X-Git-Url: http://nitlanguage.org diff --git a/src/parser/xss/lexer.xss b/src/parser/xss/lexer.xss index 4338dbc..affaa11 100644 --- a/src/parser/xss/lexer.xss +++ b/src/parser/xss/lexer.xss @@ -1,300 +1,75 @@ -/* This file is part of NIT ( http://www.nitlanguage.org ). - * - * Copyright 2008 Jean Privat - * Based on algorithms developped for ( http://www.sablecc.org/ ). - * - * Licensed under the Apache License, Version 2.0 (the "License"); - * you may not use this file except in compliance with the License. - * You may obtain a copy of the License at - * - * http://www.apache.org/licenses/LICENSE-2.0 - * - * Unless required by applicable law or agreed to in writing, software - * distributed under the License is distributed on an "AS IS" BASIS, - * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. - * See the License for the specific language governing permissions and - * limitations under the License. - */ +$ // This file is part of NIT ( http://www.nitlanguage.org ). +$ // +$ // Copyright 2008 Jean Privat +$ // Based on algorithms developped for ( http://www.sablecc.org/ ). +$ // +$ // Licensed under the Apache License, Version 2.0 (the "License"); +$ // you may not use this file except in compliance with the License. +$ // You may obtain a copy of the License at +$ // +$ // http://www.apache.org/licenses/LICENSE-2.0 +$ // +$ // Unless required by applicable law or agreed to in writing, software +$ // distributed under the License is distributed on an "AS IS" BASIS, +$ // WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +$ // See the License for the specific language governing permissions and +$ // limitations under the License. $ template make_lexer() - -# The lexer extract NIT tokens from an input stream. -# It is better user with the Parser -class Lexer - # Last peeked token - var _token: nullable Token - - # Lexer current state - var _state: Int = 0 - - # Name of the stream (as given to tokens) - readable var _filename: String - - # Input stream where character are read - var _stream: IStream - - # Pushback buffer to store unread character - var _stream_buf: Buffer - - # Number of character stored in the pushback buffer - var _stream_pos: Int - - # Current line number in the input stream - var _line: Int = 0 - - # Current column in the input stream - var _pos: Int = 0 - - # Was the last character a cariage-return? - var _cr: Bool = false - - # If the end of stream? - var _eof: Bool = false - - # Current working text read from the input stream - var _text: Buffer - -$ foreach {lexer_data/state} - # Constante state values - private fun state_${translate(@name,"ABCDEFGHIJKLMNOPQRSTUVWXYZ","abcdefghijklmnopqrstuvwxyz")}: Int do return @id end -$ end foreach - - # Create a new lexer for a stream (and a name) - init(stream: IStream, fname: String) - do - _filename = fname - _text = new Buffer - _stream = stream - _stream_pos = -1 - _stream_buf = new Buffer - build_goto_table - build_accept_table - end - - # Give the next token (but do not consume it) - fun peek: Token +redef class Lexer + redef fun make_token(accept_token, location) do - while _token == null do - _token = get_token - end - return _token.as(not null) - end - - # Give and consume the next token - fun next: Token - do - var result = _token - while result == null do - result = get_token - end - _token = null - return result.as(not null) - end - - # Get a token, or null if it is discarded - private fun get_token: nullable Token - do - var dfa_state = 0 - - var start_pos = _pos - var start_line = _line - - var accept_state = -1 - var accept_token = -1 - var accept_length = -1 - var accept_pos = -1 - var accept_line = -1 - - var goto_table = _goto_table[_state] - var accept = _accept_table[_state] - _text.clear - - while true do - var c = get_char - - if c != -1 then - if c == 10 then - if _cr then - _cr = false - else - _line = _line + 1 - _pos = 0 - end - else if c == 13 then - _line = _line + 1 - _pos = 0 - _cr = true - else - _pos = _pos + 1 - _cr = false - end - - _text.add(c.ascii) - - var first_loop = true # aka until - while dfa_state < -1 or first_loop do - var old_state = dfa_state - if dfa_state < -1 then - old_state = -2 - dfa_state - end - - dfa_state = -1 - - var tmp1 = goto_table[old_state] - var low = 0 - var high = tmp1.length - 1 - - while low <= high do - var middle = (low + high) / 2 - var tmp2 = tmp1[middle] - - if c < tmp2[0] then - high = middle - 1 - else if c > tmp2[1] then - low = middle + 1 - else - dfa_state = tmp2[2] - low = high + 1 # aka break - end - end - first_loop = false # aka until - end - else - dfa_state = -1 - end - - if dfa_state >= 0 then - if accept[dfa_state] != -1 then - accept_state = dfa_state - accept_token = accept[dfa_state] - accept_length = _text.length - accept_pos = _pos - accept_line = _line - end - else - if accept_state != -1 then $ foreach {//token} - if accept_token == ${position()-1} then - var location = new Location(_filename, start_line + 1, accept_line + 1, start_pos + 1, accept_pos) -$ if {not(@text)} -$ if {@parser_index} - var token_text = _text.substring(0, accept_length) - var token = new @ename.init_tk(token_text, location) -$ end -$ else - var token = new @ename.init_tk(location) -$ end - push_back(accept_length) - _pos = accept_pos - _line = accept_line -$ if {count(transition[@from!=@to])!=0} - var state_id = _state -$ foreach transition in {transition[@from!=@to]} - if state_id == ${/parser/lexer_data/state[@name=$transition/@from]/@id} then - _state = state_${translate(@to,"ABCDEFGHIJKLMNOPQRSTUVWXYZ","abcdefghijklmnopqrstuvwxyz")} - end -$ end -$ end if $ if {@parser_index} - return token -$ else - return null -$ end - end -$ end foreach - else - var location = new Location(_filename, start_line + 1, accept_line + 1, start_pos + 1, accept_pos) - if _text.length > 0 then - var token = new PError.init_error("Unknown token: {_text}", location) - return token - else - var token = new EOF(location) - return token - end - end + if accept_token == ${position()-1} then +$ if {count(transition[@from!=@to])!=0} + var state_id = _state +$ foreach transition in {transition[@from!=@to]} + if state_id == ${/parser/lexer_data/state[@name=$transition/@from]/@id} then + _state = state_${translate(@to,"ABCDEFGHIJKLMNOPQRSTUVWXYZ","abcdefghijklmnopqrstuvwxyz")} end +$ end +$ end if + return new @ename.init_tk(location) end - return null +$ end +$ end foreach + abort # unknown token index `accept_token` end +end - # Read the next character. - # The character is read from the stream of from the pushback buffer. - private fun get_char: Int - do - if _eof then - return -1 - end - - var result: Int - - var sp = _stream_pos - if sp >= 0 then - var res = _stream_buf[_stream_pos] - _stream_pos = sp - 1 - result = res.ascii - else - result = _stream.read_char - end - - if result == -1 then - _eof = true - end +$ end template - return result - end - # Unread some characters. - # Unread characters are stored in the pushback buffer. - private fun push_back(accept_length: Int) - do - var length = _text.length - var i = length - 1 - while i >= accept_length do - _eof = false - _stream_pos = _stream_pos + 1 - _stream_buf[_stream_pos] = _text[i] - i = i - 1 - end - end - var _goto_table: Array[Array[Array[Array[Int]]]] - private fun build_goto_table - do - _goto_table = once [ +$ template make_lexer_table() $ foreach {lexer_data/goto_table/state} - [ $ foreach {row} $ if {count(goto)!=0} - [ +static const int lexer_goto_row${position()}[] = { + ${count(goto)}, $ foreach {goto} - [@low, @high, @state] [-sep ','-] + @low, @high, @state[-sep ','-] $ end foreach - ] [-sep ','-] +}; +$ end +$ end foreach +static const int lexer_goto_row_null[] = {0}; +const int* const lexer_goto_table[] = { +$ foreach {row} +$ if {count(goto)!=0} + lexer_goto_row${position()}[-sep ','-] $ else - nil_array [-sep ','-] + lexer_goto_row_null[-sep ','-] $ end $ end foreach - ] [-sep ','-] +}; $ end foreach - ] - end - - private fun nil_array: Array[Array[Int]] - do - return once new Array[Array[Int]] - end - var _accept_table: Array[Array[Int]] - private fun build_accept_table do - _accept_table = once [ $ foreach {lexer_data/accept_table/state} - [ - [-foreach {i}-]${.} [-sep ','-] [-end foreach-] - - ] [-sep ','-] +const int lexer_accept_table[] = { + [-foreach {i}-]${.}[-sep ','-][-end foreach-] +}; $ end foreach - ] - end -end $ end template