TEST(TStr, ChangeChAll) {
TStr Str = "abac";
TStr Empty;
int Changes = 0;
Changes = Empty.ChangeChAll('a', 'c');
EXPECT_EQ(Empty, "");
EXPECT_EQ(Changes, 0);
Changes = Str.ChangeChAll('a', 'c');
EXPECT_EQ(Str, "cbcc");
EXPECT_EQ(Changes, 2);
}
// <last_name>_<first name innitial>
TStr TStrUtil::GetStdName(TStr AuthorName) {
TStr StdName;
AuthorName.ToLc();
AuthorName.ChangeChAll('\n', ' ');
AuthorName.ChangeChAll('.', ' ');
// if there is a number in the name, remove it and everything after it
int i, pos = 0;
while (pos<AuthorName.Len() && (AuthorName[pos]!='#' && !TCh::IsNum(AuthorName[pos]))) {
pos++;
}
if (pos < AuthorName.Len()) {
AuthorName = AuthorName.GetSubStr(0, pos-1).ToTrunc();
}
if (AuthorName.Empty()) {
return TStr::GetNullStr();
}
// replace everything after '('
int b = AuthorName.SearchCh('(');
if (b != -1) {
AuthorName = AuthorName.GetSubStr(0, b-1).ToTrunc();
}
// skip if contains ')'
if (AuthorName .SearchCh(')')!=-1) {
return TStr::GetNullStr();
}
// skip if it is not a name
if (AuthorName .SearchStr("figures")!=-1 || AuthorName .SearchStr("macros")!=-1
|| AuthorName .SearchStr("univ")!=-1 || AuthorName .SearchStr("institute")!=-1) {
return TStr::GetNullStr();
}
// remove all non-letters (latex tags, ...)
TChA NewName;
for (i = 0; i < AuthorName.Len(); i++) {
const char Ch = AuthorName[i];
if (TCh::IsAlpha(Ch) || TCh::IsWs(Ch) || Ch=='-') {
NewName += Ch;
}
}
StdName = NewName;
StdName.ToTrunc();
TStrV AuthNmV;
StdName.SplitOnWs(AuthNmV);
// too short -- not a name
if (! AuthNmV.Empty() && AuthNmV.Last() == "jr") AuthNmV.DelLast();
if (AuthNmV.Len() < 2) return TStr::GetNullStr();
const TStr LastNm = AuthNmV.Last();
if (! TCh::IsAlpha(LastNm[0]) || LastNm.Len() == 1) return TStr::GetNullStr();
IAssert(isalpha(AuthNmV[0][0]));
return TStr::Fmt("%s_%c", LastNm.CStr(), AuthNmV[0][0]);
}
void TStrUtil::GetStdNameV(TStr AuthorNames, TStrV& StdNameV) {
AuthorNames.ChangeChAll('\n', ' ');
AuthorNames.ToLc();
// split into author names
TStrV AuthV, TmpV, Tmp2V;
// split on 'and'
AuthorNames.SplitOnStr(" and ", TmpV);
int i;
for (i = 0; i < TmpV.Len(); i++) {
TmpV[i].SplitOnAllCh(',', Tmp2V); AuthV.AddV(Tmp2V); }
// split on '&'
TmpV = AuthV; AuthV.Clr();
for (i = 0; i < TmpV.Len(); i++) {
TmpV[i].SplitOnAllCh('&', Tmp2V); AuthV.AddV(Tmp2V); }
// split on ','
TmpV = AuthV; AuthV.Clr();
for (i = 0; i < TmpV.Len(); i++) {
TmpV[i].SplitOnAllCh(',', Tmp2V); AuthV.AddV(Tmp2V); }
// split on ';'
TmpV = AuthV; AuthV.Clr();
for (i = 0; i < TmpV.Len(); i++) {
TmpV[i].SplitOnAllCh(';', Tmp2V); AuthV.AddV(Tmp2V); }
// standardize names
StdNameV.Clr();
//printf("\n*** %s\n", AuthorNames.CStr());
for (i = 0; i < AuthV.Len(); i++) {
TStr StdName = GetStdName(AuthV[i]);
if (! StdName.Empty()) {
//printf("\t%s ==> %s\n", AuthV[i].CStr(), StdName.CStr());
StdNameV.Add(StdName);
}
}
}
/////////////////////////////////////////////////
// Google-Web-Context-Graph
PGgCtxGraph TGgCtxGraph::GetCtxGraph(const TStr& FocusUrlStr){
// create context-graph
PGgCtxGraph CtxGraph=TGgCtxGraph::New();
CtxGraph->Ok=false;
// get focus-web-page
CtxGraph->FocusUrlStr=FocusUrlStr;
{bool Ok; TStr MsgStr;
TWebFetchBlocking::GetWebPg(
CtxGraph->FocusUrlStr, Ok, MsgStr, CtxGraph->FocusWebPg, TNotify::StdNotify);
if (!Ok){return CtxGraph;}}
// get 'In' set of links
PRSet InRSet=
TGg::WebSearch(TStr("link:")+CtxGraph->FocusUrlStr, -1, TNotify::StdNotify);
for (int HitN=0; HitN<InRSet->GetHits(); HitN++){
TStr UrlStr; TStr TitleStr; TStr SrcNm; TStr CtxStr;
InRSet->GetHit(HitN, UrlStr, TitleStr, SrcNm, CtxStr);
TitleStr.ChangeChAll('\n', ' ');
CtxGraph->InUrlCtxStrPrV.Add(TStrPr(UrlStr, TitleStr));
}
// get Out set
TStrKdV OutDescUrlStrKdV;
CtxGraph->FocusWebPg->GetOutDescUrlStrKdV(OutDescUrlStrKdV);
for (int UrlN=0; UrlN<OutDescUrlStrKdV.Len(); UrlN++){
OutDescUrlStrKdV[UrlN].Key.ChangeChAll('\n', ' ');
CtxGraph->OutUrlCtxStrPrV.Add(
TStrPr(OutDescUrlStrKdV[UrlN].Dat, OutDescUrlStrKdV[UrlN].Key));
}
// set context-graph successful
CtxGraph->Ok=true;
// return context-graph
return CtxGraph;
}
void TSAppSrvFun::LogReqRes(const TStrKdV& FldNmValPrV, const PHttpResp& HttpResp)
{
if (LogRqFolder == "")
return;
try {
TDir::GenDir(LogRqFolder);
TStr TimeNow = TTm::GetCurLocTm().GetWebLogDateTimeStr(true);
TimeNow.ChangeChAll(':', '.');
PSOut Output = TFOut::New(LogRqFolder + "/" + TimeNow + ".txt", false);
Output->PutStr(FunNm.CStr()); Output->PutCh('\n');
for (int N=0; N < FldNmValPrV.Len(); N++)
Output->PutStrFmt(" %s: %s\n", FldNmValPrV[N].Key.CStr(), FldNmValPrV[N].Dat.CStr());
Output->PutCh('\n');
Output->PutStr(HttpResp->GetBodyAsStr(), false);
} catch (...) {
/*const PNotify& Notify = RqEnv->GetWebSrv()->GetNotify();
Notify->OnStatus("Unable to log request for function '" + GetFunNm() + "'!");*/
}
}
//////////////////////////////////////
// File-Download-Function
void TSASFunFile::LoadFunFileV(const TStr& FPath, TSAppSrvFunV& SrvFunV) {
TFFile File(FPath, "", false); TStr FNm;
while (File.Next(FNm)) {
TStr FExt = FNm.GetFExt();
TStr FUrl = FNm.GetSubStr(FPath.Len());
FUrl.ChangeChAll('\\', '/');
printf("%s %s %s\n", FNm.CStr(), FExt.CStr(), FUrl.CStr());
if (FExt == ".htm") { SrvFunV.Add(TSASFunFile::New(FUrl, FNm, THttp::TextHtmlFldVal)); }
else if (FExt == ".html") { SrvFunV.Add(TSASFunFile::New(FUrl, FNm, THttp::TextHtmlFldVal)); }
else if (FExt == ".js") { SrvFunV.Add(TSASFunFile::New(FUrl, FNm, THttp::TextJavaScriptFldVal)); }
else if (FExt == ".css") { SrvFunV.Add(TSASFunFile::New(FUrl, FNm, THttp::TextCssFldVal)); }
else if (FExt == ".jpg") { SrvFunV.Add(TSASFunFile::New(FUrl, FNm, THttp::ImageJpgFldVal)); }
else if (FExt == ".jpeg") { SrvFunV.Add(TSASFunFile::New(FUrl, FNm, THttp::ImageJpgFldVal)); }
else if (FExt == ".gif") { SrvFunV.Add(TSASFunFile::New(FUrl, FNm, THttp::ImageGifFldVal)); }
else {
printf("Unknown MIME type for extension '%s' for file '%s'", FExt.CStr(), FNm.CStr());
SrvFunV.Add(TSASFunFile::New(FUrl, FNm, THttp::AppOctetFldVal));
}
}
}
TFltRect TVizMapContext::PaintPointNm(PGks Gks, PVizMapPoint Point, const int& X,
const int& Y, const int& PointFontSize, const int& PointNmFontScale,
const bool& SelPointP, const bool& IsCatP) {
// get and clean point name
TStr PointNm = Point->GetPointNm();
PointNm.ChangeChAll('_', ' ');
if (PointNm.IsStrIn("[[")) {
const int StartPos = PointNm.SearchStr("[[");
PointNm = PointNm.Left(StartPos - 1);
}
// set font
TGksColor FontColor = SelPointP ? ColorSelPointFont : ColorPointFont;
const int FontSize = PointFontSize + TFlt::Round(Point->GetWgt()*PointNmFontScale);
//TFSet FontStyle = IsCatP ? (TFSet() | gfsBold) : TFSet();
//Gks->SetFont(TGksFont::New("ARIAL", FontSize, FontColor, FontStyle));
Gks->SetFont(TGksFont::New("ARIAL", FontSize, FontColor));
// refit it for the screen
TStr ScreenPointNm = Gks->BreakTxt(PointNm, " ", "", PointNmWidth, PointNmMxLines);
// calculate string position on the screen
const int HalfTxtWidth = Gks->GetTxtWidth(ScreenPointNm) / 2;
const int HalfTxtHeight = Gks->GetTxtHeight(ScreenPointNm) / 2;
// draw it!
const int MnX = X - HalfTxtWidth;
int CurrY = Y - HalfTxtHeight;
TStrV LineV; ScreenPointNm.SplitOnAllCh('\n', LineV);
for (int LineN = 0; LineN < LineV.Len(); LineN++) {
const int HalfLineWidth = Gks->GetTxtWidth(LineV[LineN]) / 2;
const int LineHeight = Gks->GetTxtHeight(LineV[LineN]);
Gks->PutTxt(LineV[LineN], MnX + (HalfTxtWidth - HalfLineWidth), CurrY);
CurrY += LineHeight-3;
}
// finish
return TFltRect(X - HalfTxtWidth, Y - HalfTxtHeight,
X + HalfTxtWidth, Y + HalfTxtHeight - LineV.Len()*3);
}
PLwOnto TLwOnto::LoadAsfaVoc(const TStr& FPath){
// normalize path
TStr NrFPath=TStr::GetNrFPath(FPath);
// create ontology
PLwOnto LwOnto=TLwOnto::New();
// create language object
int EnLangId=LwOnto->GetLangBs()->AddLang("EN", "English");
// create term-types
{PLwTermType D_TermType=TLwTermType::New(0, "Descriptor", EnLangId);
PLwTermType ND_TermType=TLwTermType::New(1, "Non-descriptor", EnLangId);
LwOnto->GetTermTypeBs()->AddTermType(D_TermType);
LwOnto->GetTermTypeBs()->AddTermType(ND_TermType);}
// create link-types
{PLwLinkType BT_LinkType=TLwLinkType::New(0, "BT", EnLangId, "Broader-Term");
PLwLinkType NT_LinkType=TLwLinkType::New(1, "NT", EnLangId, "Narrower-Term");
PLwLinkType RT_LinkType=TLwLinkType::New(2, "RT", EnLangId, "Related-Term");
PLwLinkType UF_LinkType=TLwLinkType::New(3, "UF", EnLangId, "Used-For");
PLwLinkType USE_LinkType=TLwLinkType::New(4, "USE", EnLangId, "Used-By");
LwOnto->GetLinkTypeBs()->AddLinkType(BT_LinkType);
LwOnto->GetLinkTypeBs()->AddLinkType(NT_LinkType);
LwOnto->GetLinkTypeBs()->AddLinkType(RT_LinkType);
LwOnto->GetLinkTypeBs()->AddLinkType(UF_LinkType);
LwOnto->GetLinkTypeBs()->AddLinkType(USE_LinkType);}
// load ontology file
TStr AsfaOntoFNm=NrFPath+"asfa_xml_20060522.xml";
printf("Loading '%s' ...", AsfaOntoFNm.CStr());
PXmlDoc AsfaXmlDoc=TXmlDoc::LoadTxt(AsfaOntoFNm);
IAssert(AsfaXmlDoc->IsOk());
TXmlTokV ConceptXmlTokV;
AsfaXmlDoc->GetTagTokV("THESAURUS|CONCEPT", ConceptXmlTokV);
printf(" Done.\n");
// create terms
{printf("Creating terms ...");
for (int ConceptN=0; ConceptN<ConceptXmlTokV.Len(); ConceptN++){
PXmlTok ConceptXmlTok=ConceptXmlTokV[ConceptN];
// term-name
TStr TermNm;
if (ConceptXmlTok->IsSubTag("NON-DESCRIPTOR")){
TermNm=ConceptXmlTok->GetTagTokStr("NON-DESCRIPTOR");}
else if (ConceptXmlTok->IsSubTag("DESCRIPTOR")){
TermNm=ConceptXmlTok->GetTagTokStr("DESCRIPTOR");}
// term-type
TStr TermTypeNm=ConceptXmlTok->GetTagTokStr("TYP");
int TermTypeId=LwOnto->GetTermTypeBs()->GetTermTypeId(TermTypeNm, EnLangId);
// description
TStr DescStr;
if (ConceptXmlTok->IsSubTag("SN")){
DescStr=ConceptXmlTok->GetTagTokStr("SN");
DescStr.ChangeChAll('\r', ' '); DescStr.ChangeChAll('\n', ' ');
DescStr.ChangeStrAll(" ", " "); DescStr.ToTrunc();
}
// create term
PLwTerm Term=TLwTerm::New(-1, TermNm, EnLangId, TermTypeId, DescStr);
LwOnto->GetTermBs()->AddTermGetTermId(Term);
}
printf(" Done. (%d)\n", LwOnto->GetTermBs()->GetTerms());}
// create links
{printf("Creating links ...");
for (int ConceptN=0; ConceptN<ConceptXmlTokV.Len(); ConceptN++){
PXmlTok ConceptXmlTok=ConceptXmlTokV[ConceptN];
// source-term-name
TStr TermNm1;
if (ConceptXmlTok->IsSubTag("NON-DESCRIPTOR")){
TermNm1=ConceptXmlTok->GetTagTokStr("NON-DESCRIPTOR");}
else if (ConceptXmlTok->IsSubTag("DESCRIPTOR")){
TermNm1=ConceptXmlTok->GetTagTokStr("DESCRIPTOR");}
int TermId1=LwOnto->GetTermBs()->GetTermId(TermNm1, EnLangId);
// links
for (int SubTokN=0; SubTokN<ConceptXmlTok->GetSubToks(); SubTokN++){
PXmlTok SubTok=ConceptXmlTok->GetSubTok(SubTokN);
if (SubTok->IsTag()){
TStr LinkTypeNm=SubTok->GetTagNm();
if (LwOnto->GetLinkTypeBs()->IsLinkType(LinkTypeNm, EnLangId)){
// destination-term-name
TStr TermNm2=ConceptXmlTok->GetTagTokStr(LinkTypeNm);
int TermId2=LwOnto->GetTermBs()->GetTermId(TermNm2, EnLangId);
int LinkTypeId=LwOnto->GetLinkTypeBs()->GetLinkTypeId(LinkTypeNm, EnLangId);
LwOnto->GetLinkBs()->AddLink(TermId1, LinkTypeId, TermId2);
}
}
}
}
printf(" Done. (%d)\n", LwOnto->GetLinkBs()->GetLinks());}
// return ontology
return LwOnto;
}
/// save graph into a gexf file which Gephi can read
void TAGMUtil::SaveGephi(const TStr& OutFNm, const PUNGraph& G, const TVec<TIntV>& CmtyVVAtr, const double MaxSz, const double MinSz, const TIntStrH& NIDNameH, const THash<TInt, TIntTr>& NIDColorH ) {
THash<TInt,TIntV> NIDComVHAtr;
TAGMUtil::GetNodeMembership(NIDComVHAtr, CmtyVVAtr);
FILE* F = fopen(OutFNm.CStr(), "wt");
fprintf(F, "<?xml version='1.0' encoding='UTF-8'?>\n");
fprintf(F, "<gexf xmlns='http://www.gexf.net/1.2draft' xmlns:viz='http://www.gexf.net/1.1draft/viz' xmlns:xsi='http://www.w3.org/2001/XMLSchema-instance' xsi:schemaLocation='http://www.gexf.net/1.2draft http://www.gexf.net/1.2draft/gexf.xsd' version='1.2'>\n");
fprintf(F, "\t<graph mode='static' defaultedgetype='undirected'>\n");
if (CmtyVVAtr.Len() > 0) {
fprintf(F, "\t<attributes class='node'>\n");
for (int c = 0; c < CmtyVVAtr.Len(); c++) {
fprintf(F, "\t\t<attribute id='%d' title='c%d' type='boolean'>", c, c);
fprintf(F, "\t\t<default>false</default>\n");
fprintf(F, "\t\t</attribute>\n");
}
fprintf(F, "\t</attributes>\n");
}
fprintf(F, "\t\t<nodes>\n");
for (TUNGraph::TNodeI NI = G->BegNI(); NI < G->EndNI(); NI++) {
int NID = NI.GetId();
TStr Label = NIDNameH.IsKey(NID)? NIDNameH.GetDat(NID): "";
Label.ChangeChAll('<', ' ');
Label.ChangeChAll('>', ' ');
Label.ChangeChAll('&', ' ');
Label.ChangeChAll('\'', ' ');
TIntTr Color = NIDColorH.IsKey(NID)? NIDColorH.GetDat(NID) : TIntTr(120, 120, 120);
double Size = MinSz;
double SizeStep = (MaxSz - MinSz) / (double) CmtyVVAtr.Len();
if (NIDComVHAtr.IsKey(NID)) {
Size = MinSz + SizeStep * (double) NIDComVHAtr.GetDat(NID).Len();
}
double Alpha = 1.0;
fprintf(F, "\t\t\t<node id='%d' label='%s'>\n", NID, Label.CStr());
fprintf(F, "\t\t\t\t<viz:color r='%d' g='%d' b='%d' a='%.1f'/>\n", Color.Val1.Val, Color.Val2.Val, Color.Val3.Val, Alpha);
fprintf(F, "\t\t\t\t<viz:size value='%.3f'/>\n", Size);
//specify attributes
if (NIDComVHAtr.IsKey(NID)) {
fprintf(F, "\t\t\t\t<attvalues>\n");
for (int c = 0; c < NIDComVHAtr.GetDat(NID).Len(); c++) {
int CID = NIDComVHAtr.GetDat(NID)[c];
fprintf(F, "\t\t\t\t\t<attvalue for='%d' value='true'/>\n", CID);
}
fprintf(F, "\t\t\t\t</attvalues>\n");
}
fprintf(F, "\t\t\t</node>\n");
}
fprintf(F, "\t\t</nodes>\n");
//plot edges
int EID = 0;
fprintf(F, "\t\t<edges>\n");
for (TUNGraph::TNodeI NI = G->BegNI(); NI < G->EndNI(); NI++) {
for (int e = 0; e < NI.GetOutDeg(); e++) {
if (NI.GetId() > NI.GetOutNId(e)) {
continue;
}
fprintf(F, "\t\t\t<edge id='%d' source='%d' target='%d'/>\n", EID++, NI.GetId(), NI.GetOutNId(e));
}
}
fprintf(F, "\t\t</edges>\n");
fprintf(F, "\t</graph>\n");
fprintf(F, "</gexf>\n");
fclose(F);
}