bool KeywordFinder::findNextElement(const std::string& title, Element& foundElement) const {
int foundPower = 0;
std::string::size_type keywordsSize = 0;
std::string::size_type index;
if(title.size() < min_title_length)
{
return false;
}
std::string clean_title(title);
clean_text(clean_title);
for (TokenArray::const_iterator itKeywords = tokens.begin(); itKeywords != tokens.end(); ++itKeywords) {
if ((index = clean_title.find(*itKeywords)) != std::string::npos
&& points_to_word(clean_title, index, itKeywords->size())) {
foundPower++;
keywordsSize += itKeywords->length();
}
}
if (foundPower > 0 && keywordsSize >= min_title_length) {
const std::string::size_type max_length = myKeyword.length() < title.length() ? title.length() : myKeyword.length();
keywordsSize = 100 - max_length + keywordsSize + foundPower - 1;
if (foundElement.power < foundPower
|| (foundElement.power == foundPower && keywordsSize > foundElement.KeywordsDiff)) {
foundElement.power = foundPower;
foundElement.KeywordsDiff = keywordsSize;
return true;
}
}
return false;
}
static gchar*
get_text_for_node(CongNodePtr node)
{
CongNodeType node_type;
const gchar *colour_string;
const gchar *string_colour_string;
gchar *text = NULL;
gchar *cleaned_text = NULL;
node_type = cong_node_type(node);
colour_string = cong_ui_get_colour_string(node_type);
string_colour_string = "#00FF00"; /* FIXME: get this from the theme and/or GtkSourceView settings */
switch (node_type) {
default: g_assert_not_reached();
case CONG_NODE_TYPE_UNKNOWN:
text = g_strdup(_("UNKNOWN"));
break;
case CONG_NODE_TYPE_ELEMENT:
{
xmlAttrPtr attr_iter;
const gchar* attname_col = cong_ui_get_colour_string(CONG_NODE_TYPE_ATTRIBUTE);
const gchar* attval_col = string_colour_string;
text = g_strdup_printf ("<span foreground=\"%s\"><%s",
colour_string,
cong_node_get_qualified_name (node));
/* Add attributes (if any): */
for (attr_iter = node->properties; attr_iter; attr_iter=attr_iter->next) {
gchar *attribute = g_strdup_printf(" <span foreground=\"%s\">%s=<span foreground=\"%s\">\"%s\"</span></span>",attname_col, attr_iter->name, attval_col, attr_iter->children->content);
cong_util_append(&text, attribute);
g_free(attribute);
}
cong_util_append(&text, "></span>");
}
break;
case CONG_NODE_TYPE_ATTRIBUTE:
text = g_strdup_printf("ATTRIBUTE");
break;
case CONG_NODE_TYPE_TEXT:
cleaned_text = clean_text((const gchar*)node->content);
text = g_strdup_printf("%s \"<span foreground=\"%s\">%s</span>\"",
_("Text:"), colour_string, cleaned_text);
g_free(cleaned_text);
break;
case CONG_NODE_TYPE_CDATA_SECTION:
cleaned_text = clean_text((const gchar*)node->content);
text = g_strdup_printf("%s \"<span foreground=\"%s\">%s</span>\"",
_("CDATA:"), colour_string, cleaned_text);
g_free(cleaned_text);
break;
case CONG_NODE_TYPE_ENTITY_REF:
text = g_strdup_printf("ENTITY REF: %s", node->name);
break;
case CONG_NODE_TYPE_ENTITY_NODE:
text = g_strdup_printf("ENTITY");
break;
case CONG_NODE_TYPE_PI:
text = g_strdup_printf("PI");
break;
case CONG_NODE_TYPE_COMMENT:
cleaned_text = clean_text((const gchar*)node->content);
text = g_strdup_printf("<span foreground=\"%s\"><!-- %s --></span>", colour_string, cleaned_text);
g_free(cleaned_text);
break;
case CONG_NODE_TYPE_DOCUMENT:
{
xmlDocPtr doc = (xmlDocPtr)node;
text = g_strdup_printf("<span foreground=\"%s\"><?xml version="
"<span foreground=\"%s\">\"%s\"</span>"
" encoding="
"<span foreground=\"%s\">\"%s\"</span>"
" standalone="
"<span foreground=\"%s\">\"%s\"</span>"
"></span>",
colour_string,
string_colour_string,
doc->version,
string_colour_string,
doc->encoding,
string_colour_string,
(doc->standalone?"yes":"no"));
}
break;
case CONG_NODE_TYPE_DOCUMENT_TYPE:
text = g_strdup_printf("DOCUMENT TYPE");
break;
case CONG_NODE_TYPE_DOCUMENT_FRAG:
text = g_strdup_printf("DOCUMENT FRAG");
break;
case CONG_NODE_TYPE_NOTATION:
text = g_strdup_printf("NOTATION");
break;
case CONG_NODE_TYPE_HTML_DOCUMENT:
text = g_strdup_printf("HTML DOCUMENT");
break;
case CONG_NODE_TYPE_DTD:
{
text = g_strdup_printf("<!DOCTYPE %s>", node->name);
/* FIXME: show any SYSTEM/PUBLIC DTD stuff */
}
break;
case CONG_NODE_TYPE_ELEMENT_DECL:
text = g_strdup_printf("ELEMENT DECL");
break;
case CONG_NODE_TYPE_ATRRIBUTE_DECL:
text = g_strdup_printf("ATTRIBUTE DECL");
break;
case CONG_NODE_TYPE_ENTITY_DECL:
text = g_strdup_printf("<span><!ENTITY %s ...></span>", node->name);
break;
case CONG_NODE_TYPE_NAMESPACE_DECL:
text = g_strdup_printf("NAMESPACE DECL");
break;
case CONG_NODE_TYPE_XINCLUDE_START:
text = g_strdup_printf("XINCLUDE START");
break;
case CONG_NODE_TYPE_XINCLUDE_END:
text = g_strdup_printf("XINCLUDE END");
break;
}
return (text);
}
void KeywordFinder::splitToTokens(std::string& keyword) {
clean_text(keyword);
std::transform(keyword.begin(), keyword.end(), keyword.begin(), ::tolower);
tokens = split(keyword.c_str(), ' ');
}
bool check_revision(QString id , int revision_number)
{
int pageid;
QEventLoop eventLoop;
// "quit()" the event-loop, when the network request "finished()"
QNetworkAccessManager mg;
QObject::connect(&mg, SIGNAL(finished(QNetworkReply*)), &eventLoop, SLOT(quit()));
// the HTTP request
QString url = "http://en.wikitolearn.org/api.php?action=parse&pageid="+id+"&format=json";
QNetworkRequest re( ( url ) );
QNetworkReply *reply = mg.get(re);
eventLoop.exec();
if (reply->error() == QNetworkReply::NoError) {
//success
//qDebug() << "Success" <<reply->readAll();
QString html = (QString)reply->readAll();
QJsonDocument jsonResponse = QJsonDocument::fromJson(html.toUtf8());
QJsonObject jsonObj = jsonResponse.object();
int revid = jsonObj["parse"].toObject()["revid"].toInt();
pageid = jsonObj["parse"].toObject()["pageid"].toInt();
QString page_title = jsonObj["parse"].toObject()["title"].toString();
qDebug() << jsonObj["parse"].toObject()["title"].toString();
if(revision_number == revid)
{
delete reply;
return true ;
}
else
{
qDebug() << "update page";
QString text = jsonObj["parse"].toObject()["text"].toObject()["*"].toString();
text = clean_text(text);
bool del = del_from_db(id,revid);
//check if deletion was successfull
if(del == true)
{
qDebug() << "deletion from DB done";
}
else
{
qDebug() << "error in deletion from DB";
}
QString pid = QString::number(pageid);
del_file(pid);
save_file( text , pageid , revision_number , page_title);
}
}
return true;
}
QString dbmanager::add(QString p_url)
{
QString text , page_title ;
int pageid , revid;
QString requested_url = "http://en.wikitolearn.org/api.php?action=parse&page="+p_url+"&format=json";
// create custom temporary event loop on stack
QEventLoop eventLoop;
// "quit()" the event-loop, when the network request "finished()"
QNetworkAccessManager mgr;
QObject::connect(&mgr, SIGNAL(finished(QNetworkReply*)), &eventLoop, SLOT(quit()));
// the HTTP request
QNetworkRequest req( requested_url );
QNetworkReply *reply = mgr.get(req);
eventLoop.exec();
if (reply->error() == QNetworkReply::NoError) {
//success
//qDebug() << "Success" <<reply->readAll();
QString html = (QString)reply->readAll();
QJsonDocument jsonResponse = QJsonDocument::fromJson(html.toUtf8());
QJsonObject jsonObj = jsonResponse.object();
/********************************************************************************
* API returns JSON . JSON needs to be parsed to get the contents from it *
* contents that we need are html contents , pageid , revision number of page *
* text == contains the html content of the specified page *
* pageid == contains the page id of the page . *
* revid == contains revision number of the page *
* page_title == contains the title of the page *
* ******************************************************************************
*/
text = jsonObj["parse"].toObject()["text"].toObject()["*"].toString();
pageid = jsonObj["parse"].toObject()["pageid"].toInt();
revid = jsonObj["parse"].toObject()["revid"].toInt();
page_title = jsonObj["parse"].toObject()["title"].toString();
//clean the result from the API
text = clean_text(text);
// qDebug() << text;
qDebug() <<pageid;
delete reply;
}
else {
//failure
qDebug() << "Failure" <<reply->errorString();
delete reply;
}
// ******************* save file here ****************
save_file(text , pageid , revid , page_title);
static auto i = 0;
return QString("%1: %2").arg(++i).arg(p_url);
}
