// parse 28/03/03 and 28-MAY-03 formats
TSecTm TSecTm::GetDtTmFromDmyStr(const TStr& DmyStr){
int DmyStrLen=DmyStr.Len();
// day
TChA ChA; int ChN=0;
while ((ChN<DmyStrLen)&&(DmyStr[ChN]!='/')&&(DmyStr[ChN]!='-')){
ChA+=DmyStr[ChN]; ChN++;}
TStr DayStr=ChA;
// month
ChA.Clr(); ChN++;
while ((ChN<DmyStrLen)&&(DmyStr[ChN]!='/')&&(DmyStr[ChN]!='-')){
ChA+=DmyStr[ChN]; ChN++;}
TStr MonthStr=ChA;
// year
ChA.Clr(); ChN++;
while (ChN<DmyStrLen){
ChA+=DmyStr[ChN]; ChN++;}
TStr YearStr=ChA;
// transform to numbers
int DayN=DayStr.GetInt(-1);
int MonthN=MonthStr.GetInt(-1);
int YearN=YearStr.GetInt(-1);
if (MonthN == -1){
MonthN = TTmInfo::GetMonthN(MonthStr.ToCap()); }
if ((DayN==-1)||(MonthN==-1)||(YearN==-1)){
return TSecTm();
} else {
if (YearN<1000){
if (YearN<70){YearN+=2000;} else {YearN+=1900;}}
// construct the date
return GetDtTm(YearN, MonthN, DayN);
}
return TSecTm();
}
int TDateWnd::GetFtr(const TTm& Val) const {
EAssert(InitP);
const uint ValUnit = TSecTm(Val).GetInUnits(TmUnit);
if (ValUnit < StartUnit) { return 0; }
if (ValUnit > EndUnit) { return (int)(EndUnit - StartUnit); }
return (int)(ValUnit - StartUnit);
}
TSecTm::TSecTm(const PXmlTok& XmlTok) {
const int Year = XmlTok->GetIntArgVal("Year");
const int Month = XmlTok->GetIntArgVal("Month");
const int Day = XmlTok->GetIntArgVal("Day");
const int Hour = XmlTok->GetIntArgVal("Hour");
const int Min = XmlTok->GetIntArgVal("Min");
const int Sec = XmlTok->GetIntArgVal("Sec");
AbsSecs = TSecTm(Year, Month, Day, Hour, Min, Sec).GetAbsSecs();
}
bool TDateWnd::Update(const TTm& Val) {
// if first time, use it to initialize
if (!InitP) {
StartUnit = TSecTm(Val).GetInUnits(TmUnit);
EndUnit = TSecTm(Val).GetInUnits(TmUnit);
InitP = true;
return true;
}
// check if we moved start or end boundary
const uint ValUnit = TSecTm(Val).GetInUnits(TmUnit);
if (StartUnit > ValUnit) {
StartUnit = ValUnit; return true;
}
if (EndUnit < ValUnit) {
EndUnit = ValUnit; return true;
}
// nope, we are fine
return false;
}
TFfGGen::TStopReason TFfGGen::GenGraph(const int& GraphNodes, PGStatVec& EvolStat, const bool& FloodStop) {
int GrowthStatNodes = 100;
Graph = PNGraph::New();
AddNodes(StartNodes);
TStopReason SR = srUndef;
while (Graph->GetNodes() < GraphNodes) {
SR = AddNodes(GrowthStatNodes, FloodStop);
if (SR != srOk) { return SR; }
EvolStat->Add(Graph, TSecTm(Graph->GetNodes()));
GrowthStatNodes = int(1.5*GrowthStatNodes);
}
return SR;
}
TSecTm TSecTm::GetCurTm(){
const time_t TmSec = time(NULL);
struct tm LocTm;
uint AbsSec = TUInt::Mx;
#if defined(GLib_MSN)
localtime_s(&LocTm, &TmSec);
#elif defined(GLib_BCB)
LocTm = *localtime(&TmSec);
#else
LocTm = *localtime(&TmSec);
#endif
IAssert(TSecTm::GetTmSec(LocTm, AbsSec));
return TSecTm(AbsSec);
}
// parse:
// 10:16, 16 Sep 2004
// 10:20, 2004 Sep 16
// 2005-07-07 20:30:35
// 23:24:07, 2005-07-10
// 9 July 2005 14:38
// 21:16, July 9, 2005
// 06:02, 10 July 2005
bool TStrUtil::GetTmFromStr(const char* TmStr, TSecTm& Tm) {
static TStrV MonthV1, MonthV2;
if (MonthV1.Empty()) {
TStr("january|february|march|april|may|june|july|august|september|october|november|december").SplitOnAllCh('|', MonthV1);
TStr("jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec").SplitOnAllCh('|', MonthV2);
}
TChA Tmp(TmStr);
Tmp.ToLc();
TVec<char *> WrdV;
const char* End = Tmp.CStr()+Tmp.Len();
int Col = -1, Cols=0;
for (char *b = Tmp.CStr(); b <End; ) {
WrdV.Add(b);
while (*b && ! (*b==' ' || *b=='-' || *b==':' || *b==',')) { b++; }
if (*b==':') { if(Col==-1) { Col=WrdV.Len(); } Cols++; }
*b=0; b++;
while (*b && (*b==' ' || *b=='-' || *b==':' || *b==',')) { b++; }
}
if (Cols == 2) {
if (Col+1 >= WrdV.Len()) { return false; }
WrdV.Del(Col+1);
}
if (Col<1) { return false; }
const int Hr = atoi(WrdV[Col-1]);
const int Min = atoi(WrdV[Col]);
WrdV.Del(Col); WrdV.Del(Col-1);
if (WrdV.Len() != 3) { return false; }
int y=0,m=1,d=2, Mon=-1;
if (TCh::IsAlpha(WrdV[0][0])) {
y=2; m=0; d=1;
} else if (TCh::IsAlpha(WrdV[1][0])) {
y=2; m=1; d=0;
} else if (TCh::IsAlpha(WrdV[2][0])) {
y=0; m=2; d=1;
} else {
y=0; m=1; d=2;
Mon = atoi(WrdV[m]);
}
int Day = atoi(WrdV[d]);
if (Mon <= 0) { Mon = MonthV1.SearchForw(WrdV[m])+1; }
if (Mon <= 0) { Mon = MonthV2.SearchForw(WrdV[m])+1; }
if (Mon == 0) { return false; }
int Year = atoi(WrdV[y]);
if (Day > Year) { ::Swap(Day, Year); }
//printf("%d-%02d-%02d %02d:%02d\n", Year, Mon, Day, Hr, Min);
Tm = TSecTm(Year, Mon, Day, Hr, Min, 0);
return true;
}
uint TSecTm::GetInUnits(const TTmUnit& TmUnit) const {
static const int DayZero = TJulianDate::GetJulianDateN(1, 1, 1970);
if (TmUnit == tmu1Sec) { return AbsSecs; }
struct tm Time;
IAssert(IsDef() && GetTmStruct(AbsSecs(), Time));
switch (TmUnit) {
case tmu1Min : return TSecTm(Time.tm_year+1900, Time.tm_mon+1, Time.tm_mday, Time.tm_hour, Time.tm_min, 0).GetAbsSecs()/60;
case tmu10Min : return TSecTm(Time.tm_year+1900, Time.tm_mon+1, Time.tm_mday, Time.tm_hour, 10*(Time.tm_min/10), 0).GetAbsSecs()/(10*60);
case tmu15Min : return TSecTm(Time.tm_year+1900, Time.tm_mon+1, Time.tm_mday, Time.tm_hour, 15*(Time.tm_min/15), 0).GetAbsSecs()/(15*60);
case tmu30Min : return TSecTm(Time.tm_year+1900, Time.tm_mon+1, Time.tm_mday, Time.tm_hour, 30*(Time.tm_min/30), 0).GetAbsSecs()/(30*60);
case tmu1Hour : return TSecTm(Time.tm_year+1900, Time.tm_mon+1, Time.tm_mday, Time.tm_hour, 0, 0).GetAbsSecs()/3600;
case tmu2Hour : return TSecTm(Time.tm_year+1900, Time.tm_mon+1, Time.tm_mday, 2*(Time.tm_hour/2), 0, 0).GetAbsSecs()/(2*3600);
case tmu4Hour : return TSecTm(Time.tm_year+1900, Time.tm_mon+1, Time.tm_mday, 4*(Time.tm_hour/4), 0, 0).GetAbsSecs()/(4*3600);
case tmu6Hour : return TSecTm(Time.tm_year+1900, Time.tm_mon+1, Time.tm_mday, 6*(Time.tm_hour/6), 0, 0).GetAbsSecs()/(6*3600);
case tmu12Hour : return TSecTm(Time.tm_year+1900, Time.tm_mon+1, Time.tm_mday, 12*(Time.tm_hour/12), 0, 0).GetAbsSecs()/(12*3600);
case tmuDay : return TJulianDate::GetJulianDateN(Time.tm_mday, Time.tm_mon+1, 1900+Time.tm_year) - DayZero;
case tmuWeek : return (TJulianDate::GetJulianDateN(Time.tm_mday, Time.tm_mon+1, 1900+Time.tm_year)-DayZero)/7;
case tmuMonth : return 12*(Time.tm_year-70)+Time.tm_mon+1;
case tmuYear : return Time.tm_year+1900;
default : Fail;
}
return TUInt::Mx;
}
// Parse strings of the form 2006-08-28 14:11:16 or 14:11:16 08/28/2008
// Non-numeric characters act as separators (there can be many consecutive separating characters)
// Variables give indexes of the date fields
TSecTm TSecTm::GetDtTmFromStr(const TChA& YmdHmsPmStr, const int& YearId, const int& MonId,
const int& DayId, const int& HourId, const int& MinId, const int& SecId) {
TChA Tmp = YmdHmsPmStr;
TVec<char *> FldV;
// get the sequences of numbers
for (char *c = (char *) Tmp.CStr(); *c; c++) {
if (TCh::IsNum(*c)) {
FldV.Add(c);
while (TCh::IsNum(*c)) { c++; }
c--;
} else { *c = 0; }
}
const int Y = atoi(FldV[YearId]);
const int M = atoi(FldV[MonId]);
const int D = atoi(FldV[DayId]);
const int H = atoi(FldV[HourId]);
const int m = atoi(FldV[MinId]);
const int S = atoi(FldV[SecId]);
IAssert(Y>0 && M>0 && D>0 && M<13 && D<32);
IAssert(H>=0 && H<24 && m>=0 && m<60 && S>=0 && S<60);
return TSecTm(Y,M,D,H,m,S);
}
int main(int argc, char* argv[]) {
// create a graph and save it
{ PNGraph Graph = TNGraph::New();
for (int i = 0; i < 10; i++) {
Graph->AddNode(i); }
for (int i = 0; i < 10; i++) {
Graph->AddEdge(i, TInt::Rnd.GetUniDevInt(10)); }
TSnap::SaveEdgeList(Graph, "graph.txt", "Edge list format"); }
// load a graph
PNGraph Graph;
Graph = TSnap::LoadEdgeList<PNGraph>("graph.txt", 0, 1);
// traverse nodes
for (TNGraph::TNodeI NI = Graph->BegNI(); NI < Graph->EndNI(); NI++) {
printf("NodeId: %d, InDegree: %d, OutDegree: %d\n", NI.GetId(), NI.GetInDeg(), NI.GetOutDeg());
printf("OutNodes: ");
for (int e = 0; e < NI.GetOutDeg(); e++) { printf(" %d", NI.GetOutNId(e)); }
printf("\nInNodes: ");
for (int e = 0; e < NI.GetInDeg(); e++) { printf(" %d", NI.GetInNId(e)); }
printf("\n\n");
}
// graph statistic
TSnap::PrintInfo(Graph, "Graph info");
PNGraph MxWcc = TSnap::GetMxWcc(Graph);
TSnap::PrintInfo(MxWcc, "Largest Weakly connected component");
// random graph
PNGraph RndGraph = TSnap::GenRndGnm<PNGraph>(100, 1000);
TGStat GraphStat(RndGraph, TSecTm(1), TGStat::AllStat(), "Gnm graph");
GraphStat.PlotAll("RndGraph", "Random graph on 1000 nodes");
// Forest Fire graph
{ TFfGGen ForestFire(false, 1, 0.35, 0.30, 1.0, 0.0, 0.0);
ForestFire.GenGraph(100);
PNGraph FfGraph = ForestFire.GetGraph(); }
// network
TPt<TNodeEDatNet<TStr, TStr> > Net = TNodeEDatNet<TStr, TStr>::New();
Net->AddNode(0, "zero");
Net->AddNode(1, "one");
Net->AddEdge(0, 1, "zero to one");
return 0;
}
int main(int argc, char* argv[])
{
TExeTm ExeTm;
THash< TStr , CascadeElementV > quotesFiltered;
double* vol_me;
uint period = 9 * 3600; // 9 days because of NIFTY paper
printf("((((( Starting The Filtering Cascades CODE )))))\n");
try
{
Env = TEnv(argc, argv, TNotify::StdNotify);
Env.PrepArgs(TStr::Fmt("\nFiltering Memes Cascades. build: %s, %s. Time: %s", __TIME__, __DATE__, TExeTm::GetCurTm()));
// ---== Loading Data ==---
TZipIn ZquotesIn("QuotesPreprocessedData_NIFTY.rar"); ///("/agbs/cluster/oaskaris/Data_Preparing_Codes/RESULTS/QuotesPreprocessedData_NIFTY.rar");
THash< TStr , CascadeElementV > quotes;
quotes.Load(ZquotesIn);
printf("Loaded QuotesPreprocessedData_NIFTY has instances: %d\n\n\n",quotes.Len());
// NIFTY Method for Filtering by Peaks
uint begin = TSecTm(2008,7,31,0,0,0).GetAbsSecs();
uint end = TSecTm(2009,10,1,0,0,0).GetAbsSecs();
TSecTmV memesTimes;
int bins = (end - begin) / period;
for(int c=0;c<quotes.Len();c++)
{
memesTimes.Clr();
for(int i=0;i<quotes[c].Len();i++)
{
memesTimes.Add(quotes[c][i].time);
}
vol_me = Tools::calculateHistOfCascade(memesTimes,begin,period,false);
// calculating mean and standard deviation
double mean = 0;
for(int i=0;i<bins;i++)
{
mean += vol_me[i];
}
mean /= bins;
double std = 0;
for(int i=0;i<bins;i++)
{
std += pow(vol_me[i]-mean , 2);
}
std = sqrt(std / (bins-1));
// peak definition by NIFTY: a point is a peak if its volume in 9 days binning is 1 standard deviation higher than the average frequency
double maxVolume = mean + std;
int peakCnt = 0;
for(int i=0;i<bins;i++)
{
if(vol_me[i] > maxVolume)
{
peakCnt++;
}
}
// if there is more than 5 peaks ignore this quote, since it is not a meme
if(peakCnt > 5)
{
delete[] vol_me;
continue;
}
quotesFiltered.AddDat(quotes.GetKey(c),quotes[c]);
delete[] vol_me;
}
TZipOut mout("QuotesPreprocessedData_NIFTY_FINALFILTERED.rar");
quotesFiltered.Save(mout);
printf("Saved QuotesPreprocessedData_NIFTY_FINALFILTERED has instances: %d\n\n\n",quotesFiltered.Len());
printf("\nThe Meme Filter for plotting had been done successfully.\n");
}
catch(exception& ex)
{
printf("\nError1 happened, it was: %s\n\n",ex.what());
}
catch(TPt<TExcept>& ex)
{
printf("\nError2 happened: %s\n\n",ex[0].GetStr().CStr());
}
printf("\nrun time: %s (%s)\n", ExeTm.GetTmStr(), TSecTm::GetCurTm().GetTmStr().CStr());
return 0;
}
TSecTm TTm::GetSecTmFromDateTimeInt(const uint& DateTimeInt) {
if (DateTimeInt == 0) { return TSecTm(); }
return TSecTm(DateTimeInt);
}
TSecTm::TSecTm(const TTm& Tm): AbsSecs(
TSecTm(Tm.GetYear(), Tm.GetMonth(), Tm.GetDay(), Tm.GetHour(),
Tm.GetMin(), Tm.GetSec()).GetAbsSecs()) { }
TSecTm TSecTm::LoadTxt(TILx& Lx){
return TSecTm(Lx.GetInt());
}
TSecTm TSecTm::GetDtTm(const int& YearN, const int& MonthN, const int& DayN){
uint AbsSecs;
TSecTm::GetTmSec(YearN, MonthN, DayN, 0, 0, 0, AbsSecs);
return TSecTm(AbsSecs);
}
TSecTm::TSecTm(const TTm& Tm): AbsSecs(
TSecTm(Tm.GetYear(), Tm.GetMonth(), Tm.GetDay(), Tm.GetHour(),
Tm.GetMin(), int(TMath::Round(Tm.GetSec()*1000+Tm.GetMSec()))).GetAbsSecs()){}
uint TTm::GetDateTimeInt(const int& Year, const int& Month,
const int& Day, const int& Hour, const int& Min, const int& Sec) {
return TSecTm(Year, Month, Day, Hour, Min, Sec).GetAbsSecs();
}
TSecTm TSecTm::Round(const TTmUnit& TmUnit) const {
if (TmUnit == tmu1Sec) { return *this; }
struct tm Time;
IAssert(IsDef() && GetTmStruct(AbsSecs(), Time));
switch (TmUnit) {
case tmu1Min : return TSecTm(Time.tm_year+1900, Time.tm_mon+1, Time.tm_mday, Time.tm_hour, Time.tm_min, 0);
case tmu10Min : return TSecTm(Time.tm_year+1900, Time.tm_mon+1, Time.tm_mday, Time.tm_hour, 10*(Time.tm_min/10), 0);
case tmu15Min : return TSecTm(Time.tm_year+1900, Time.tm_mon+1, Time.tm_mday, Time.tm_hour, 15*(Time.tm_min/15), 0);
case tmu30Min : return TSecTm(Time.tm_year+1900, Time.tm_mon+1, Time.tm_mday, Time.tm_hour, 30*(Time.tm_min/30), 0);
case tmu1Hour : return TSecTm(Time.tm_year+1900, Time.tm_mon+1, Time.tm_mday, Time.tm_hour, 0, 0);
case tmu2Hour : return TSecTm(Time.tm_year+1900, Time.tm_mon+1, Time.tm_mday, 2*(Time.tm_hour/2), 0, 0);
case tmu4Hour : return TSecTm(Time.tm_year+1900, Time.tm_mon+1, Time.tm_mday, 4*(Time.tm_hour/4), 0, 0);
case tmu6Hour : return TSecTm(Time.tm_year+1900, Time.tm_mon+1, Time.tm_mday, 6*(Time.tm_hour/6), 0, 0);
case tmu12Hour : return TSecTm(Time.tm_year+1900, Time.tm_mon+1, Time.tm_mday, 12*(Time.tm_hour/12), 0, 0);
case tmuDay : return TSecTm(Time.tm_year+1900, Time.tm_mon+1, Time.tm_mday, 0, 0, 0);
case tmuMonth : return TSecTm(Time.tm_year+1900, Time.tm_mon+1, 1, 0, 0, 0);
case tmuYear : return TSecTm(Time.tm_year+1900, 1, 1, 0, 0, 0);
case tmuWeek : { int dd=1, mm=1, yy=1;
// week starts on Thursday, since 1.1.1970 is Thursday
const int Day = TJulianDate::GetJulianDateN(Time.tm_mday, Time.tm_mon+1, 1900+Time.tm_year);
TJulianDate::GetCalendarDate(3+7*(Day/7), dd, mm, yy); return TSecTm(yy, mm, dd, 0, 0, 0); }
default : Fail;
}
return TSecTm();
}
