//*
// * Returns the length of the maximum number of elements matching
// * the string starting at the specified position. This method
// * allows no backtracking, i.e. no skips..
// *
// * @param m the matcher being used
// * @param input the input character stream to match
// * @param start the starting position
// * @param count the start count, normally zero (0)
// *
// * @return the length of the longest matching string, or
// * -1 if no match was found
// *
// * @throws IOException if an I/O error occurred
//
private int MatchPossessive(Matcher m, LookAheadReader input, int start, int count)
{
int length = 0;
int subLength = 1;
// Match as many elements as possible
while (subLength > 0 && count < max)
{
subLength = elem.Match(m, input, start + length, 0);
if (subLength >= 0)
{
count += 1;
length += subLength;
}
}
// Return result
if (min <= count && count <= max)
{
return(length);
}
else
{
return(-1);
}
}
//*
// * Returns the length of a matching string starting at the
// * specified position. The number of matches to skip can also
// * be specified, but numbers higher than zero (0) cause a
// * failed match for any element that doesn't attempt to
// * combine other elements.
// *
// * @param m the matcher being used
// * @param input the input character stream to match
// * @param start the starting position
// * @param skip the number of matches to skip
// *
// * @return the length of the longest matching string, or
// * -1 if no match was found
// *
// * @throws IOException if an I/O error occurred
//
public override int Match(Matcher m, LookAheadReader input, int start, int skip)
{
int c = 0;
if (skip != 0)
{
return(-1);
}
for (int i = 0; i <= value.Length - 1; i++)
{
c = input.Peek(start + i);
if (c < 0)
{
m.SetReadEndOfString();
return(-1);
}
if (m.IsCaseInsensitive())
{
c = Convert.ToInt32(Char.ToLower(Convert.ToChar(c)));
}
if (c != Convert.ToInt32(value[i]))
{
return(-1);
}
}
return(value.Length);
}
//*
// * Returns the length of the longest possible matching string
// * starting at the specified position. The number of matches
// * to skip can also be specified.
// *
// * @param m the matcher being used
// * @param input the input character stream to match
// * @param start the starting position
// * @param skip the number of matches to skip
// *
// * @return the length of the longest matching string, or
// * -1 if no match was found
// *
// * @throws IOException if an I/O error occurred
//
private int MatchGreedy(Matcher m, LookAheadReader input, int start, int skip)
{
// Check for simple case
if (skip == 0)
{
return(MatchPossessive(m, input, start, 0));
}
// Find all matches
if (matchStart != start)
{
matchStart = start;
matches = new BitArray(10);
FindMatches(m, input, start, 0, 0, 0);
}
for (int i = matches.Count - 1; i >= 0; i += -1)
{
// Find first non-skipped match
if (matches[i])
{
if (skip == 0)
{
return(i);
}
skip -= 1;
}
}
return(-1);
}
//*
// * Returns the length of a matching string starting at the
// * specified position. The number of matches to skip can also be
// * specified.
// *
// * @param m the matcher being used
// * @param input the input character stream to match
// * @param start the starting position
// * @param skip the number of matches to skip
// *
// * @return the length of the matching string, or
// * -1 if no match was found
// *
// * @throws IOException if an I/O error occurred
//
public override int Match(Matcher m, LookAheadReader input, int start, int skip)
{
if (skip == 0)
{
matchStart = -1;
matches = null;
}
switch (type)
{
case RepeatType.GREEDY:
return(MatchGreedy(m, input, start, skip));
case RepeatType.RELUCTANT:
return(MatchReluctant(m, input, start, skip));
case RepeatType.POSSESSIVE:
if (skip == 0)
{
return(MatchPossessive(m, input, start, 0));
}
break;
}
return(-1);
}
//*
// * Creates a new matcher with the specified element.
// *
// * @param e the base regular expression element
// * @param input the input character stream to work with
// * @param ignoreCase the character case ignore flag
//
internal Matcher(Element e, LookAheadReader input, bool ignoreCase)
{
this.element = e;
this.input = input;
this.ignoreCase = ignoreCase;
this.Reset();
}
//*
// * Returns the length of a matching string starting at the
// * specified position. The number of matches to skip can also
// * be specified, but numbers higher than zero (0) cause a
// * failed match for any element that doesn't attempt to
// * combine other elements.
// *
// * @param m the matcher being used
// * @param input the input character stream to match
// * @param start the starting position
// * @param skip the number of matches to skip
// *
// * @return the length of the longest matching string, or
// * -1 if no match was found
// *
// * @throws IOException if an I/O error occurred
//
public override int Match(Matcher m, LookAheadReader input, int start, int skip)
{
int length = 0;
int length1 = -1;
int length2 = -1;
int skip1 = 0;
int skip2 = 0;
while (length >= 0 && skip1 + skip2 <= skip)
{
length1 = elem1.Match(m, input, start, skip1);
length2 = elem2.Match(m, input, start, skip2);
if (length1 >= length2)
{
length = length1;
skip1 += 1;
}
else
{
length = length2;
skip2 += 1;
}
}
return(length);
}
//*
// * Returns the length of a matching string starting at the
// * specified position. The number of matches to skip can also be
// * specified, but numbers higher than zero (0) cause a failed
// * match for any element that doesn't attempt to combine other
// * elements.
// *
// * @param m the matcher being used
// * @param input the input character stream to match
// * @param start the starting position
// * @param skip the number of matches to skip
// *
// * @return the length of the longest matching string, or
// * -1 if no match was found
// *
// * @throws IOException if an I/O error occurred
//
public override int Match(Matcher m, LookAheadReader input, int start, int skip)
{
int length1 = -1;
int length2 = 0;
int skip1 = 0;
int skip2 = 0;
while (skip >= 0)
{
length1 = elem1.Match(m, input, start, skip1);
if (length1 < 0)
{
return(-1);
}
length2 = elem2.Match(m, input, start + length1, skip2);
if (length2 < 0)
{
skip1 += 1;
skip2 = 0;
}
else
{
skip2 += 1;
skip -= 1;
}
}
return(length1 + length2);
}
//*
// * Resets this tokenizer for usage with another input stream.
// * This method will clear all the internal state in the
// * tokenizer as well as close the previous input stream. It is
// * normally called in order to reuse a parser and tokenizer
// * pair for parsing another input stream.
// *
// * @param input the new input stream to read
// *
// * @since 1.5
//
public void Reset(TextReader input)
{
this.input.Close();
this.input = new LookAheadReader(input);
this.previousToken = null;
stringMatcher.Reset();
for (int i = 0; i <= regexpMatchers.Count - 1; i++)
{
((RegExpTokenMatcher)regexpMatchers[i]).Reset(this.input);
}
}
//*
// * Checks if the automaton matches an input stream. The
// * matching will be performed from a specified position. This
// * method will not read any characters from the stream, just
// * peek ahead. The comparison can be done either in
// * case-sensitive or case-insensitive mode.
// *
// * @param input the input stream to check
// * @param pos the starting position
// * @param caseInsensitive the case-insensitive flag
// *
// * @return the match value, or
// * null if no match was found
// *
// * @throws IOException if an I/O error occurred
//
public object MatchFrom(LookAheadReader input, int pos, bool caseInsensitive)
{
object result = null;
Automaton state = default(Automaton);
int c = 0;
c = input.Peek(pos);
if (tree != null && c >= 0)
{
state = tree.Find(Convert.ToChar(c), caseInsensitive);
if (state != null)
{
result = state.MatchFrom(input, pos + 1, caseInsensitive);
}
}
return((result == null) ? value : result);
}
//*
// * Returns the length of a matching string starting at the
// * specified position. The number of matches to skip can also be
// * specified, but numbers higher than zero (0) cause a failed
// * match for any element that doesn't attempt to combine other
// * elements.
// *
// * @param m the matcher being used
// * @param input the input character stream to match
// * @param start the starting position
// * @param skip the number of matches to skip
// *
// * @return the length of the matching string, or
// * -1 if no match was found
// *
// * @throws IOException if an I/O error occurred
//
public override int Match(Matcher m, LookAheadReader input, int start, int skip)
{
int c = 0;
if (skip != 0)
{
return(-1);
}
c = input.Peek(start);
if (c < 0)
{
m.SetReadEndOfString();
return(-1);
}
if (m.IsCaseInsensitive())
{
c = Convert.ToInt32(Char.ToLower(Convert.ToChar(c)));
}
return(InSet(Convert.ToChar(c)) ? 1 : -1);
}
//*
// * Finds all matches and adds the lengths to the matches set.
// *
// * @param m the matcher being used
// * @param input the input character stream to match
// * @param start the starting position
// * @param length the match length at the start position
// * @param count the number of sub-elements matched
// * @param attempt the number of match attempts here
// *
// * @throws IOException if an I/O error occurred
//
private void FindMatches(Matcher m, LookAheadReader input, int start, int length, int count, int attempt)
{
int subLength = 0;
// Check match ending here
if (count > max)
{
return;
}
if (min <= count && attempt == 0)
{
if (matches.Length <= length)
{
matches.Length = length + 10;
}
matches[length] = true;
}
// Check element match
subLength = elem.Match(m, input, start, attempt);
if (subLength < 0)
{
return;
}
else if (subLength == 0)
{
if (min == count + 1)
{
if (matches.Length <= length)
{
matches.Length = length + 10;
}
matches[length] = true;
}
return;
}
// Find alternative and subsequent matches
FindMatches(m, input, start, length, count, attempt + 1);
FindMatches(m, input, start + subLength, length + subLength, count + 1, 0);
}
//*
// * Returns the length of the shortest possible matching string
// * starting at the specified position. The number of matches to
// * skip can also be specified.
// *
// * @param m the matcher being used
// * @param input the input character stream to match
// * @param start the starting position
// * @param skip the number of matches to skip
// *
// * @return the length of the shortest matching string, or
// * -1 if no match was found
// *
// * @throws IOException if an I/O error occurred
//
private int MatchReluctant(Matcher m, LookAheadReader input, int start, int skip)
{
// Find all matches
if (matchStart != start)
{
matchStart = start;
matches = new BitArray(10);
FindMatches(m, input, start, 0, 0, 0);
}
for (int i = 0; i <= matches.Count - 1; i++)
{
// Find first non-skipped match
if (matches[i])
{
if (skip == 0)
{
return(i);
}
skip -= 1;
}
}
return(-1);
}
//*
// * Creates a new regular expression token matcher.
// *
// * @param pattern the pattern to match
// * @param ignoreCase the character case ignore flag
// * @param input the input stream to check
// *
// * @throws RegExpException if the regular expression couldn't
// * be created properly
//
public RegExpTokenMatcher(TokenPattern pattern, bool ignoreCase, LookAheadReader input)
{
this.pattern = pattern;
this.regExp = new RegExp(pattern.Pattern, ignoreCase);
this.matcher = regExp.Matcher(input);
}
//*
// * Returns the length of a matching string starting at the
// * specified position. The number of matches to skip can also
// * be specified, but numbers higher than zero (0) cause a
// * failed match for any element that doesn't attempt to
// * combine other elements.
// *
// * @param m the matcher being used
// * @param input the input character stream to match
// * @param start the starting position
// * @param skip the number of matches to skip
// *
// * @return the length of the matching string, or
// * -1 if no match was found
// *
// * @throws IOException if an I/O error occurred
//
public abstract int Match(Matcher m, LookAheadReader input, int start, int skip);
//*
// * Checks if the token pattern matches the input stream. This
// * method will also reset all flags in this matcher.
// *
// * @param input the input stream to match
// *
// * @return true if a match was found, or
// * false otherwise
// *
// * @throws IOException if an I/O error occurred
//
public bool Match(LookAheadReader input)
{
this.Reset();
m_match = (TokenPattern)start.MatchFrom(input, 0, ignoreCase);
return(m_match != null);
}
//*
// * Creates a new tokenizer for the specified input stream. The
// * tokenizer can be set to process tokens either in
// * case-sensitive or case-insensitive mode.
// *
// * @param input the input stream to read
// * @param ignoreCase the character case ignore flag
// *
// * @since 1.5
//
public Tokenizer(TextReader input, bool ignoreCase)
{
this.stringMatcher = new StringTokenMatcher(ignoreCase);
this.input = new LookAheadReader(input);
this.ignoreCase = ignoreCase;
}
//*
// * Resets the matcher for use with a new look-ahead character
// * input stream. This will clear all flags and set the match
// * length to a negative value.
// *
// * @param input the character input stream
// *
// * @since 1.5
//
private void Reset(LookAheadReader input)
{
this.input = input;
this.Reset();
}
//*
// * Creates a new matcher for the specified look-ahead
// * character input stream.
// *
// * @param input the character input stream
// *
// * @return the regular expresion matcher
// *
// * @since 1.5
//
private Matcher Matcher(LookAheadReader input)
{
return(new Matcher((Element)element.Clone(), input, ignoreCase));
}
//*
// * Resets the matcher for another character input stream. This
// * will clear the results of the last match.
// *
// * @param input the new input stream to check
//
public void Reset(LookAheadReader input)
{
matcher.Reset(input);
}
