char OSISReferenceLinks::processText(std::string &text, const SWKey *key, const SWModule *module) {
(void) key;
(void) module;
if (option) return 0;
std::string token;
bool intoken = false;
bool stripThisToken = false;
std::string orig = text;
const char *from = orig.c_str();
for (text = ""; *from; ++from) {
if (*from == '<') {
intoken = true;
token = "";
continue;
}
else if (*from == '>') { // process tokens
intoken = false;
if (std::strncmp(token.c_str(), "reference", 9)
&& std::strncmp(token.c_str(), "/reference", 10)) {
text.push_back('<');
text.append(token);
text.push_back('>');
}
else {
XMLTag tag;
tag = token.c_str();
if (!tag.isEndTag() && type == tag.getAttribute("type") && (!subType.size() || subType == tag.getAttribute("subType"))) {
stripThisToken = true;
continue;
}
else if (tag.isEndTag() && stripThisToken) {
stripThisToken = false;
continue;
}
text.push_back('<');
text.append(token);
text.push_back('>');
}
continue;
}
if (intoken) { //copy token
token.push_back(*from);
}
else { //copy text which is not inside a token
text.push_back(*from);
}
}
return 0;
}
char OSISFootnotes::processText(SWBuf &text, const SWKey *key, const SWModule *module) {
SWBuf token;
bool intoken = false;
bool hide = false;
SWBuf tagText;
XMLTag startTag;
SWBuf refs = "";
int footnoteNum = 1;
char buf[254];
SWKey *p = (module) ? module->createKey() : (key) ? key->clone() : new VerseKey();
VerseKey *parser = SWDYNAMIC_CAST(VerseKey, p);
if (!parser) {
delete p;
parser = new VerseKey();
}
*parser = key->getText();
SWBuf orig = text;
const char *from = orig.c_str();
XMLTag tag;
bool strongsMarkup = false;
for (text = ""; *from; ++from) {
// remove all newlines temporarily to fix kjv2003 module
if ((*from == 10) || (*from == 13)) {
if ((text.length()>1) && (text[text.length()-2] != ' ') && (*(from+1) != ' '))
text.append(' ');
continue;
}
if (*from == '<') {
intoken = true;
token = "";
continue;
}
if (*from == '>') { // process tokens
intoken = false;
if (!strncmp(token, "note", 4) || !strncmp(token.c_str(), "/note", 5)) {
tag = token;
if (!tag.isEndTag()) {
if (tag.getAttribute("type") && (!strcmp("x-strongsMarkup", tag.getAttribute("type"))
|| !strcmp("strongsMarkup", tag.getAttribute("type"))) // deprecated
) {
tag.setEmpty(false); // handle bug in KJV2003 module where some note open tags were <note ... />
strongsMarkup = true;
}
if (!tag.isEmpty()) {
// if ((!tag.isEmpty()) || (SWBuf("strongsMarkup") == tag.getAttribute("type"))) {
refs = "";
startTag = tag;
hide = true;
tagText = "";
continue;
}
}
if (hide && tag.isEndTag()) {
if (module->isProcessEntryAttributes() && !strongsMarkup) { //don`t parse strongsMarkup to EntryAttributes as Footnote
sprintf(buf, "%i", footnoteNum++);
StringList attributes = startTag.getAttributeNames();
for (StringList::const_iterator it = attributes.begin(); it != attributes.end(); it++) {
module->getEntryAttributes()["Footnote"][buf][it->c_str()] = startTag.getAttribute(it->c_str());
}
module->getEntryAttributes()["Footnote"][buf]["body"] = tagText;
startTag.setAttribute("swordFootnote", buf);
if ((startTag.getAttribute("type")) && (!strcmp(startTag.getAttribute("type"), "crossReference"))) {
if (!refs.length())
refs = parser->parseVerseList(tagText.c_str(), *parser, true).getRangeText();
module->getEntryAttributes()["Footnote"][buf]["refList"] = refs.c_str();
}
}
hide = false;
if (option || (startTag.getAttribute("type") && !strcmp(startTag.getAttribute("type"), "crossReference"))) { // we want the tag in the text; crossReferences are handled by another filter
text.append(startTag);
// text.append(tagText); // we don't put the body back in because it is retrievable from EntryAttributes["Footnotes"][]["body"].
}
else continue;
}
strongsMarkup = false;
}
// if not a heading token, keep token in text
//if ((!strcmp(tag.getName(), "reference")) && (!tag.isEndTag())) {
// SWBuf osisRef = tag.getAttribute("osisRef");
if (!strncmp(token, "reference", 9)) {
if (refs.length()) {
refs.append("; ");
}
const char* attr = strstr(token.c_str() + 9, "osisRef=\"");
const char* end = attr ? strchr(attr+9, '"') : 0;
if (attr && end) {
refs.append(attr+9, end-(attr+9));
}
}
if (!hide) {
text.append('<');
text.append(token);
text.append('>');
}
else {
tagText.append('<');
tagText.append(token);
tagText.append('>');
}
continue;
}
if (intoken) { //copy token
token.append(*from);
}
else if (!hide) { //copy text which is not inside a token
text.append(*from);
}
else tagText.append(*from);
}
delete parser;
return 0;
}
char OSISScripref::processText(std::string &text, const SWKey *key, const SWModule *module) {
(void) key;
(void) module;
std::string token;
bool intoken = false;
bool hide = false;
std::string tagText;
XMLTag startTag;
std::string orig = text;
const char *from = orig.c_str();
XMLTag tag;
for (text = ""; *from; ++from) {
if (*from == '<') {
intoken = true;
token = "";
continue;
}
if (*from == '>') { // process tokens
intoken = false;
tag = token.c_str();
if (!std::strncmp(token.c_str(), "note", 4) || !std::strncmp(token.c_str(), "/note", 5)) {
if (!tag.isEndTag() && !tag.isEmpty()) {
startTag = tag;
if (tag.attribute("type") == "crossReference") {
hide = true;
tagText = "";
if (option) { // we want the tag in the text
text.push_back('<');
text.append(token);
text.push_back('>');
}
continue;
}
}
if (hide && tag.isEndTag()) {
hide = false;
if (option) { // we want the tag in the text
text.append(tagText); // end tag gets added further down
}
else continue; // don't let the end tag get added to the text
}
}
// if not a heading token, keep token in text
if (!hide) {
text.push_back('<');
text.append(token);
text.push_back('>');
}
else {
tagText.push_back('<');
tagText.append(token);
tagText.push_back('>');
}
continue;
}
if (intoken) { //copy token
token.push_back(*from);
}
else if (!hide) { //copy text which is not inside a token
text.push_back(*from);
}
else tagText.push_back(*from);
}
return 0;
}
char ThMLHeadings::processText(SWBuf &text, const SWKey *key, const SWModule *module) {
SWBuf token;
bool intoken = false;
bool isheader = false;
bool hide = false;
bool preverse = false;
bool withinDiv = false;
SWBuf header;
int headerNum = 0;
int pvHeaderNum = 0;
char buf[254];
XMLTag startTag;
SWBuf orig = text;
const char *from = orig.c_str();
XMLTag tag;
for (text = ""; *from; ++from) {
if (*from == '<') {
intoken = true;
token = "";
continue;
}
if (*from == '>') { // process tokens
intoken = false;
if (!strnicmp(token.c_str(), "div", 3) || !strnicmp(token.c_str(), "/div", 4)) {
withinDiv = (!strnicmp(token.c_str(), "div", 3));
tag = token;
if (hide && tag.isEndTag()) {
if (module->isProcessEntryAttributes() && (option || (!preverse))) {
if (preverse) {
sprintf(buf, "%i", pvHeaderNum++);
module->getEntryAttributes()["Heading"]["Preverse"][buf] = header;
}
else {
sprintf(buf, "%i", headerNum++);
module->getEntryAttributes()["Heading"]["Interverse"][buf] = header;
if (option) { // we want the tag in the text
text.append(header);
}
}
StringList attributes = startTag.getAttributeNames();
for (StringList::const_iterator it = attributes.begin(); it != attributes.end(); it++) {
module->getEntryAttributes()["Heading"][buf][it->c_str()] = startTag.getAttribute(it->c_str());
}
}
hide = false;
if (!option || preverse) { // we don't want the tag in the text anymore
preverse = false;
continue;
}
preverse = false;
}
if (tag.getAttribute("class") && ((!stricmp(tag.getAttribute("class"), "sechead"))
|| (!stricmp(tag.getAttribute("class"), "title")))) {
isheader = true;
if (!tag.isEndTag()) { //start tag
if (!tag.isEmpty()) {
startTag = tag;
/* how do we tell a ThML preverse title from one that should be in the text? probably if any text is before the title... just assuming all are preverse for now
}
if (tag.getAttribute("subtype") && !stricmp(tag.getAttribute("subtype"), "x-preverse")) {
*/
hide = true;
preverse = true;
header = "";
continue;
} // move back up under startTag = tag
}
/* this is where non-preverse will go eventually
if (!tag.isEndTag()) { //start tag
hide = true;
header = "";
if (option) { // we want the tag in the text
text.append('<');
text.append(token);
text.append('>');
}
continue;
}
*/
}
else
isheader = false;
}
if (withinDiv && isheader) {
header.append('<');
header.append(token);
header.append('>');
} else {
// if not a heading token, keep token in text
if (!hide) {
text.append('<');
text.append(token);
text.append('>');
}
}
continue;
}
if (intoken) { //copy token
token.append(*from);
}
else if (!hide) { //copy text which is not inside a token
text.append(*from);
}
else header.append(*from);
}
return 0;
}
// Return true if the content was handled or is to be ignored.
// false if the what has been seen is to be accumulated and considered later.
bool handleToken(std::string & text, XMLTag & token) {
// The start token for the current entry;
static XMLTag startTag;
// Flags to indicate whether we are in a entry, entryFree or superentry
static bool inEntry = false;
static bool inEntryFree = false;
static bool inSuperEntry = false;
std::string const & tokenName = token.name();
static char const * splitPtr;
static char const * splitPtr2 = nullptr;
static std::array<char, 4096> splitBuffer;
static SWKey tmpKey;
//-- START TAG -------------------------------------------------------------------------
if (!token.isEndTag()) {
// If we are not in an "entry" and we see one, then enter it.
if (!inEntry && !inEntryFree && !inSuperEntry) {
inEntry = (tokenName == "entry");
inEntryFree = (tokenName == "entryFree");
inSuperEntry = (tokenName == "superentry");
if (inEntry || inEntryFree || inSuperEntry) {
#ifdef DEBUG
cout << "Entering " << tokenName << endl;
#endif
startTag = token;
text = "";
keyStr = token.attribute("n"); // P5 with linking and/or non-URI chars
if (keyStr.empty()) {
keyStr = token.attribute("sortKey"); // P5 otherwise
if (keyStr.empty()) {
keyStr = token.attribute("key"); // P4
}
}
return false; // make tag be part of the output
}
}
}
//-- EMPTY and END TAG ---------------------------------------------------------------------------------------------
else {
// ENTRY end
// If we see the end of an entry that we are in, then leave it
if ((inEntry && (tokenName == "entry" )) ||
(inEntryFree && (tokenName == "entryFree" )) ||
(inSuperEntry && (tokenName == "superentry"))) {
#ifdef DEBUG
cout << "Leaving " << tokenName << endl;
#endif
// Only one is false coming into here,
// but all must be on leaving.
inEntry = false;
inEntryFree = false;
inSuperEntry = false;
text += token.toString();
entryCount++;
#ifdef DEBUG
cout << "keyStr: " << keyStr << endl;
#endif
splitPtr = std::strchr(keyStr.c_str(), '|');
if (splitPtr) {
std::strncpy (splitBuffer.data(), keyStr.c_str(), splitPtr - keyStr.c_str());
splitBuffer[splitPtr - keyStr.c_str()] = 0;
currentKey->setText(splitBuffer.data());
#ifdef DEBUG
cout << "splitBuffer: " << splitBuffer.data() << endl;
cout << "currentKey: " << currentKey->getText() << endl;
#endif
writeEntry(*currentKey, text);
#if 1
while (splitPtr) {
splitPtr += 1;
splitPtr2 = std::strstr(splitPtr, "|");
entryCount++;
if (splitPtr2) {
std::strncpy (splitBuffer.data(), splitPtr, splitPtr2 - splitPtr);
splitBuffer[splitPtr2 - splitPtr] = 0;
#ifdef DEBUG
cout << "splitBuffer: " << splitBuffer.data() << endl;
cout << "currentKey: " << currentKey->getText() << endl;
#endif
linkToEntry(currentKey->getText(), splitBuffer.data());
splitPtr = splitPtr2;
}
else {
std::strcpy(splitBuffer.data(), splitPtr);
#ifdef DEBUG
cout << "splitBuffer: " << splitBuffer.data() << endl;
cout << "currentKey: " << currentKey->getText() << endl;
#endif
linkToEntry(currentKey->getText(), splitBuffer.data());
splitPtr = nullptr;
}
}
#endif
}
else {
currentKey->setText(keyStr);
writeEntry(*currentKey, text);
}
// Since we consumed the text, clear it
// and tell the caller that the tag was consumed.
text = "";
return true;
}
}
return false;
}
char OSISMorphSegmentation::processText(SWBuf &text, const SWKey * /*key*/, const SWModule *module) {
SWBuf token;
bool intoken = false;
bool hide = false;
SWBuf orig( text );
const char *from = orig.c_str();
XMLTag tag;
SWBuf tagText = "";
unsigned int morphemeNum = 0;
bool inMorpheme = false;
SWBuf buf;
for (text = ""; *from; ++from) {
if (*from == '<') {
intoken = true;
token = "";
continue;
}
if (*from == '>') { // process tokens
intoken = false;
if (!strncmp(token.c_str(), "seg ", 4) || !strncmp(token.c_str(), "/seg", 4)) {
tag = token;
if (!tag.isEndTag() && tag.getAttribute("type") && !strcmp("morph", tag.getAttribute("type"))) { //<seg type="morph"> start tag
hide = !option; //only hide if option is Off
tagText = "";
inMorpheme = true;
}
if (tag.isEndTag()) {
buf.setFormatted("%.3d", morphemeNum++);
module->getEntryAttributes()["Morpheme"][buf]["body"] = tagText;
inMorpheme = false;
}
if (hide) { //hides start and end tags as long as hide is set
if (tag.isEndTag()) { //</seg>
hide = false;
}
continue; //leave out the current token
}
} //end of seg tag handling
text.append('<');
text.append(token);
text.append('>');
if (inMorpheme) {
tagText.append('<');
tagText.append(token);
tagText.append('>');
}
hide = false;
continue;
} //end of intoken part
if (intoken) { //copy token
token.append(*from);
}
else { //copy text which is not inside of a tag
text.append(*from);
if (inMorpheme) {
tagText.append(*from);
}
}
}
return 0;
}
