void CCharacterScanJob::start()
{
// open the output file for the character table
CSString filename= NLMISC::CPath::getFullPath(STAT_GLOBALS::getOutputFilePath(getOutputPath()))+_OutputName+"tbl.csv";
nlinfo("Opening output file: %s",filename.c_str());
_CharTblFile=fopen(filename.c_str(),"wb");
if (_CharTblFile==NULL)
{
nlwarning("Failed to open output file: %s",filename.c_str());
_State=ERROR;
return;
}
// write a header line to the _CharTblFile file
CSString rowTxt;
for (uint32 i=0;i<_TblCols.size();++i)
{
rowTxt+=_TblCols[i]+',';
}
fprintf(_CharTblFile,"%s\n",rowTxt.rightCrop(1).c_str());
fflush(_CharTblFile);
// setup the file list before the job begins
nlinfo("Scanning for input files...");
_Files.clear();
getFileList(_Files);
nlinfo("Input file scan completed: %d files found ... %s",_Files.size(),_Files.empty()?"Nothing to do!":"starting work");
// set the state to 'WORK' meaning we've finished init and now it's ok to do some work
_State=WORK;
}
/*
* Set the result vector with strings corresponding to the input string:
* - If inputStr is partially or completely found in the keys, all the matching <key,words> are returned;
* - If inputStr is partially or completely in the words, all the matching <key, words> are returned.
* The following tags can modify the behaviour of the search algorithm:
* - ^mystring returns mystring only if it is at the beginning of a key or word
* - mystring$ returns mystring only if it is at the end of a key or word
* All returned words are in UTF8.
*/
void CWordsDictionary::lookup( const CSString& inputStr, CVectorSString& resultVec ) const
{
// Prepare search string
if ( inputStr.empty() )
return;
CSString searchStr = inputStr;
bool findAtBeginning = false, findAtEnd = false;
if ( searchStr[0] == '^' )
{
searchStr = searchStr.substr( 1 );
findAtBeginning = true;
}
if ( searchStr[searchStr.size()-1] == '$' )
{
searchStr = searchStr.rightCrop( 1 );
findAtEnd = true;
}
// Search
const vector<string> &vec = reinterpret_cast<const vector<string>&>(_Keys);
// for ( CVectorSString::const_iterator ivs=_Keys.begin(); ivs!=_Keys.end(); ++ivs )
for ( vector<string>::const_iterator ivs=vec.begin(); ivs!=vec.end(); ++ivs )
{
const CSString& key = *ivs;
string::size_type p;
if ( (p = key.findNS( searchStr.c_str() )) != string::npos )
{
if ( ((!findAtBeginning) || (p==0)) && ((!findAtEnd) || (p==key.size()-searchStr.size())) )
resultVec.push_back( makeResult( key, _Words[ivs-vec.begin()] ) );
}
}
for ( CVectorSString::const_iterator ivs=_Words.begin(); ivs!=_Words.end(); ++ivs )
{
const CSString& word = *ivs;
string::size_type p;
if ( (p = word.findNS( searchStr.c_str() )) != string::npos )
{
if ( ((!findAtBeginning) || (p==0)) && ((!findAtEnd) || (p==word.size()-searchStr.size())) )
resultVec.push_back( makeResult( _Keys[ivs-_Words.begin()], word ) );
}
}
}