unsigned int
ControlFile::findOnOffStatus(Logger::LogLevel level, const char *levelsString)
{
const char *name = Logger::levelName(level);
if (hasWord(name, levelsString)
|| (!hasNegWord(name, levelsString)
&& hasWord("all", levelsString))) {
return CHARS_TO_UINT(' ', ' ', 'O', 'N');
} else {
return CHARS_TO_UINT(' ', 'O', 'F', 'F');
}
}
set<WordScore> getBestKClustersForWord(string currentWord,
int maxAllotment, int minConnections, unsigned int maxFreq) {
set<WordScore> bestClusters;
for (map<string, set<WordScore> >::iterator i = clusters.begin();
i != clusters.end(); ++i) {
string topic = i->first;
set<WordScore> clusterWords = clusters[topic];
if (hasWord(clusterWords, currentWord)) {
continue;
}
unsigned int curFreq = getClusterFrequency(clusterWords);
unsigned int wordFreq = wfUtil->getFrequency(currentWord);
if (curFreq + wordFreq <= maxFreq) {
int numConnections = 0;
double clusterAffinity = getClusterAffinity(currentWord,
clusterWords, minConnections, numConnections);
if ((clusterAffinity/numConnections) >= affinityThreshold_) {
bestClusters.insert(WordScore(topic, clusterAffinity/numConnections));
if (bestClusters.size() > maxAllotment) {
bestClusters.erase(*(bestClusters.rbegin()));
}
}
}
}
return bestClusters;
}
bool hasWord(Trie trieToInspect, const char *wordToLookFor) {
D("looking for \"%s\"\n", wordToLookFor);
bool foundWord;
if(trieToInspect == NULL) {
// word doesn't exist
foundWord = FALSE;
} else if(wordToLookFor[FIRST_INDEX] == END_OF_STRING) {
D("string ended.\n");
if(trieToInspect->wordEndsHere) {
// word exists
foundWord = TRUE;
} else {
// reached the end of the string; still no success
// therefore the word doesn't exist
foundWord = FALSE;
}
} else {
// go down to the next level
foundWord = hasWord( getChild(trieToInspect, wordToLookFor[FIRST_INDEX]),
nextLetterInWord(wordToLookFor) );
}
return foundWord;
}
/* CENTRAL METHOD to parse and render html request*/
int handle(cchar* q0,int conn){
int len=(int)strlen(q0);
if(len>1000){
p("checkSanity len>1000");
return 0;// SAFETY!
}
char* q=editable(q0);
checkSanity(q,len);
while(q[0]=='/')q++;
enum result_format format = html;//txt; html DANGER WITH ROBOTS
enum result_verbosity verbosity = normal;
if (eq(q, "favicon.ico"))return 0;
if(contains(q,"robots.txt")){
Writeline(conn,"User-agent: *\n");
Writeline("Disallow: /\n");
return 0;
}
char* jsonp=strstr(q,"jsonp");// ?jsonp=fun
if(jsonp){
jsonp[-1]=0;
jsonp+=6;
format = json;
}
else jsonp=(char*)"parseResults";
if (endsWith(q, ".json")) {
format = json;
q[len-5]=0;
}
if (endsWith(q, ".xml")) {
format = xml;
q[len-4]=0;
}
if (endsWith(q, ".csv")||endsWith(q, ".tsv")) {
format = csv;
q[len-4]=0;
}
if (endsWith(q, ".txt")) {
format = txt;
q[len-4]=0;
}
if (endsWith(q, ".html")) {
format = html;
q[len-5]=0;
}
if (startsWith(q, ".js")) {
q[len-3]=0;
Writeline(conn, jsonp);
Writeline(conn, "(");
format = js;
}
// todo : dedup!!
if (startsWith(q, "all/")) {
cut_to(q," +");
cut_to(q," -");
q = q + 4;
showExcludes=false;
verbosity = alle;
}
if (startsWith(q, "long/")){
verbosity = longer;
q = q + 5;
}
if (startsWith(q, "full/")) {
verbosity = verbose;
q = q + 5;
}
if (startsWith(q, "verbose/")) {
verbosity = verbose;
q = q + 8;
}
if (startsWith(q, "short/")) {
verbosity = shorter;
q = q + 6;
}
if (startsWith(q, "html/")) {
format = html;
if(!contains(q,".")&&!contains(q,":"))
verbosity=verbose;
q = q + 5;
}
if (startsWith(q, "plain/")) {
format = txt;
q = q + 6;
}
if (startsWith(q, "text/")) {
format = txt;
q = q + 5;
}
if (startsWith(q, "txt/")) {
format = txt;
q = q + 4;
}
if (startsWith(q, "xml/")) {
format = xml;
q = q + 4;
}
if (startsWith(q, "csv/")||startsWith(q, "tsv/")) {
format = csv;
q = q + 4;
}
if (startsWith(q, "json/")) {
format = json;
q = q + 5;
}
if (startsWith(q, "js/")) {
q = q + 3;
Writeline(conn, jsonp);
Writeline(conn, "(");
format = js;
}
if (startsWith(q, "long/")) {
verbosity = longer;
q = q + 5;
}
if (startsWith(q, "verbose/")) {
verbosity = verbose;
q = q + 8;
}
if (startsWith(q, "short/")) {
verbosity = shorter;
q = q + 6;
}
if (startsWith(q, "excludes/")||startsWith(q, "includes/")||startsWith(q, "excluded/")||startsWith(q, "included/")||startsWith(q, "showview/")) {
showExcludes=true;
verbosity=longer;
q = q + 9;
}
else showExcludes=false;
excluded.clear();
included.clear();
if(contains(q,"statement count")){Writeline(conn,itoa((int)context->statementCount).data());return 0;}
if(contains(q,"node count")){Writeline(conn,itoa(context->nodeCount).data());return 0;}
if (startsWith(q, "all/")) {
cut_to(q," +");
cut_to(q," -");
q = q + 4;
showExcludes=false;
verbosity = alle;
}
// bool get_topic=false;
bool get_topic=true;
bool sort=false;
if (startsWith(q, "ee/")||startsWith(q, "ee ")) {
q[2]=' ';
get_topic=true;
}
if (startsWith(q, "entities/")) {
q[8]=' ';
get_topic=true;
// verbosity=longer;
}
if(hasWord(q)) loadView(q);
if(contains(q,"exclude")||contains(q,"include")){
verbosity=normal;
showExcludes=true;
}
p(q);
// !!!!!!!!!!!!!!!!!!!!!!!!!!!
//
NodeVector all = parse(q); // <<<<<<<< HANDLE QUERY WITH NETBASE!
//
// !!!!!!!!!!!!!!!!!!!!!!!!!!!
autoIds=false;
int size=(int)all.size();
if(showExcludes){
for (int i = 0; i < size; i++) {
// todo : own routine!!!
Node* node = (Node*) all[i];
if(!contains(all,getAbstract(node->name)))
all.push_back(getAbstract(node->name));
N parent= getType(node);
if(parent){
if(!contains(all,parent))all.push_back(parent);
N abs= getAbstract(parent->name);
if(parent&&!contains(all,abs))all.push_back(abs);
}
}
show(excluded);
}
const char* html_block="<!DOCTYPE html><html><head><META HTTP-EQUIV='CONTENT-TYPE' CONTENT='text/html; charset=UTF-8'/></head>"\
"<body><div id='netbase_results'></div>\n<script>var results=";
// if((int)all.size()==0)Writeline("0");
// Writeline(conn,q);
char buff[10000];
bool use_json= format == json || format == js || format == html;
if (format == xml && (startsWith(q,"select")||contains(q," where "))){Writeline(conn,query2(q));return 0;}
if (format == xml)Writeline(conn, "<results>\n");
if (format == html)Writeline(conn,html_block);
if (use_json)Writeline(conn, "{\"results\":[\n");
const char* statement_format_xml = " <statement id='%d' subject=\"%s\" predicate=\"%s\" object=\"%s\" sid='%d' pid='%d' oid='%d'/>\n";
const char* statement_format_text = " $%d %s %s %s %d->%d->%d\n";
const char* statement_format_json = " { \"id\":%d, \"subject\":\"%s\", \"predicate\":\"%s\", \"object\":\"%s\", \"sid\":%d, \"pid\":%d, \"oid\":%d}";
const char* statement_format_csv = "%d\t%s\t%s\t%s\t%d\t%d\t%d\n";
const char* statement_format = 0;
if (format == xml)statement_format = statement_format_xml;
if (format == html)statement_format = statement_format_json;
if (format == json)statement_format = statement_format_json;
if (format == txt)statement_format = statement_format_text;
if (format == csv)statement_format = statement_format_csv;
const char* entity_format = 0;
const char* entity_format_txt = "%s #%d (statements:%d) %s\n";
const char* entity_format_xml = "<entity name=\"%s\" id='%d' statementCount='%d' description='%s'>\n";
const char* entity_format_json = " {\"name\":\"%s\", \"id\":%d, \"statementCount\":%d, \"description\":\"%s\"";
const char* entity_format_csv = "%s\t%d\t%d\t%s\n";
if(all.size()==1)entity_format_csv = "";//statements!
if (format == xml)entity_format = entity_format_xml;
if (format == txt)entity_format = entity_format_txt;
if (format == csv)entity_format = entity_format_csv;
if (use_json) entity_format = entity_format_json;
Node* last=0;
warnings=0;
char* entity=0;
if(startsWith(q,"all")){
entity=(char*)cut_to(q," ");
entity=keep_to(entity,"limit");
}
sortNodes(all);
int count=(int)all.size();
int good=0;
for (int i = 0; i < count && i<resultLimit; i++) {
Node* node = (Node*) all[i];
if(!checkNode(node))continue;
if(node->id==0)continue;
if(last==node)continue;
if(eq(node->name,"◊"))continue;
last=node;
if(verbosity ==normal && entity&& eq(entity,node->name))continue;
char* text=getText(node);
// if(use_json && get_topic){
// if(empty(text))continue;//! no description = no entity? BAD for amazon etc
// if(isAbstract(node))continue;
// N t=getTopic(node);
// }
good++;
if (use_json)if(good>1)Writeline(conn, "},\n");
sprintf(buff, entity_format, node->name, node->id,node->statementCount,text);
Writeline(conn, buff);
// if(verbosity != alle && !get_topic)
// loadView(node);
bool got_topic=false;
if(use_json && get_topic){
N c=getClass(node);
N t=getTopic(node);
N ty=getType(node);
// if(!c)c=t;
if(!t)t=ty;
if(t==node)t=ty;
if(t!=Entity && checkNode(t)){
got_topic=true;
Writeline(conn, ",\n\t \"topicid\":"+itoa(t->id));
Writeline(conn, ", \"topic\":\""+string(t->name)+"\"");
}
if(c && c!=t){
Writeline(conn, ",\n\t \"classid\":"+itoa(c->id));
Writeline(conn, ", \"class\":\""+string(c->name)+"\"");
}
if(ty&& c!=ty && ty!=t){
Writeline(conn, ",\n\t \"typeid\":"+itoa(ty->id));
Writeline(conn, ", \"type\":\""+string(ty->name)+"\"");
}
}
if(use_json)// && (verbosity==verbose||verbosity==shorter))// lol // just name
Writeline(conn, ", \"kind\":"+itoa(node->kind));
if((use_json)&&!showExcludes&&node->statementCount>1 && getImage(node)!="")
Writeline(", \"image\":\""+replace_all(replace_all(getImage(node,150,/*thumb*/true),"'","%27"),"\"","%22")+"\"");
// if((use_json)&&getText(node)[0]!=0)
// Writeline(", \"description\":\""+string(getText(node))+"\"");
Statement* s = 0;
if (format==csv|| verbosity == verbose || verbosity == longer|| verbosity == alle || showExcludes || ( all.size() == 1 && !(verbosity == shorter))) {
int count=0;
// Writeline(",image:\""+getImage(node->name)+"\"");
if (use_json)Writeline(conn, ",\n\t \"statements\":[\n");
// sortStatements(
deque<Statement*> statements;// sort
while ((s = nextStatement(node, s))&&count++<lookupLimit){// resultLimit
if (!checkStatement(s))break;
// if(!got_topic &&( s->predicate==_Type|| s->predicate==_SuperClass)){
// addStatementToNode(node, s->id(), true);// next time
// }
if(get_topic &&!got_topic && verbosity != verbose && (s->predicate>100 || s->predicate<-100))
continue;// only important stuff here!
// filter statements
if(s->object==0)continue;
// if(eq(s->Predicate()->name,"Offizielle Website") && !contains(s->Object()->name,"www"))
// continue;
if (s->subject==node->id and s->predicate!=4)//_instance
statements.push_front(s);
else statements.push_back(s);
}
// if(get_topic && verbosity!=shorter){
// NV topics=getTopics(node);
// N s=topics[0];
// for (int j = 0; j < topics.size() && j<=resultLimit; j++) {
// N s=topics[j];
// Temporary statement (node,topic,s)
// statements.push_front(s);
// }
// }
int good=0;
for (int j = 0; j < statements.size() && j<=resultLimit; j++) {
s=statements.at(j);
// while ((s = nextStatement(node, s))&&count++<resultLimit) {
if(format==csv&&all.size()>1)break;// entities vs statements
p(s);
if(verbosity!=alle&&checkHideStatement(s)){warnings++;continue;}
fixLabels(s);
if(!(verbosity==verbose||verbosity==alle) && (s->Predicate()==Instance||s->Predicate()==Type))continue;
if(use_json && good>0)Writeline(conn, ",\n");
char* objectName=s->Object()->name;
if(s->Predicate()==Instance){
N type=findProperty(s->Object(),Type->name,0,50);
if( checkNode(type))
objectName=(char*)(concat(concat(objectName, ": "),type->name));
}
sprintf(buff, statement_format, s->id(), s->Subject()->name, s->Predicate()->name, objectName, s->Subject()->id, s->Predicate()->id, s->Object()->id);
Writeline(conn, buff);
good++;
}
if (use_json)Writeline(conn, "]");
}
if (format == xml)Writeline(conn, "</entity>\n");
// string img=getImage(node->name);
// if(img!="")Writeline(conn,"<img src=\""+img+"\"/>");
}
if (use_json || format == html || format == js)Writeline(conn,good>0?"}\n]}":"]}");
if (format == xml)Writeline(conn, "</results>\n");
if(format == js)Writeline(conn, ")");// jsonp
const char* html_end=";\n</script>\n<script src='http://pannous.net/netbase.js'></script></body></html>\n";
if(format == html)Writeline(conn, html_end);
// sprintf(buff, "<script src='/js/%s'></script>",q0);
// Writeline(conn, buff);
// }
pf("Warnings/excluded: %d\n",warnings);
return 0;// 0K
}
