SentenceAnnotation::SentenceAnnotation(uima::CAS& aCas, size_t begin, size_t end, const TokenAnnotation& firstToken, const TokenAnnotation& lastToken, const ParagraphAnnotation& paragraph)
{
FSIndexRepository& indexRep = aCas.getIndexRepository();
annotation = aCas.createAnnotation(tSentenceAnnotation, begin, end);
setFirstToken(firstToken);
setLastToken(lastToken);
setParagraph(paragraph);
indexRep.addFS(annotation);
}
SentenceAnnotation::SentenceAnnotation(CAS& aCas, size_t begin, size_t end, const TokenAnnotation& firstToken, const TokenAnnotation& lastToken) :
ContextAreaAnnotation(aCas)
{
FSIndexRepository& indexRep = aCas.getIndexRepository();
annotation = aCas.createAnnotation(tSentenceAnnotation, begin, end);
setFirstToken(firstToken);
setLastToken(lastToken);
indexRep.addFS(annotation);
}
bool parseForNextToken()
{
size_t pos = getCurrentPosition();
assert( pos < codestr.size() );
//
// Consume whitespace
//
for ( ;
pos < codestr.size() && isWhitespace(codestr[pos]);
++pos )
{
// do nothing, just consume space.
}
//
// End of the line?
//
if ( pos == codestr.size() )
{
setLastToken( "", TOK_EOF );
return false;
}
size_t startPos = pos;
//
// Search for the next token
//
TokenType type = TOK_UNKNOWN;
bool done = false;
bool error = false;
for ( ;
(!done) && pos < codestr.size() && (!isWhitespace(codestr[pos]))
&& ((pos == startPos) || !isEndOfToken(codestr[pos]) );
++pos )
{
char c = codestr[pos];
switch ( type )
{
case TOK_UNKNOWN:
if ( c == '(' )
{
type = TOK_OPEN;
done = true;
}
else if ( c == ')' )
{
type = TOK_CLOSE;
done = true;
}
else if ( c == ',' )
{
type = TOK_SEP;
done = true;
}
else if ( isNumeric( c ) )
{
type = TOK_NUMERIC;
}
else if ( isIdent( c ) )
{
type = TOK_IDENT;
}
else
{
error = true;
}
break;
case TOK_NUMERIC:
if ( isNumeric( c ) == false )
{
error = true;
}
break;
case TOK_IDENT:
if ( isIdent( c ) == false )
{
error = true;
}
break;
default:
error = true;
}
//
// Was there an error while parsing the token?
//
if ( error )
{
std::cerr << "Error while parsing token. "
<< "start=" << startPos << ", "
<< "pos=" << pos << ", "
<< "type=" << toString(type) << ", "
<< "value: "
<< codestr.substr( startPos, pos - startPos+1 )
<< std::endl;
return false;
}
}
setLastToken( codestr.substr(startPos, pos - startPos), type );
setCurrentPosition( pos );
return !isEOF();
}