-/* This file is part of EdiTeX, an editor of mathematical
- * expressions based on TeX syntax.
- *
- * Copyright (C) 2002-2003 Luca Padovani <lpadovan@cs.unibo.it>,
- * 2003 Paolo Marinelli <pmarinel@cs.unibo.it>.
- *
- * This library is free software; you can redistribute it and/or
- * modify it under the terms of the GNU Lesser General Public
- * License as published by the Free Software Foundation; either
- * version 2.1 of the License, or (at your option) any later version.
- *
- * This library is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
- * Lesser General Public License for more details.
- *
- * You should have received a copy of the GNU Lesser General Public
- * License along with this library; if not, write to the Free Software
- * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
- *
- * For more information, please visit the project's home page
- * http://helm.cs.unibo.it/editex/
- * or send an email to <lpadovan@cs.unibo.it>
- */
-
-#include <string>
-#include <cctype>
-#include <cassert>
-
-#include "ALogger.hh"
-#include "TToken.hh"
-#include "LPushLexer.hh"
-#include "APushParser.hh"
-
-LPushLexer::LPushLexer(ALogger& l, APushParser& p) : APushLexer(l, p)
-{
- state = ACCEPT;
-}
-
-void
-LPushLexer::reset()
-{
- buffer.erase();
- state = ACCEPT;
-
- displayCursor();
-}
-
-void
-LPushLexer::flush()
-{
- push(-1);
-}
-
-void
-LPushLexer::transaction(char ch, State newState)
-{
- switch (ch)
- {
- case '{': parser.push(TToken(TToken::BEGIN)); break;
- case '}': parser.push(TToken(TToken::END)); break;
- case '$': parser.push(TToken(TToken::SHIFT)); break;
- case '&': parser.push(TToken(TToken::ALIGN)); break;
- case '\n':
- case '\r': parser.push(TToken(TToken::EOL, ch)); break;
- case '^': parser.push(TToken(TToken::SUPERSCRIPT)); break;
- case '_': parser.push(TToken(TToken::SUBSCRIPT)); break;
- case '\t': parser.push(TToken(TToken::IGNORABLE_SPACE, ch)); break;
- case ' ': parser.push(TToken(TToken::SPACE, ch)); break;
- case '~': parser.push(TToken(TToken::ACTIVE, ch)); break;
- case '%': parser.push(TToken(TToken::COMMENT)); break;
- default: parser.push(TToken(TToken::OTHER, ch)); break;
- }
- state = newState;
-}
-
-void
-LPushLexer::push(char ch)
-{
- switch (state)
- {
- case ACCEPT:
- if (ch == '\\') state = ESCAPE;
- else if (ch == '#') state = PARAMETER;
- else if (ch == -1) ;
- else if (isalpha(ch))
- {
- buffer.push_back(ch);
- state = IDENTIFIER;
- }
- else if (isdigit(ch))
- {
- buffer.push_back(ch);
- state = NUMBER;
- }
- else transaction(ch, ACCEPT);
- break;
- case ESCAPE:
- if (isalpha(ch))
- {
- buffer.push_back(ch);
- state = MACRO;
- }
- else if (ch == -1) error();
- else if (isdigit(ch))
- {
- // in this case, the previous '\' is ignored
- buffer.push_back(ch);
- state = NUMBER;
- }
- else
- {
- parser.push(TToken(TToken::CONTROL, ch));
- state = ACCEPT;
- }
- break;
- case MACRO:
- if (ch == '\\')
- {
- parser.push(TToken(TToken::CONTROL, buffer));
- buffer.erase();
- state = ESCAPE;
- }
- else if (ch == '#')
- {
- parser.push(TToken(TToken::CONTROL, buffer));
- buffer.erase();
- state = PARAMETER;
- }
- else if (isalpha(ch))
- buffer.push_back(ch);
- else if (ch == -1)
- {
- parser.push(TToken(TToken::CONTROL, buffer));
- buffer.erase();
- state = ACCEPT;
- }
- else if (isspace(ch))
- {
- // we don't call transaction, because a white space is useful to exit from the macro,
- // without "side effect". It's the TeX syntax.
- parser.push(TToken(TToken::CONTROL, buffer));
- buffer.erase();
- state = ACCEPT;
- }
- else if (isdigit(ch))
- {
- parser.push(TToken(TToken::CONTROL, buffer));
- buffer.erase();
- buffer.push_back(ch);
- state = NUMBER;
- }
- else
- {
- parser.push(TToken(TToken::CONTROL, buffer));
- buffer.erase();
- transaction(ch, ACCEPT);
- }
- break;
- case PARAMETER:
- if (ch == -1) error();
- else
- {
- parser.push(TToken(TToken::PARAMETER, ch));
- state = ACCEPT;
- }
- break;
- case IDENTIFIER:
- if (ch == -1)
- {
- parser.push(TToken(TToken::LETTER, buffer));
- buffer.erase();
- state = ACCEPT;
- }
- else if (isalpha(ch) || isdigit(ch))
- {
- buffer.push_back(ch);
- }
- else if (ch == '\\') state = ESCAPED_CHARACTER;
- else if (ch == '#')
- {
- parser.push(TToken(TToken::LETTER, buffer));
- buffer.erase();
- state = PARAMETER;
- }
- else
- {
- parser.push(TToken(TToken::LETTER, buffer));
- buffer.erase();
- transaction(ch, ACCEPT);
- }
- break;
- case ESCAPED_CHARACTER:
- if ((ch == '-') || (ch == '_') || (ch == '/'))
- {
- buffer.push_back(ch);
- state = IDENTIFIER;
- }
- else if (isalpha(ch))
- {
- parser.push(TToken(TToken::LETTER, buffer));
- buffer.erase();
- buffer.push_back(ch);
- state = MACRO;
- }
- else if (ch == -1) error();
- else if (isdigit(ch))
- {
- parser.push(TToken(TToken::LETTER, buffer));
- buffer.erase();
- buffer.push_back(ch);
- state = NUMBER;
- }
- else
- {
- parser.push(TToken(TToken::LETTER, buffer));
- buffer.erase();
- parser.push(TToken(TToken::CONTROL, ch));
- state = ACCEPT;
- }
- break;
- case NUMBER:
- if (isdigit(ch)) buffer.push_back(ch);
- else if (isalpha(ch))
- {
- parser.push(TToken(TToken::DIGIT, buffer));
- buffer.erase();
- buffer.push_back(ch);
- state = IDENTIFIER;
- }
- else if (ch == -1)
- {
- parser.push(TToken(TToken::DIGIT, buffer));
- buffer.erase();
- state = ACCEPT;
- }
- else if (ch == '\\')
- {
- parser.push(TToken(TToken::DIGIT, buffer));
- buffer.erase();
- state = ESCAPE;
- }
- else if (ch == '#')
- {
- parser.push(TToken(TToken::DIGIT, buffer));
- buffer.erase();
- state = PARAMETER;
- }
- else
- {
- parser.push(TToken(TToken::DIGIT, buffer));
- buffer.erase();
- transaction(ch, ACCEPT);
- }
- break;
- default:
- assert(0);
- break;
- }
-
- displayCursor();
-
-}
-
-void
-LPushLexer::drop(bool alt)
-{
- std::string restore = "";
-
- switch (state)
- {
- case ACCEPT:
- {
- restore = parser.drop(alt);
- long bs_pos = restore.find('\\');
- if ((restore.length() > 0) && (bs_pos != std::string::npos))
- {
- // in this case we have to control the blackslash's position
- if (bs_pos == 0)
- {
- //logger.debug(restore);
- buffer = std::string(restore, 1, restore.length() - 1);
- state = (buffer.length() > 0) ? MACRO : ESCAPE;
- }
- else
- {
- assert(bs_pos == restore.length() - 1);
- buffer = std::string(restore, 0, bs_pos);
- state = ESCAPED_CHARACTER;
- }
- }
- else if (restore.length() > 0 && isdigit(restore[0]))
- {
- buffer = restore;
- state = NUMBER;
- }
- else if (restore.length() > 0 && isalpha(restore[0]))
- {
- buffer = restore;
- state = IDENTIFIER;
- }
- }
- break;
-/* if (restore.length() > 0 && restore[0] == '\\')
- {
- logger.debug(restore);
- buffer = std::string(restore, 1, restore.length() - 1);
- state = (buffer.length() > 0) ? MACRO : ESCAPE;
- }
- else if (restore.length() > 0 && isdigit(restore[0]))
- {
- buffer = restore;
- state = NUMBER;
- }
- else if (restore.length() > 0 && isalpha(restore[0]))
- {
- buffer = restore;
- state = IDENTIFIER;
- }
- break;*/
- case ESCAPED_CHARACTER:
- state = IDENTIFIER;
- break;
- case ESCAPE:
- state = ACCEPT;
- break;
- case MACRO:
- if (alt) buffer.erase();
- else buffer.erase(buffer.length() - 1, 1);
- if (buffer.length() == 0) state = ESCAPE;
- break;
- case IDENTIFIER:
- switch (buffer[buffer.length() - 1])
- {
- case '-':
- case '_':
- buffer.erase(buffer.length() - 1, 1);
- if (alt) state = ESCAPED_CHARACTER;
- break;
- default:
- if (alt) buffer.erase();
- else buffer.erase(buffer.length() - 1, 1);
- if (buffer.length() == 0) state = ACCEPT;
- break;
- }
- break;
- case NUMBER:
- if (alt) buffer.erase();
- else buffer.erase(buffer.length() - 1, 1);
- if (buffer.length() == 0) state = ACCEPT;
- break;
- case PARAMETER:
- default:
- //assert(0);
- error();
- break;
- }
-
- displayCursor();
-
-}
-
-void
-LPushLexer::displayCursor()
-{
- switch (state)
- {
- case ESCAPE: parser.setCursorHint("\\"); break;
- case ESCAPED_CHARACTER: parser.setCursorHint(buffer + "\\"); break;
- case MACRO: parser.setCursorHint("\\" + buffer); break;
- case PARAMETER: parser.setCursorHint("#"); break;
- case IDENTIFIER: parser.setCursorHint(buffer); break;
- case NUMBER: parser.setCursorHint(buffer); break;
- default: parser.setCursorHint(""); break;
- }
-}
-
-bool
-LPushLexer::error() const
-{
- return false;
-}