void TCpDoc::SaveAcmTechNewsToCpd(
const TStr& InFPath, const TStr& OutCpdFNm, const int& MxDocs){
// create output file
PSOut SOut=TFOut::New(OutCpdFNm);
// processing xml files
TFFile FFile(TStrV()+InFPath, TStrV()+".Html"+".Htm", "", true);
TStr FNm; int Docs=0;
while (FFile.Next(FNm)){
printf("Processing file '%s'\r", FNm.CStr());
PSIn SIn=TFIn::New(FNm);
THtmlLx Lx(SIn);
while (Lx.GetSym()!=hsyEof){
//printf("%d\r", Docs);
if ((MxDocs!=-1)&&(Docs>MxDocs)){break;}
Lx.MoveToBTagOrEof("<SPAN>");
if (Lx.GetArg("CLASS")!="title"){continue;}
Lx.MoveToBTagOrEof("<A>");
TStr TitleStr=Lx.GetStrToETag("<A>", false); TitleStr="";
Lx.MoveToETagOrEof("<SPAN>");
Lx.MoveToBTagOrEof("<P>");
TStr ParStr=Lx.GetStrToETag("<P>", false);
if (!ParStr.Empty()){
Docs++;
PCpDoc CpDoc=TCpDoc::New(TInt::GetStr(Docs), TitleStr, ParStr);
CpDoc->Save(*SOut);
}
}
}
printf("\n");
}
PNGramBs TNGramBs::GetNGramBsFromReuters21578(
const TStr& FPath, const int& MxDocs,
const int& MxNGramLen, const int& MnNGramFq,
const PSwSet& SwSet, const PStemmer& Stemmer){
// create n-gram-base
/* printf("Generating frequent n-grams (MaxLen:%d MinFq:%d) ...\n", MxNGramLen, MnNGramFq); */
PNGramBs NGramBs=TNGramBs::New(MxNGramLen, MnNGramFq, SwSet, Stemmer);
// interations over document-set
while (!NGramBs->IsFinished()){
TFFile FFile(FPath, ".SGM", false); TStr FNm; int Docs=0;
while (FFile.Next(FNm)){
/* printf("Processing file '%s'\n", FNm.CStr()); */
TXmlDocV LDocV; TXmlDoc::LoadTxt(FNm, LDocV);
for (int LDocN=0; LDocN<LDocV.Len(); LDocN++){
Docs++;
/* printf(" Pass %2d: %6d Docs\r", NGramBs->GetPassN(), Docs); */
if ((MxDocs!=-1)&&(Docs>=MxDocs)){break;}
PXmlDoc Doc=LDocV[LDocN];
PXmlTok DocTok=Doc->GetTok();
// get html string
PXmlTok Tok=Doc->GetTagTok("REUTERS|TEXT");
TStr HtmlStr=Tok->GetTokStr(false);
// extract words & update n-gram-base
_UpdateNGramBsFromHtmlStr(NGramBs, HtmlStr, SwSet, Stemmer);
}
if ((MxDocs!=-1)&&(Docs>=MxDocs)){break;}
}
NGramBs->ConcPass();
/* printf(" Pass %2d: %6d Docs\r", NGramBs->GetPassN(), Docs); */
}
/* printf("\nDone.\n"); */
// return
return NGramBs;
}
PNGramBs TNGramBs::GetNGramBsFromHtmlFPathV(
const TStr& FPath, const bool& RecurseDirP, const int& MxDocs,
const int& MxNGramLen, const int& MnNGramFq,
const PSwSet& SwSet, const PStemmer& Stemmer){
// create n-gram-base
/* printf("Generating frequent n-grams (MaxLen:%d MinFq:%d) ...\n", MxNGramLen, MnNGramFq); */
PNGramBs NGramBs=TNGramBs::New(MxNGramLen, MnNGramFq, SwSet, Stemmer);
// interations over document-set
while (!NGramBs->IsFinished()){
// prepare file-directory traversal
TFFile FFile(FPath, "", RecurseDirP);
// traverse files
TStr FNm; int Docs=0;
while (FFile.Next(FNm)){
Docs++; if ((MxDocs!=-1)&&(Docs>MxDocs)){break;}
/* if (Docs%100==0){ */
/* printf(" Pass %d: %6d\r", NGramBs->GetPassN(), Docs);} */
if (TFile::Exists(FNm)) {
// load html
TStr HtmlStr=TStr::LoadTxt(FNm);
// extract words & update ngram-base
_UpdateNGramBsFromHtmlStr(NGramBs, HtmlStr, SwSet, Stemmer);
}
}
NGramBs->ConcPass();
/* printf(" Pass %d: %6d\n", NGramBs->GetPassN()-1, Docs); */
}
/* printf("Done.\n"); */
// return
return NGramBs;
}
void TFile::DelWc(const TStr& WcStr, const bool& RecurseDirP){
// collect file-names
TStrV FNmV;
TFFile FFile(WcStr, RecurseDirP); TStr FNm;
while (FFile.Next(FNm)){
FNmV.Add(FNm);}
// delete files
for (int FNmN=0; FNmN<FNmV.Len(); FNmN++){
Del(FNmV[FNmN], false);}
}
/////////////////////////////////////////////////
// SkyGrid-Document
void TSkyGridBinDoc::SaveBinDocV(
const TStr& InXmlFPath, const TStr& OutBinFNm, const int& MxDocs){
printf("Processing SkyGrid-News-Xml files from '%s'...\n", InXmlFPath.CStr());
TFOut SOut(OutBinFNm);
TFFile FFile(InXmlFPath, true); TStr FNm;
int Docs=0; int DateDocs=0; uint64 PrevTm=0;
while (FFile.Next(FNm)){
if ((MxDocs!=-1)&&(Docs>=MxDocs)){break;}
//printf(" Processing '%s' ...", FNm.CStr());
PXmlDoc XmlDoc=TXmlDoc::LoadTxt(FNm);
PXmlTok ContentTok=XmlDoc->GetTagTok("item|content");
TStr SwIdStr=ContentTok->GetTagTok("swid")->GetArgVal("value");
TStr UrlStr=ContentTok->GetTagTok("url")->GetTokStr(false);
TStr TitleStr=ContentTok->GetTagTok("title")->GetTokStr(false);
TStr FetchedValStr=ContentTok->GetTagTok("fetched")->GetArgVal("value");
TXmlTokV EntityTokV; ContentTok->GetTagTokV("annotations|entity", EntityTokV);
TStr BodyStr=ContentTok->GetTagTok("body")->GetTokStr(false);
// extract date
TStr DateStr=SwIdStr.GetSubStr(0, 7);
TStr YearStr=DateStr.GetSubStr(0, 3);
TStr MonthStr=DateStr.GetSubStr(4, 5);
TStr DayStr=DateStr.GetSubStr(6, 7);
TTm DateTm(YearStr.GetInt(), MonthStr.GetInt(), DayStr.GetInt());
uint64 Tm=TTm::GetMSecsFromTm(DateTm);
// extract entities
TStrIntH EntNmToFqH;
for (int EntityTokN=0; EntityTokN<EntityTokV.Len(); EntityTokN++){
PXmlTok EntityTok=EntityTokV[EntityTokN];
if (!EntityTok->IsTag("entity")){continue;}
TStr CanonicalNm=EntityTok->GetArgVal("canonical", "");
TStr TextStr=EntityTok->GetArgVal("text", "");
TStr TypeNm=EntityTok->GetArgVal("type", "");
TStr EntNm=CanonicalNm.Empty() ? TextStr : CanonicalNm;
EntNmToFqH.AddDat(EntNm)++;
}
TIntStrPrV FqEntNmPrV; EntNmToFqH.GetDatKeyPrV(FqEntNmPrV); FqEntNmPrV.Sort(false);
// extract headline
TChA HeadlineChA=BodyStr.GetSubStr(0, 250);
while ((HeadlineChA.Len()>0)&&(HeadlineChA.LastCh()!=' ')){
HeadlineChA.Trunc(HeadlineChA.Len()-1);}
HeadlineChA+="...";
// create document
TSkyGridBinDoc Doc(SwIdStr, Tm, TitleStr, HeadlineChA, FqEntNmPrV);
// save document
Doc.Save(SOut);
// screen log
if (PrevTm!=Tm){
if (PrevTm!=0){printf("\n");}
PrevTm=Tm; DateDocs=0;
}
Docs++; DateDocs++;
printf(" %s [Day:%d / All:%d]\r", DateStr.CStr(), DateDocs, Docs);
}
printf("\nDone.\n");
}
PBowDocBs TBowFl::LoadReuters21578Txt(
const TStr& FPath, const int& MxDocs,
const TStr& SwSetTypeNm, const TStr& StemmerTypeNm,
const int& MxNGramLen, const int& MnNGramFq, const bool& SaveDocP,
const PNotify& Notify){
// prepare stop-words
PSwSet SwSet=TSwSet::GetSwSet(SwSetTypeNm);
// prepare stemmer
PStemmer Stemmer=TStemmer::GetStemmer(StemmerTypeNm);
// create ngrams
PNGramBs NGramBs;
if (!((MxNGramLen==1)&&(MnNGramFq==1))){
NGramBs=TNGramBs::GetNGramBsFromReuters21578(
FPath, MxDocs,
MxNGramLen, MnNGramFq, SwSet, Stemmer);
}
// create document-base
PBowDocBs BowDocBs=TBowDocBs::New(SwSet, Stemmer, NGramBs);
// traverse directory
TFFile FFile(FPath, ".SGM", false); TStr FNm; int Docs=0;
while (FFile.Next(FNm)){
printf("Processing file '%s'\n", FNm.CStr());
TIntH DocWIdToFqH(100);
TXmlDocV LDocV; TXmlDoc::LoadTxt(FNm, LDocV);
for (int LDocN=0; LDocN<LDocV.Len(); LDocN++){
Docs++; if (Docs%100==0){printf("%d\r", Docs);}
if ((MxDocs!=-1)&&(Docs>=MxDocs)){break;}
// get document-name
PXmlDoc Doc=LDocV[LDocN];
PXmlTok DocTok=Doc->GetTok();
TStr DocNm=DocTok->GetArgVal("NEWID");
// get document-categories
TStrV CatNmV;
TXmlTokV TopicsTokV; Doc->GetTagTokV("REUTERS|TOPICS|D", TopicsTokV);
for (int TopicsTokN=0; TopicsTokN<TopicsTokV.Len(); TopicsTokN++){
TStr CatNm=TopicsTokV[TopicsTokN]->GetTokStr(false);
CatNmV.Add(CatNm);
}
// get document-contents
PXmlTok DocStrTok=Doc->GetTagTok("REUTERS|TEXT");
TStr DocStr=DocStrTok->GetTokStr(false);
// add document to bow
int DId=BowDocBs->AddHtmlDoc(DocNm, CatNmV, DocStr, SaveDocP);
// train & test data
if ((DocTok->GetArgVal("LEWISSPLIT")=="TRAIN")&&(DocTok->GetArgVal("TOPICS")=="YES")){
BowDocBs->AddTrainDId(DId);}
if ((DocTok->GetArgVal("LEWISSPLIT")=="TEST")&&(DocTok->GetArgVal("TOPICS")=="YES")){
BowDocBs->AddTestDId(DId);}
}
if ((MxDocs!=-1)&&(Docs>=MxDocs)){break;}
}
// return results
BowDocBs->AssertOk();
return BowDocBs;
}
void TCpDoc::SaveReuters21578ToCpd(
const TStr& InFPath, const TStr& OutCpdFNm, const int& MxDocs){
// create output file
PSOut SOut=TFOut::New(OutCpdFNm);
// traverse directory with .sgm files
TFFile FFile(InFPath, ".SGM", false); TStr FNm; int Docs=0;
while (FFile.Next(FNm)){
printf("Processing file '%s'\n", FNm.CStr());
TXmlDocV LDocV; TXmlDoc::LoadTxt(FNm, LDocV);
for (int LDocN=0; LDocN<LDocV.Len(); LDocN++){
Docs++; if (Docs%100==0){printf("%d\r", Docs);}
if ((MxDocs!=-1)&&(Docs>MxDocs)){break;}
// create reuters document
PCpDoc CpDoc=TCpDoc::New();
// load xml document
PXmlDoc Doc=LDocV[LDocN];
PXmlTok DocTok=Doc->GetTok();
// document id
CpDoc->DocNm=DocTok->GetArgVal("NEWID");
// date
CpDoc->DateStr=Doc->GetTagTok("REUTERS|DATE")->GetTokStr(false);
// document title
PXmlTok TitleTok=Doc->GetTagTok("REUTERS|TEXT|TITLE");
if (!TitleTok.Empty()){
CpDoc->TitleStr=TitleTok->GetTokStr(false);}
// dateline
PXmlTok DatelineTok=Doc->GetTagTok("REUTERS|TEXT|DATELINE");
if (!DatelineTok.Empty()){
CpDoc->DatelineStr=DatelineTok->GetTokStr(false);}
// get text string
TStr TextStr;
PXmlTok BodyTok=Doc->GetTagTok("REUTERS|TEXT|BODY");
if (!BodyTok.Empty()){
TextStr=BodyTok->GetTokStr(false);
} else {
// if <BODY> doesn't exist, take the whole <TEXT>
PXmlTok TextTok=Doc->GetTagTok("REUTERS|TEXT");
if (!TextTok.Empty()){
TextStr=TextTok->GetTokStr(false);
}
}
CpDoc->ParStrV.Add(TextStr, 1);
// topic categories
TXmlTokV TopCatTokV; Doc->GetTagTokV("REUTERS|TOPICS|D", TopCatTokV);
for (int TokN=0; TokN<TopCatTokV.Len(); TokN++){
TStr CatNm=TopCatTokV[TokN]->GetTokStr(false);
CpDoc->TopCdNmV.Add(CatNm);
}
// save cpd document
CpDoc->Save(*SOut);
}
if ((MxDocs!=-1)&&(Docs>MxDocs)){break;}
}
}
void TFFile::GetFNmV(
const TStr& FPath, const TStrV& FExtV, const bool& RecurseP, TStrV& FNmV){
// prepare file-directory traversal
TStrV FPathV; FPathV.Add(FPath);
TFFile FFile(FPathV, FExtV, "", RecurseP); TStr FNm;
// traverse directory
FNmV.Clr();
while (FFile.Next(FNm)){
FNmV.Add(FNm);
}
}
PAlignPairBs TAlignPairBs::LoadAcXml(const TStr& FPath, const int& MxSents) {
// prepare base
PAlignPairBs AlignPairBs = TAlignPairBs::New();
// iterate over files and load language to language alignments
TFFile FFile(FPath, ".xml", false); TStr FNm;
while (FFile.Next(FNm)) {
PAlignPair AlignPair = TAlignPair::LoadAcXml(FNm, MxSents);
AlignPairBs->AddAlignPair(AlignPair);
}
// finish
return AlignPairBs;
}
void FData::NewDataFile(int _Type, int _Size)
{
/*
Cria um buffer para novo arquivo
*/
if(!ExistFileType(_Type))
{
Files.push_back(FFile());
Files[Files.size()-1].Type = _Type;
int Id = GetId(_Type);
Files[Id].Buffer = new char[_Size];
Files[Id].SizeBuffer = _Size;
}
UpdateMaxFileType();
DefSerieMain();
}
TDzsBs::TDzsBs(const TStr& FPath, const TStr& WebAlias):
TBook("Dzs-Base", "Dzs", "Lexicon"),
DzsBsDocV(), Bix(TBix::New()){
TStrV FPathV; FPathV.Add(FPath);
TStrV FExtV;
FExtV.Add("xml"); FExtV.Add("html"); FExtV.Add("htm"); FExtV.Add("txt");
TFFile FFile(FPathV, FExtV, true); TStr FNm; int FNmN=0;
while (FFile.Next(FNm)){
printf("%d\r", ++FNmN);
PDzsBsDoc DzsBsDoc;
if (FNm.GetFExt().GetUc()==".XML"){
PXmlDoc XmlDoc=TXmlDoc::LoadTxt(FNm);
DzsBsDoc=TDzsBsDoc::GetDzsBsDoc(FNm, XmlDoc, FPath, WebAlias);
} else {
PHtmlDoc HtmlDoc=THtmlDoc::LoadTxt(FNm, hdtAll, false);
DzsBsDoc=TDzsBsDoc::GetDzsBsDoc(FNm, HtmlDoc, FPath, WebAlias);
}
AddDoc(DzsBsDoc);
}
}
/////////////////////////////////////////////////
// BagOfWords-Files
void TBowFl::LoadHtmlTxt(
PBowDocBs BowDocBs, const TStr& FPath, TIntV& NewDIdV,
const bool& RecurseDirP, const int& MxDocs,
const bool& SaveDocP, const PNotify& Notify) {
// prepare file-directory traversal
TStr LcNrFPath=TStr::GetNrFPath(FPath).GetLc();
Notify->OnStatus("Creating Bow from file-path " + FPath + " ...");
TFFile FFile(FPath, "", RecurseDirP);
// traverse files
TStr FNm; int Docs=0; NewDIdV.Clr();
while (FFile.Next(FNm)){
Docs++; if ((MxDocs!=-1)&&(Docs>MxDocs)){break;}
Notify->OnStatus(TStr::Fmt("%d\r", Docs));
// prepare document-name
if (TFile::Exists(FNm)) { //B:
TStr DocNm=FNm.GetLc();
if (DocNm.IsPrefix(LcNrFPath)){
DocNm=DocNm.GetSubStr(LcNrFPath.Len(), DocNm.Len()-1);}
// categories
TStrV CatNmV; TStr CatNm;
if (DocNm.IsChIn('/')){
TStr Str; DocNm.SplitOnCh(CatNm, '/', Str);
} else if (DocNm.IsChIn('\\')){
TStr Str; DocNm.SplitOnCh(CatNm, '\\', Str);
}
if (!CatNm.Empty()){
CatNmV.Add(CatNm);}
// load document-content
TStr DocStr=TStr::LoadTxt(FNm);
// add document to bow
NewDIdV.Add(BowDocBs->AddHtmlDoc(DocNm, CatNmV, DocStr, SaveDocP));
}
}
Notify->OnStatus(TStr::Fmt("%d", Docs));
// return results
Notify->OnStatus("Done.");
BowDocBs->AssertOk();
}
void TCpDoc::SaveHtmlToCpd(
const TStr& InHtmlFPath, const TStr& OutCpdFNm,
const bool& /*RecurseDirP*/, const int& MxDocs){
// create output file
PSOut SOut=TFOut::New(OutCpdFNm);
// prepare file-directory traversal
TStrV FPathV; FPathV.Add(InHtmlFPath);
TStrV FExtV; FExtV.Add("html"); FExtV.Add("htm"); FExtV.Add("xml");
TFFile FFile(FPathV, FExtV, "", true); TStr FNm;
// traverse files
printf("Processing '%s' ...\n", InHtmlFPath.CStr());
int Docs=0;
while (FFile.Next(FNm)){
// get file-name
if ((MxDocs!=-1)&&(FFile.GetFNmN()>MxDocs)){break;}
Docs++; if (FFile.GetFNmN()%10==0){printf("%d\r", Docs);}
// load html
PCpDoc CpDoc=TCpDoc::LoadHtmlDoc(FNm);
// save cpd document
CpDoc->Save(*SOut);
}
printf("%d\nDone.\n", Docs);
}
PPlBs TPlBs::LoadTxtPl(const TStr& FNmWc){
PPlBs PlBs=TPlBs::New();
printf("Loading Prolog from '%s'...\n", FNmWc.CStr());
TFFile FFile(FNmWc); TStr FNm;
while (FFile.Next(FNm)){
printf(" ...loading '%s'\n", FNm.CStr());
// process prolog-file
PSIn SIn=TFIn::New(FNm);
TILx Lx(SIn, TFSet()|/*iloList|*/iloCmtAlw|iloCsSens|iloUniStr|iloExcept);
Lx.GetSym(TFSet(TupExpect)|syEof);
while (Lx.Sym!=syEof){
TPlVal TupVal=TPlBs::ParseTup(Lx, TFSet()|syPeriod, PlBs);
int FuncId=PlBs->TupV[TupVal.GetTupId()]->GetFuncId();
int Arity=PlBs->TupV[TupVal.GetTupId()]->GetVals();
PlBs->FuncIdArityPrToTupIdVH.
AddDat(TIntPr(FuncId, Arity)).Add(TupVal.GetTupId());
Lx.GetSym(TFSet(TupExpect)|syEof);
}
//break;
}
printf("Done.\n");
return PlBs;
}
void TCpDoc::SaveReuters2000ToCpd(
const TStr& InFPath, const TStr& OutCpdFNm, const int& MxDocs){
// create output file
PSOut SOut=TFOut::New(OutCpdFNm);
// processing xml files
TStrStrH DocIdToDateStrH;
TFFile FFile(InFPath, ".XML", true); TStr FNm;
while (FFile.Next(FNm)){
if ((MxDocs!=-1)&&(FFile.GetFNmN()>=MxDocs)){break;}
if ((1+FFile.GetFNmN())%100==0){
printf("Processing file '%s' (%d)\r", FNm.CStr(), 1+FFile.GetFNmN());}
PXmlDoc Doc=TXmlDoc::LoadTxt(FNm);
// get document
PCpDoc CpDoc=TCpDoc::New();
LoadReuters2000DocFromXml(FNm,
CpDoc->DocNm, CpDoc->DateStr, CpDoc->TitleStr,
CpDoc->HeadlineStr, CpDoc->BylineStr, CpDoc->DatelineStr,
CpDoc->ParStrV,
CpDoc->TopCdNmV, CpDoc->GeoCdNmV, CpDoc->IndCdNmV);
// save cpd document
CpDoc->Save(*SOut);
}
printf("\n");
}
void TCpDoc::SaveAsfaToCpd(const TStr& InFPath, const TStr& OutCpdFNm){
// create output file
PSOut SOut=TFOut::New(OutCpdFNm);
// traverse files
TStrH AccessionIdH;
TFFile FFile(TStr::GetNrFPath(InFPath)+"*.Asfa"); TStr AsfaFNm;
while (FFile.Next(AsfaFNm)){
printf("Processing file '%s'\n", AsfaFNm.CStr());
PSIn SIn=TFIn::New(AsfaFNm);
TILx Lx(SIn, TFSet(iloRetEoln, iloExcept));
Lx.GetSym(syLn, syEof);
while (Lx.Sym!=syEof){
// Query Line
TStr QueryLnStr=Lx.Str;
TStrV QueryStrV; QueryLnStr.SplitOnAllCh('\t', QueryStrV, false);
IAssert(QueryStrV[0]=="Query");
// RecordNo Line
Lx.GetSym(syLn); TStr RecNoLnStr=Lx.Str;
TStrV RecNoStrV; RecNoLnStr.SplitOnAllCh('\t', RecNoStrV, false);
IAssert(RecNoStrV[0]=="RecordNo");
//int RecN=RecNoStrV[1].GetInt();
// fields (format: Short-Name Tab Long-Name Tab Value-String)
TStr TitleStr, AbstractStr, PublicationYearStr, AccessionId;
TStrV AuthorNmV; TStrV TermNmV1, TermNmV2;
while (true){
Lx.GetSym(syLn); TStr FldLnStr=Lx.Str;
TStrV FldStrV; FldLnStr.SplitOnAllCh('\t', FldStrV, false);
if (FldStrV[0]=="----"){
if (!AccessionIdH.IsKey(AccessionId)){
AccessionIdH.AddKey(AccessionId);
// create & save cpd document
PCpDoc CpDoc=TCpDoc::New();
CpDoc->DocNm=AccessionId;
CpDoc->DateStr=PublicationYearStr;
CpDoc->TitleStr=TitleStr;
CpDoc->ParStrV.Add(AbstractStr);
CpDoc->TopCdNmV=TermNmV1;
CpDoc->GeoCdNmV=TermNmV2;
CpDoc->IndCdNmV=AuthorNmV;
CpDoc->Save(*SOut);
} else {/*printf("[%s]", AccessionId.CStr());*/}
break;
} else
if (FldStrV[0]=="TI"){
TitleStr=FldStrV[2];
} else if (FldStrV[0]=="TI"){
TitleStr=FldStrV[2];
} else if (FldStrV[0]=="AU"){
FldStrV[2].SplitOnAllCh(';', AuthorNmV);
for (int StrN=0; StrN<AuthorNmV.Len(); StrN++){AuthorNmV[StrN].ToTrunc();}
} else if (FldStrV[0]=="AB"){
AbstractStr=FldStrV[2];
} else if (FldStrV[0]=="PY"){
PublicationYearStr=FldStrV[2];
} else if (FldStrV[0]=="DE"){
FldStrV[2].SplitOnAllCh(';', TermNmV1);
for (int StrN=0; StrN<TermNmV1.Len(); StrN++){TermNmV1[StrN].ToTrunc();}
} else if (FldStrV[0]=="CL"){
FldStrV[2].SplitOnAllCh(';', TermNmV2);
for (int StrN=0; StrN<TermNmV2.Len(); StrN++){TermNmV2[StrN].ToTrunc();}
} else if (FldStrV[0]=="AN"){
AccessionId=FldStrV[2];
}
}
printf("%d\r", AccessionIdH.Len());
Lx.GetSym(syLn, syEof);
}
}
}
