ANTLR_BEGIN_NAMESPACE() template TokenSource::TokenSource() :m_eofToken( ImplTraits::CommonTokenType::TOKEN_EOF), m_skipToken( ImplTraits::CommonTokenType::TOKEN_INVALID) { } template ANTLR_INLINE typename TokenSource::CommonTokenType& TokenSource::get_eofToken() { return m_eofToken; } template ANTLR_INLINE const typename TokenSource::TokenType& TokenSource::get_eofToken() const { return m_eofToken; } template ANTLR_INLINE typename TokenSource::CommonTokenType& TokenSource::get_skipToken() { return m_skipToken; } template ANTLR_INLINE typename TokenSource::StringType& TokenSource::get_fileName() { return m_fileName; } template ANTLR_INLINE void TokenSource::set_fileName( const StringType& fileName ) { m_fileName = fileName; } template typename TokenSource::LexerType* TokenSource::get_super() { return static_cast(this); } template typename TokenSource::TokenType* TokenSource::nextTokenStr() { typedef typename LexerType::RecognizerSharedStateType RecognizerSharedStateType; typedef typename LexerType::InputStreamType InputStreamType; typedef typename LexerType::IntStreamType IntStreamType; LexerType* lexer; RecognizerSharedStateType* state; InputStreamType* input; IntStreamType* istream; lexer = this->get_super(); state = lexer->get_rec()->get_state(); input = lexer->get_input(); istream = input->get_istream(); /// Loop until we get a non skipped token or EOF /// for (;;) { // Get rid of any previous token (token factory takes care of // any de-allocation when this token is finally used up. // state->set_token_present(false); state->set_error(false); // Start out without an exception state->set_failed(false); // Now call the matching rules and see if we can generate a new token // for (;;) { // Record the start of the token in our input stream. // state->set_channel( TOKEN_DEFAULT_CHANNEL ); state->set_tokenStartCharIndex( (ANTLR_MARKER)input->get_nextChar() ); state->set_tokenStartCharPositionInLine( input->get_charPositionInLine() ); state->set_tokenStartLine( input->get_line() ); state->set_text(""); if (istream->_LA(1) == ANTLR_CHARSTREAM_EOF) { // Reached the end of the current stream, nothing more to do if this is // the last in the stack. // TokenType& teof = m_eofToken; teof.set_startIndex(lexer->getCharIndex()); teof.set_stopIndex(lexer->getCharIndex()); teof.set_line(lexer->getLine()); return &teof; } state->set_token_present( false ); state->set_error(false); // Start out without an exception state->set_failed(false); // Call the generated lexer, see if it can get a new token together. // lexer->mTokens(); if (state->get_error() == true) { // Recognition exception, report it and try to recover. // state->set_failed(true); lexer->get_rec()->reportError(); lexer->recover(); } else { if ( !state->get_token_present() ) { // Emit the real token, which adds it in to the token stream basically // lexer->emit(); } else if ( *(state->get_token()) == m_skipToken ) { // A real token could have been generated, but "Computer say's naaaaah" and it // it is just something we need to skip altogether. // continue; } // Good token, not skipped, not EOF token // return state->get_token(); } } } } template typename TokenSource::TokenType* TokenSource::nextToken() { return this->nextToken( BoolForwarder() ); } template typename TokenSource::CommonTokenType* TokenSource::nextToken( BoolForwarder /*isFiltered*/ ) { LexerType* lexer; typename LexerType::RecognizerSharedStateType* state; lexer = this->get_super(); state = lexer->get_lexstate(); /* Get rid of any previous token (token factory takes care of * any deallocation when this token is finally used up. */ state->set_token_present( false ); state->set_error( false ); /* Start out without an exception */ state->set_failed(false); /* Record the start of the token in our input stream. */ state->set_tokenStartCharIndex( lexer->index() ); state->set_tokenStartCharPositionInLine( lexer->getCharPositionInLine() ); state->set_tokenStartLine( lexer->getLine() ); state->set_text(""); /* Now call the matching rules and see if we can generate a new token */ for (;;) { if (lexer->LA(1) == ANTLR_CHARSTREAM_EOF) { /* Reached the end of the stream, nothing more to do. */ CommonTokenType& teof = m_eofToken; teof.set_startIndex(lexer->getCharIndex()); teof.set_stopIndex(lexer->getCharIndex()); teof.set_line(lexer->getLine()); return &teof; } state->set_token_present(false); state->set_error(false); /* Start out without an exception */ { ANTLR_MARKER m; m = lexer->get_istream()->mark(); state->set_backtracking(1); /* No exceptions */ state->set_failed(false); /* Call the generated lexer, see if it can get a new token together. */ lexer->mTokens(); state->set_backtracking(0); /* mTokens backtracks with synpred at BACKTRACKING==2 and we set the synpredgate to allow actions at level 1. */ if(state->get_failed()) { lexer->rewind(m); lexer->consume(); // } else { lexer->emit(); /* Assemble the token and emit it to the stream */ TokenType* tok = state->get_token(); return tok; } } } } template typename TokenSource::CommonTokenType* TokenSource::nextToken( BoolForwarder /*isFiltered*/ ) { // Find the next token in the current stream // CommonTokenType* tok = this->nextTokenStr(); // If we got to the EOF token then switch to the previous // input stream if there were any and just return the // EOF if there are none. We must check the next token // in any outstanding input stream we pop into the active // role to see if it was sitting at EOF after PUSHing the // stream we just consumed, otherwise we will return EOF // on the reinstalled input stream, when in actual fact // there might be more input streams to POP before the // real EOF of the whole logical inptu stream. Hence we // use a while loop here until we find somethign in the stream // that isn't EOF or we reach the actual end of the last input // stream on the stack. // while(tok->get_type() == CommonTokenType::TOKEN_EOF) { typename ImplTraits::LexerType* lexer; lexer = static_cast( this->get_super() ); if ( lexer->get_rec()->get_state()->get_streams().size() > 0) { // We have another input stream in the stack so we // need to revert to it, then resume the loop to check // it wasn't sitting at EOF itself. // lexer->popCharStream(); tok = this->nextTokenStr(); } else { // There were no more streams on the input stack // so this EOF is the 'real' logical EOF for // the input stream. So we just exit the loop and // return the EOF we have found. // break; } } // return whatever token we have, which may be EOF // return tok; } template TokenStream::TokenStream() { m_tokenSource = NULL; m_debugger = NULL; m_initialStreamState = false; } template typename TokenStream::IntStreamType* TokenStream::get_istream() { return this; } template TokenStream::TokenStream(TokenSourceType* source, DebugEventListenerType* debugger) { m_initialStreamState = false; m_tokenSource = source; m_debugger = debugger; } template CommonTokenStream::CommonTokenStream(ANTLR_UINT32 , TokenSourceType* source, DebugEventListenerType* debugger) : CommonTokenStream::BaseType( source, debugger ) { m_p = -1; m_channel = TOKEN_DEFAULT_CHANNEL; m_discardOffChannel = false; m_nissued = 0; } template typename CommonTokenStream::TokensType& CommonTokenStream::get_tokens() { return m_tokens; } template const typename CommonTokenStream::TokensType& CommonTokenStream::get_tokens() const { return m_tokens; } template typename CommonTokenStream::DiscardSetType& CommonTokenStream::get_discardSet() { return m_discardSet; } template const typename CommonTokenStream::DiscardSetType& CommonTokenStream::get_discardSet() const { return m_discardSet; } template ANTLR_INLINE ANTLR_INT32 CommonTokenStream::get_p() const { return m_p; } template ANTLR_INLINE void CommonTokenStream::set_p( ANTLR_INT32 p ) { m_p = p; } template ANTLR_INLINE void CommonTokenStream::inc_p() { ++m_p; } template ANTLR_INLINE void CommonTokenStream::dec_p() { --m_p; } template ANTLR_INLINE ANTLR_MARKER CommonTokenStream::index_impl() { return m_p; } // Reset a token stream so it can be used again and can reuse it's // resources. // template void CommonTokenStream::reset() { // Free any resources that ar most like specifc to the // run we just did. // m_discardSet.clear(); m_channelOverrides.clear(); // Now, if there were any existing tokens in the stream, // then we just reset the vector count so that it starts // again. We must traverse the entries unfortunately as // there may be free pointers for custom token types and // so on. However that is just a quick NULL check on the // vector entries. // m_tokens.clear(); // Reset to defaults // m_discardOffChannel = false; m_channel = ImplTraits::CommonTokenType::TOKEN_DEFAULT_CHANNEL; m_p = -1; } template void TokenStream::setDebugListener(DebugEventListenerType* debugger) { m_debugger = debugger; m_initialStreamState = false; } template const typename TokenStream::TokenType* TokenStream::_LT(ANTLR_INT32 k) { ANTLR_INT32 i; ANTLR_INT32 n; TokenStreamType* cts; cts = this->get_super(); if(k < 0) { return cts->LB(-k); } ANTLR_INT32 req_idx = cts->get_p() + k - 1; ANTLR_INT32 cached_size = static_cast(this->get_istream()->get_cachedSize()); if( (cts->get_p() == -1) || ( ( req_idx >= cached_size ) && ( (cached_size % ImplTraits::TOKEN_FILL_BUFFER_INCREMENT) == 0 ) ) ) { cts->fillBuffer(); } // Here we used to check for k == 0 and return 0, but this seems // a superfluous check to me. LT(k=0) is therefore just undefined // and we won't waste the clock cycles on the check // cached_size = static_cast(this->get_istream()->get_cachedSize()); if ( req_idx >= cached_size ) { TokenType& teof = cts->get_tokenSource()->get_eofToken(); teof.set_startIndex( this->get_istream()->index()); teof.set_stopIndex( this->get_istream()->index()); return &teof; } i = cts->get_p(); n = 1; /* Need to find k good tokens, skipping ones that are off channel */ while( n < k) { /* Skip off-channel tokens */ i = cts->skipOffTokenChannels(i+1); /* leave p on valid token */ n++; } if( ( i >= cached_size ) && ( (cached_size % ImplTraits::TOKEN_FILL_BUFFER_INCREMENT) == 0 ) ) { cts->fillBuffer(); } if ( (ANTLR_UINT32) i >= this->get_istream()->get_cachedSize() ) { TokenType& teof = cts->get_tokenSource()->get_eofToken(); teof.set_startIndex(this->get_istream()->index()); teof.set_stopIndex(this->get_istream()->index()); return &teof; } // Here the token must be in the input vector. Rather then incur // function call penalty, we just return the pointer directly // from the vector // return cts->getToken(i); } template const typename CommonTokenStream::TokenType* CommonTokenStream::LB(ANTLR_INT32 k) { ANTLR_INT32 i; ANTLR_INT32 n; if (m_p == -1) { this->fillBuffer(); } if (k == 0) { return NULL; } if ((m_p - k) < 0) { return NULL; } i = m_p; n = 1; /* Need to find k good tokens, going backwards, skipping ones that are off channel */ while (n <= k) { /* Skip off-channel tokens */ i = this->skipOffTokenChannelsReverse(i - 1); /* leave p on valid token */ n++; } if (i < 0) { return NULL; } // Here the token must be in the input vector. Rather then incut // function call penalty, we jsut return the pointer directly // from the vector // return this->getToken(i); } template const typename CommonTokenStream::TokenType* CommonTokenStream::getToken(ANTLR_MARKER i) { return this->get(i); } template const typename CommonTokenStream::TokenType* CommonTokenStream::get(ANTLR_MARKER i) { return this->getToken( static_cast(i), BoolForwarder() ); } template const typename CommonTokenStream::TokenType* CommonTokenStream::getToken( ANTLR_MARKER tok_idx, BoolForwarder /*tokens_accessed_from_owning_rule*/ ) { typename TokensType::iterator iter = m_tokens.find(tok_idx); if( iter == m_tokens.end() ) { TokenAccessException ex; throw ex; } const TokenType& tok = iter->second; return &tok; } template const typename CommonTokenStream::TokenType* CommonTokenStream::getToken( ANTLR_MARKER tok_idx, BoolForwarder /*tokens_accessed_from_owning_rule*/ ) { TokenType& tok = m_tokens.at( static_cast(tok_idx) ); return &tok; } template typename TokenStream::TokenSourceType* TokenStream::get_tokenSource() const { return m_tokenSource; } template void TokenStream::set_tokenSource( TokenSourceType* tokenSource ) { m_tokenSource = tokenSource; } template typename TokenStream::StringType TokenStream::toString() { TokenStreamType* cts = static_cast(this); if (cts->get_p() == -1) { cts->fillBuffer(); } return this->toStringSS(0, this->get_istream()->size()); } template typename TokenStream::StringType TokenStream::toStringSS(ANTLR_MARKER start, ANTLR_MARKER stop) { StringType string; TokenSourceType* tsource; const TokenType* tok; TokenStreamType* cts; cts = this->get_super(); if (cts->get_p() == -1) { cts->fillBuffer(); } if (stop >= this->get_istream()->size()) { stop = this->get_istream()->size() - 1; } /* Who is giving us these tokens? */ tsource = cts->get_tokenSource(); if (tsource != NULL && !cts->get_tokens().empty() ) { /* Finally, let's get a string */ for (ANTLR_MARKER i = start; i <= stop; i++) { tok = cts->get(i); if (tok != NULL) { string.append( tok->getText() ); } } return string; } return ""; } template typename TokenStream::StringType TokenStream::toStringTT(const TokenType* start, const TokenType* stop) { if (start != NULL && stop != NULL) { return this->toStringSS( start->get_tokenIndex(), stop->get_tokenIndex()); } else { return ""; } } /** A simple filter mechanism whereby you can tell this token stream * to force all tokens of type ttype to be on channel. For example, * when interpreting, we cannot execute actions so we need to tell * the stream to force all WS and NEWLINE to be a different, ignored, * channel. */ template void CommonTokenStream::setTokenTypeChannel ( ANTLR_UINT32 ttype, ANTLR_UINT32 channel) { /* We add one to the channel so we can distinguish NULL as being no entry in the * table for a particular token type. */ m_channelOverrides[ttype] = (ANTLR_UINT32)channel + 1; } template void CommonTokenStream::discardTokenType(ANTLR_INT32 ttype) { /* We add one to the channel so we can distinguish NULL as being no entry in the * table for a particular token type. We could use bitsets for this I suppose too. */ m_discardSet.insert(ttype); } template void CommonTokenStream::discardOffChannelToks(bool discard) { m_discardOffChannel = discard; } template typename CommonTokenStream::TokensType* CommonTokenStream::getTokens() { if (m_p == -1) { this->fillBuffer(); } return &m_tokens; } template void CommonTokenStream::getTokenRange(ANTLR_UINT32 start, ANTLR_UINT32 stop, TokensListType& tokenRange) { return this->getTokensSet(start, stop, NULL, tokenRange); } /** Given a start and stop index, return a List of all tokens in * the token type BitSet. Return null if no tokens were found. This * method looks at both on and off channel tokens. */ template void CommonTokenStream::getTokensSet(ANTLR_UINT32 start, ANTLR_UINT32 stop, BitsetType* types, TokensListType& filteredList ) { ANTLR_UINT32 i; ANTLR_UINT32 n; TokenType* tok; if ( m_p == -1) { this->fillBuffer(); } if (stop > this->get_istream()->size()) { stop = this->get_istream()->size(); } if (start > stop) { return; } /* We have the range set, now we need to iterate through the * installed tokens and create a new list with just the ones we want * in it. We are just moving pointers about really. */ for(i = start, n = 0; i<= stop; i++) { tok = this->get(i); if ( types == NULL || (types->isMember( tok->get_type() ) == true ) ) { filteredList.push_back(tok); } } return ; } template void CommonTokenStream::getTokensList(ANTLR_UINT32 start, ANTLR_UINT32 stop, const IntListType& list, TokensListType& newlist) { BitsetType* bitSet; bitSet = Bitset::BitsetFromList(list); this->getTokensSet(start, stop, bitSet, newlist); delete bitSet; } template void CommonTokenStream::getTokensType(ANTLR_UINT32 start, ANTLR_UINT32 stop, ANTLR_UINT32 type, TokensListType& newlist ) { BitsetType* bitSet; bitSet = BitsetType::BitsetOf(type, -1); this->getTokensSet(start, stop, bitSet, newlist); delete bitSet; } template void CommonTokenStream::fillBufferExt() { this->fillBuffer(); } template bool CommonTokenStream::hasReachedFillbufferTarget( ANTLR_UINT32 cnt, BoolForwarder ) { return ( cnt >= ImplTraits::TOKEN_FILL_BUFFER_INCREMENT ); } template bool CommonTokenStream::hasReachedFillbufferTarget( ANTLR_UINT32, BoolForwarder ) { return false; } template void CommonTokenStream::fillBuffer() { ANTLR_UINT32 index; TokenType* tok; bool discard; /* Start at index 0 of course */ ANTLR_UINT32 cached_p = (m_p < 0) ? 0 : m_p; index = m_nissued; ANTLR_UINT32 cnt = 0; /* Pick out the next token from the token source * Remember we just get a pointer (reference if you like) here * and so if we store it anywhere, we don't set any pointers to auto free it. */ tok = this->get_tokenSource()->nextToken(); while ( tok->get_type() != TokenType::TOKEN_EOF ) { discard = false; /* Assume we are not discarding */ /* I employ a bit of a trick, or perhaps hack here. Rather than * store a pointer to a structure in the override map and discard set * we store the value + 1 cast to a void *. Hence on systems where NULL = (void *)0 * we can distinguish "not being there" from "being channel or type 0" */ if ( m_discardSet.find(tok->get_type()) != m_discardSet.end() ) { discard = true; } else if ( m_discardOffChannel == true && tok->get_channel() != m_channel ) { discard = true; } else if (!m_channelOverrides.empty()) { /* See if this type is in the override map */ typename ChannelOverridesType::iterator iter = m_channelOverrides.find( tok->get_type() + 1 ); if (iter != m_channelOverrides.end()) { /* Override found */ tok->set_channel( ANTLR_UINT32_CAST(iter->second) - 1); } } /* If not discarding it, add it to the list at the current index */ if (discard == false) { /* Add it, indicating that we will delete it and the table should not */ tok->set_tokenIndex(index); ++m_p; this->insertToken(*tok); index++; m_nissued++; cnt++; } if( !this->hasReachedFillbufferTarget( cnt, BoolForwarder() ) ) tok = this->get_tokenSource()->nextToken(); else break; } /* Cache the size so we don't keep doing indirect method calls. We do this as * early as possible so that anything after this may utilize the cached value. */ this->get_istream()->set_cachedSize( m_nissued ); /* Set the consume pointer to the first token that is on our channel, we just read */ m_p = cached_p; m_p = this->skipOffTokenChannels( m_p ); } /// Given a starting index, return the index of the first on-channel /// token. /// template ANTLR_UINT32 CommonTokenStream::skipOffTokenChannels(ANTLR_INT32 i) { ANTLR_INT32 n; n = this->get_istream()->get_cachedSize(); while (i < n) { const TokenType* tok = this->getToken(i); if (tok->get_channel() != m_channel ) { i++; } else { return i; } } return i; } template ANTLR_UINT32 CommonTokenStream::skipOffTokenChannelsReverse(ANTLR_INT32 x) { while (x >= 0) { const TokenType* tok = this->getToken(x); if( tok->get_channel() != m_channel ) { x--; } else { return x; } } return x; } template void CommonTokenStream::discardTokens( ANTLR_MARKER start, ANTLR_MARKER stop ) { this->discardTokens( start, stop, BoolForwarder< ImplTraits::TOKENS_ACCESSED_FROM_OWNING_RULE >() ); } template void CommonTokenStream::discardTokens( ANTLR_MARKER start, ANTLR_MARKER stop, BoolForwarder /*tokens_accessed_from_owning_rule */ ) { typename TokensType::iterator iter1 = m_tokens.lower_bound(start); typename TokensType::iterator iter2 = m_tokens.upper_bound(stop); m_tokens.erase( iter1, iter2 ); } template void CommonTokenStream::discardTokens( ANTLR_MARKER start, ANTLR_MARKER stop, BoolForwarder /*tokens_accessed_from_owning_rule*/ ) { m_tokens.erase( m_tokens.begin() + start, m_tokens.begin() + stop ); } template void CommonTokenStream::insertToken( const TokenType& tok ) { this->insertToken( tok, BoolForwarder< ImplTraits::TOKENS_ACCESSED_FROM_OWNING_RULE >() ); } template void CommonTokenStream::insertToken( const TokenType& tok, BoolForwarder /*tokens_accessed_from_owning_rule*/ ) { assert( m_tokens.find( tok.get_index() ) == m_tokens.end() ); assert( tok.get_index() == m_nissued ); m_tokens[ tok.get_index() ] = tok; } template void CommonTokenStream::insertToken( const TokenType& tok, BoolForwarder /*tokens_accessed_from_owning_rule*/ ) { m_tokens.push_back( tok ); } template CommonTokenStream::~CommonTokenStream() { m_tokens.clear(); } ANTLR_END_NAMESPACE()