+/* 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"
{
buffer.erase();
state = ACCEPT;
+
+ displayCursor();
}
void
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':
+ 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:
- if (isalpha(ch)) parser.push(TToken(TToken::LETTER, ch));
- else if (isdigit(ch)) parser.push(TToken(TToken::DIGIT, ch));
- else parser.push(TToken(TToken::OTHER, ch));
- break;
+ default: parser.push(TToken(TToken::OTHER, ch)); break;
}
state = newState;
}
else if (isalpha(ch))
{
buffer.push_back(ch);
- state = LONG_IDENTIFIER;
+ state = IDENTIFIER;
+ }
+ else if (isdigit(ch))
+ {
+ buffer.push_back(ch);
+ state = NUMBER;
}
else transaction(ch, ACCEPT);
break;
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));
buffer.erase();
state = ACCEPT;
}
- else
- {
- parser.push(TToken(TToken::CONTROL, buffer));
- buffer.erase();
- if (isspace(ch)) state = IGNORE_SPACE;
- else transaction(ch, 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;
}
- break;
- case IGNORE_SPACE:
- if (ch == '\\') state = ESCAPE;
- else if (ch == '#') state = PARAMETER;
- else if (isspace(ch)) ;
- else if (ch == -1) state = ACCEPT;
- else if (isalpha(ch))
+ else if (isdigit(ch))
{
+ parser.push(TToken(TToken::CONTROL, buffer));
+ buffer.erase();
buffer.push_back(ch);
- state = LONG_IDENTIFIER;
+ state = NUMBER;
}
- else transaction(ch, ACCEPT);
+ else
+ {
+ parser.push(TToken(TToken::CONTROL, buffer));
+ buffer.erase();
+ transaction(ch, ACCEPT);
+ }
break;
case PARAMETER:
if (ch == -1) error();
state = ACCEPT;
}
break;
- case LONG_IDENTIFIER:
+ case IDENTIFIER:
if (ch == -1)
{
parser.push(TToken(TToken::LETTER, buffer));
{
buffer.push_back(ch);
}
- else if (isspace(ch))
+ else if (ch == '\\') state = ESCAPED_CHARACTER;
+ else if (ch == '#')
{
parser.push(TToken(TToken::LETTER, buffer));
-
- // the parser ignores spaces. But in this case, the space
- // is a significant. So, we transform this space in the MACRO
- // \; which will not be ignored.
- // This is not a good solution. Having a special token, that will be
- // interpreted as "function application" is a better one.
buffer.erase();
- buffer = ";";
- parser.push(TToken(TToken::CONTROL, buffer));
+ state = PARAMETER;
+ }
+ else
+ {
+ parser.push(TToken(TToken::LETTER, buffer));
buffer.erase();
- state = IGNORE_SPACE;
+ transaction(ch, ACCEPT);
}
- else if (ch == '\\')
+ 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::LETTER, buffer));
+ parser.push(TToken(TToken::DIGIT, buffer));
buffer.erase();
transaction(ch, ACCEPT);
}
break;
}
- switch (state)
- {
- case ESCAPE: parser.setCursorHint("\\"); break;
- case MACRO: parser.setCursorHint("\\" + buffer); break;
- case PARAMETER: parser.setCursorHint("#"); break;
- case LONG_IDENTIFIER: parser.setCursorHint(buffer); break;
- default: parser.setCursorHint(""); break;
- }
+ displayCursor();
+
}
void
switch (state)
{
case ACCEPT:
- case IGNORE_SPACE:
- restore = parser.drop();
- if (restore.length() > 0 && restore[0] == '\\')
+ {
+ 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;
else buffer.erase(buffer.length() - 1, 1);
if (buffer.length() == 0) state = ESCAPE;
break;
- case LONG_IDENTIFIER:
- buffer.erase(buffer.length() - 1, 1);
+ 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);
+ //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;
}
}