = "";
continue;
}
if (*from == '>') { // process tokens
intoken = false;
if (token.startsWith("w ")) { // Word
XMLTag wtag(token);
const char *l = wtag.getAttribute("gloss");
if (l) {
wtag.setAttribute("gloss", 0);
token = wtag;
token.trim();
// drop <>
token << 1;
token--;
}
}
// keep token in text
text.append('<');
text.append(token);
text.append('>');
continue;
}
if (intoken) {
token += *from;
}
else {
text.append(*from);
}
}
}
return 0;
}
int main(int argc, char **argv) {
SWMgr library;
SWModule *book = library.getModule("KJV");
VerseKey *vk = (VerseKey *) book->getKey();
for (;!vk->popError();vk->setChapter(vk->getChapter()+1)) {
vk->setVerse(vk->getVerseMax());
SWBuf text = book->stripText();
text = text.trim();
if (text.endsWith(",")) {
cout << vk->getText() << ":\n\n";
cout << text << endl;
}
}
return 0;
}
void setEntryText(RawGenBook *book) {
SWBuf body;
TreeKeyIdx *treeKey = (TreeKeyIdx *)(SWKey *)(*book);
if (treeKey->getOffset()) {
char buf[1023];
std::cout << "Enter New Entry Text ('.' on a line by itself to end): \n";
do {
fgets(buf, 1000, stdin);
SWBuf text = buf;
text.trim();
if ((text[0] == '.') && (text[1] == 0))
break;
body += text;
body += "\n";
} while (true);
(*book) << body.c_str();
}
else std::cout << "Can't add entry text to root node\n";
}
int main(int argc, char **argv) {
if (argc != 2) {
fprintf(stderr, "usage: %s <tree/key/data/path>\n", *argv);
exit(-1);
}
TreeKeyIdx *treeKey = new TreeKeyIdx(argv[1]);
if (treeKey->Error()) {
RawGenBook::createModule(argv[1]);
}
delete treeKey;
RawGenBook *book = new RawGenBook(argv[1]);
TreeKeyIdx root = *((TreeKeyIdx *)((SWKey *)(*book)));
treeKey = (TreeKeyIdx *)(SWKey *)(*book);
SWBuf input;
char line[1024];
do {
std::cout << "[" << treeKey->getText() << "] > ";
fgets(line, 1000, stdin);
input = line;
input.trim();
if (input.length() > 0) {
switch (input[0]) {
case 'n': printLocalName(treeKey); break;
case 's': setLocalName(treeKey); break;
case 'g': gotoPath(treeKey); break;
case 'G': assurePath(treeKey); break;
case 'p': root.root(); printTree(root, treeKey); break;
case 'a': appendSibbling(treeKey); break;
case 'c': appendChild(treeKey); break;
case 'd': deleteNode(treeKey); break;
case 'j': treeKey->nextSibling(); break;
case 'k': treeKey->previousSibling(); break;
case 'h': treeKey->parent(); break;
case 'l': treeKey->firstChild(); break;
case 'r': treeKey->root(); break;
case 't': setEntryText(book); break;
case 'v': viewEntryText(book); break;
case 'q': break;
case '?':
default:
std::cout << "\n p - print tree\n";
std::cout << " n - get local name\n";
std::cout << " s - set local name\n";
std::cout << " j - next sibbling\n";
std::cout << " k - previous sibbling\n";
std::cout << " h - parent\n";
std::cout << " l - first child\n";
std::cout << " r - root\n";
std::cout << " g - goto path\n";
std::cout << " G goto path; create if it doesn't exist\n";
std::cout << " a - append sibbling\n";
std::cout << " c - append child\n";
std::cout << " d - delete node\n";
std::cout << " v - view entry text\n";
std::cout << " t - set entry text\n";
std::cout << " q - quit\n\n";
break;
}
}
}
while (input.compare("q"));
delete treeKey;
return 0;
}
int main(int argc, char **argv) {
// handle options
if (argc < 2) usage(*argv);
const char *progName = argv[0];
const char *inFileName = argv[1];
SWBuf v11n = "KJV";
SWBuf outPath = "./";
SWBuf locale = "en";
bool fourByteSize = false;
bool append = false;
int iType = 4;
SWBuf cipherKey = "";
SWCompress *compressor = 0;
SWBuf compType = "";
for (int i = 2; i < argc; i++) {
if (!strcmp(argv[i], "-a")) {
append = true;
}
else if (!strcmp(argv[i], "-z")) {
if (fourByteSize) usage(*argv, "Cannot specify both -z and -4");
compType = "ZIP";
if (i+1 < argc && argv[i+1][0] != '-') {
switch (argv[++i][0]) {
case 'l': compType = "LZSS"; break;
case 'z': compType = "ZIP"; break;
case 'b': compType = "BZIP2"; break;
case 'x': compType = "XZ"; break;
}
}
}
else if (!strcmp(argv[i], "-Z")) {
if (compType.size()) usage(*argv, "Cannot specify both -z and -Z");
if (fourByteSize) usage(*argv, "Cannot specify both -Z and -4");
compType = "LZSS";
}
else if (!strcmp(argv[i], "-4")) {
fourByteSize = true;
}
else if (!strcmp(argv[i], "-b")) {
if (i+1 < argc) {
iType = atoi(argv[++i]);
if ((iType >= 2) && (iType <= 4)) continue;
}
usage(*argv, "-b requires one of <2|3|4>");
}
else if (!strcmp(argv[i], "-o")) {
if (i+1 < argc) outPath = argv[++i];
else usage(progName, "-o requires <output_path>");
}
else if (!strcmp(argv[i], "-v")) {
if (i+1 < argc) v11n = argv[++i];
else usage(progName, "-v requires <v11n>");
}
else if (!strcmp(argv[i], "-l")) {
if (i+1 < argc) locale = argv[++i];
else usage(progName, "-l requires <locale>");
}
else if (!strcmp(argv[i], "-c")) {
if (i+1 < argc) cipherKey = argv[++i];
else usage(*argv, "-c requires <cipher_key>");
}
else usage(progName, (((SWBuf)"Unknown argument: ")+ argv[i]).c_str());
}
// -----------------------------------------------------
const VersificationMgr::System *v = VersificationMgr::getSystemVersificationMgr()->getVersificationSystem(v11n);
if (!v) std::cout << "Warning: Versification " << v11n << " not found. Using KJV versification...\n";
if (compType == "LZSS") {
compressor = new LZSSCompress();
}
else if (compType == "ZIP") {
#ifndef EXCLUDEZLIB
compressor = new ZipCompress();
#else
usage(*argv, "ERROR: SWORD library not compiled with ZIP compression support.\n\tBe sure libz is available when compiling SWORD library");
#endif
}
else if (compType == "BZIP2") {
#ifndef EXCLUDEBZIP2
compressor = new Bzip2Compress();
#else
usage(*argv, "ERROR: SWORD library not compiled with bzip2 compression support.\n\tBe sure libbz2 is available when compiling SWORD library");
#endif
}
else if (compType == "XZ") {
#ifndef EXCLUDEXZ
compressor = new XzCompress();
#else
usage(*argv, "ERROR: SWORD library not compiled with xz compression support.\n\tBe sure liblzma is available when compiling SWORD library");
#endif
}
// setup module
if (!append) {
if (compressor) {
if (zText::createModule(outPath, iType, v11n)) {
fprintf(stderr, "ERROR: %s: couldn't create module at path: %s \n", *argv, outPath.c_str());
exit(-1);
}
}
else {
if (!fourByteSize)
RawText::createModule(outPath, v11n);
else RawText4::createModule(outPath, v11n);
}
}
SWModule *module = 0;
if (compressor) {
// Create a compressed text module allowing very large entries
// Taking defaults except for first, fourth, fifth and last argument
module = new zText(
outPath, // ipath
0, // iname
0, // idesc
iType, // iblockType
compressor, // icomp
0, // idisp
ENC_UNKNOWN, // enc
DIRECTION_LTR, // dir
FMT_UNKNOWN, // markup
0, // lang
v11n // versification
);
}
else {
module = (!fourByteSize)
? (SWModule *)new RawText(outPath, 0, 0, 0, ENC_UNKNOWN, DIRECTION_LTR, FMT_UNKNOWN, 0, v11n)
: (SWModule *)new RawText4(outPath, 0, 0, 0, ENC_UNKNOWN, DIRECTION_LTR, FMT_UNKNOWN, 0, v11n);
}
SWFilter *cipherFilter = 0;
if (cipherKey.length()) {
fprintf(stderr, "Adding cipher filter with phrase: %s\n", cipherKey.c_str() );
cipherFilter = new CipherFilter(cipherKey.c_str());
module->addRawFilter(cipherFilter);
}
// -----------------------------------------------------
// setup locale manager
LocaleMgr::getSystemLocaleMgr()->setDefaultLocaleName(locale);
// setup module key to allow full range of possible values, and then some
VerseKey *vkey = (VerseKey *)module->createKey();
vkey->setIntros(true);
vkey->setAutoNormalize(false);
vkey->setPersist(true);
module->setKey(*vkey);
// -----------------------------------------------------
// process input file
FileDesc *fd = FileMgr::getSystemFileMgr()->open(inFileName, FileMgr::RDONLY);
SWBuf lineBuffer;
SWBuf keyBuffer;
SWBuf entBuffer;
bool more = true;
do {
more = FileMgr::getLine(fd, lineBuffer)!=0;
if (lineBuffer.startsWith("$$$")) {
if ((keyBuffer.size()) && (entBuffer.size())) {
writeEntry(module, keyBuffer, entBuffer);
}
keyBuffer = lineBuffer;
keyBuffer << 3;
keyBuffer.trim();
entBuffer.size(0);
}
else {
if (keyBuffer.size()) {
entBuffer += lineBuffer;
entBuffer += "\n";
}
}
} while (more);
if ((keyBuffer.size()) && (entBuffer.size())) {
writeEntry(module, keyBuffer, entBuffer);
}
delete module;
if (cipherFilter)
delete cipherFilter;
delete vkey;
FileMgr::getSystemFileMgr()->close(fd);
return 0;
}
char OSISLemma::processText(SWBuf &text, const SWKey *key, const SWModule *module) {
SWBuf token;
bool intoken = false;
const SWBuf orig = text;
const char * from = orig.c_str();
if (!option) {
for (text = ""; *from; ++from) {
if (*from == '<') {
intoken = true;
token = "";
continue;
}
if (*from == '>') { // process tokens
intoken = false;
if (token.startsWith("w ")) { // Word
XMLTag wtag(token);
// always save off lemma if we haven't yet
if (!wtag.getAttribute("savlm")) {
const char *l = wtag.getAttribute("lemma");
if (l) {
wtag.setAttribute("savlm", l);
}
}
int count = wtag.getAttributePartCount("lemma", ' ');
for (int i = 0; i < count; i++) {
SWBuf a = wtag.getAttribute("lemma", i, ' ');
const char *prefix = a.stripPrefix(':');
if ((!prefix) || ((SWBuf)prefix).startsWith("lemma.")) {
// remove attribute part
wtag.setAttribute("lemma", 0, i, ' ');
i--;
count--;
}
}
token = wtag;
token.trim();
// drop <>
token << 1;
token--;
}
// keep token in text
text.append('<');
text.append(token);
text.append('>');
continue;
}
if (intoken) {
token += *from;
}
else {
text.append(*from);
}
}
}
return 0;
}
char OSISStrongs::processText(SWBuf &text, const SWKey *key, const SWModule *module) {
SWBuf token;
bool intoken = false;
int wordNum = 1;
char wordstr[5];
const char *wordStart = 0;
SWBuf page = ""; // some modules include <seg> page info, so we add these to the words
const SWBuf orig = text;
const char * from = orig.c_str();
for (text = ""; *from; ++from) {
if (*from == '<') {
intoken = true;
token = "";
continue;
}
if (*from == '>') { // process tokens
intoken = false;
// possible page seg --------------------------------
if (token.startsWith("seg ")) {
XMLTag stag(token);
SWBuf type = stag.getAttribute("type");
if (type == "page") {
SWBuf number = stag.getAttribute("subtype");
if (number.length()) {
page = number;
}
}
}
// ---------------------------------------------------
if (token.startsWith("w ")) { // Word
XMLTag wtag(token);
if (module->isProcessEntryAttributes()) {
wordStart = from+1;
char gh = 0;
VerseKey *vkey = 0;
if (key) {
vkey = SWDYNAMIC_CAST(VerseKey, key);
}
SWBuf lemma = "";
SWBuf morph = "";
SWBuf src = "";
SWBuf morphClass = "";
SWBuf lemmaClass = "";
const char *attrib;
sprintf(wordstr, "%03d", wordNum);
// why is morph entry attribute processing done in here? Well, it's faster. It makes more local sense to place this code in osismorph.
// easier to keep lemma and morph in same wordstr number too maybe.
if ((attrib = wtag.getAttribute("morph"))) {
int count = wtag.getAttributePartCount("morph", ' ');
int i = (count > 1) ? 0 : -1; // -1 for whole value cuz it's faster, but does the same thing as 0
do {
SWBuf mClass = "";
SWBuf mp = "";
attrib = wtag.getAttribute("morph", i, ' ');
if (i < 0) i = 0; // to handle our -1 condition
const char *m = strchr(attrib, ':');
if (m) {
int len = m-attrib;
mClass.append(attrib, len);
attrib += (len+1);
}
if ((mClass == "x-Robinsons") || (mClass == "x-Robinson") || (mClass == "Robinson")) {
mClass = "robinson";
}
if (i) { morphClass += " "; morph += " "; }
mp += attrib;
morphClass += mClass;
morph += mp;
if (count > 1) {
SWBuf tmp;
tmp.setFormatted("Morph.%d", i+1);
module->getEntryAttributes()["Word"][wordstr][tmp] = mp;
tmp.setFormatted("MorphClass.%d", i+1);
module->getEntryAttributes()["Word"][wordstr][tmp] = mClass;
}
} while (++i < count);
}
if ((attrib = wtag.getAttribute("lemma"))) {
int count = wtag.getAttributePartCount("lemma", ' ');
int i = (count > 1) ? 0 : -1; // -1 for whole value cuz it's faster, but does the same thing as 0
do {
gh = 0;
SWBuf lClass = "";
SWBuf l = "";
attrib = wtag.getAttribute("lemma", i, ' ');
if (i < 0) i = 0; // to handle our -1 condition
const char *m = strchr(attrib, ':');
if (m) {
int len = m-attrib;
lClass.append(attrib, len);
attrib += (len+1);
}
if ((lClass == "x-Strongs") || (lClass == "strong") || (lClass == "Strong")) {
if (isdigit(attrib[0])) {
if (vkey) {
gh = vkey->getTestament() ? 'H' : 'G';
}
}
else {
gh = *attrib;
attrib++;
}
lClass = "strong";
}
if (gh) l += gh;
l += attrib;
if (i) { lemmaClass += " "; lemma += " "; }
lemma += l;
lemmaClass += lClass;
if (count > 1) {
SWBuf tmp;
tmp.setFormatted("Lemma.%d", i+1);
module->getEntryAttributes()["Word"][wordstr][tmp] = l;
tmp.setFormatted("LemmaClass.%d", i+1);
module->getEntryAttributes()["Word"][wordstr][tmp] = lClass;
}
} while (++i < count);
module->getEntryAttributes()["Word"][wordstr]["PartCount"].setFormatted("%d", count);
}
if ((attrib = wtag.getAttribute("src"))) {
int count = wtag.getAttributePartCount("src", ' ');
int i = (count > 1) ? 0 : -1; // -1 for whole value cuz it's faster, but does the same thing as 0
do {
SWBuf mp = "";
attrib = wtag.getAttribute("src", i, ' ');
if (i < 0) i = 0; // to handle our -1 condition
if (i) src += " ";
mp += attrib;
src += mp;
if (count > 1) {
SWBuf tmp;
tmp.setFormatted("Src.%d", i+1);
module->getEntryAttributes()["Word"][wordstr][tmp] = mp;
}
} while (++i < count);
}
if (lemma.length())
module->getEntryAttributes()["Word"][wordstr]["Lemma"] = lemma;
if (lemmaClass.length())
module->getEntryAttributes()["Word"][wordstr]["LemmaClass"] = lemmaClass;
if (morph.length())
module->getEntryAttributes()["Word"][wordstr]["Morph"] = morph;
if (morphClass.length())
module->getEntryAttributes()["Word"][wordstr]["MorphClass"] = morphClass;
if (src.length())
module->getEntryAttributes()["Word"][wordstr]["Src"] = src;
if (page.length())
module->getEntryAttributes()["Word"][wordstr]["Page"] = page;
if (wtag.isEmpty()) {
int j;
for (j = token.length()-1; ((j>0) && (strchr(" /", token[j]))); j--);
token.size(j+1);
}
token += " wn=\"";
token += wordstr;
token += "\"";
if (wtag.isEmpty()) {
token += "/";
}
wordNum++;
}
if (!option) {
/*
* Code which handles multiple lemma types. Kindof works but breaks at least WEBIF filters for strongs.
*
int count = wtag.getAttributePartCount("lemma", ' ');
for (int i = 0; i < count; i++) {
SWBuf a = wtag.getAttribute("lemma", i, ' ');
const char *prefix = a.stripPrefix(':');
if ((prefix) && (!strcmp(prefix, "x-Strongs") || !strcmp(prefix, "strong") || !strcmp(prefix, "Strong"))) {
// remove attribute part
wtag.setAttribute("lemma", 0, i, ' ');
i--;
count--;
}
}
* Instead the codee below just removes the lemma attribute
*****/
const char *l = wtag.getAttribute("lemma");
if (l) {
SWBuf savlm = l;
wtag.setAttribute("lemma", 0);
wtag.setAttribute("savlm", savlm);
token = wtag;
token.trim();
// drop <>
token << 1;
token--;
}
}
}
if (token.startsWith("/w")) { // Word End
if (module->isProcessEntryAttributes()) {
if (wordStart) {
SWBuf tmp;
tmp.append(wordStart, (from-wordStart)-3);
sprintf(wordstr, "%03d", wordNum-1);
module->getEntryAttributes()["Word"][wordstr]["Text"] = tmp;
}
}
wordStart = 0;
}
// keep token in text
text.append('<');
text.append(token);
text.append('>');
continue;
}
if (intoken) {
token += *from;
}
else {
text.append(*from);
}
}
return 0;
}
