*
* REMatch myMatch = myExpression.getMatch(myString);
*/
public String toString ()
{
return matchedText;
}
/**
* Returns the index within the input text where the match in its entirety
* began.
*/
public int getStartIndex ()
{
return offset + start[0];
}
/**
* Returns the index within the input string where the match in
* its entirety ends. The return value is the next position after
* the end of the string; therefore, a match created by the
* following call:
*
*
* if (myMatch != null) System.out.println("Regexp found: "+myMatch);
*
* REMatch myMatch = myExpression.getMatch(myString);
*
* can be viewed (given that myMatch is not null) by creating *
* String theMatch = myString.substring(myMatch.getStartIndex(),
* myMatch.getEndIndex());
*
* But you can save yourself that work, since the toString()
* method (above) does exactly that for you.
*/
public int getEndIndex ()
{
return offset + end[0];
}
/**
* Returns the string matching the given subexpression. The subexpressions
* are indexed starting with one, not zero. That is, the subexpression
* identified by the first set of parentheses in a regular expression
* could be retrieved from an REMatch by calling match.toString(1).
*
* @param sub Index of the subexpression.
*/
public String toString (int sub)
{
if ((sub >= start.length) || sub < 0)
throw new IndexOutOfBoundsException ("No group " + sub);
if (start[sub] == -1)
return null;
if (start[sub] >= 0 && end[sub] <= matchedText.length ())
return (matchedText.substring (start[sub], end[sub]));
else
{
// This case occurs with RETokenLookAhead or RETokenLookBehind.
CPStringBuilder sb = new CPStringBuilder ();
int s = start[sub];
int e = end[sub];
if (s < 0)
s += 1;
if (e < 0)
e += 1;
for (int i = start[0] + s; i < start[0] + e; i++)
sb.append (matchedCharIndexed.charAt (i));
return sb.toString ();
}
}
/**
* Returns the index within the input string used to generate this match
* where subexpression number sub begins, or -1 if
* the subexpression does not exist. The initial position is zero.
*
* @param sub Subexpression index
* @deprecated Use getStartIndex(int) instead.
*/
public int getSubStartIndex (int sub)
{
if (sub >= start.length)
return -1;
int x = start[sub];
return (x == -1) ? x : (x >= 0) ? offset + x : offset + x + 1;
}
/**
* Returns the index within the input string used to generate this match
* where subexpression number sub begins, or -1 if
* the subexpression does not exist. The initial position is zero.
*
* @param sub Subexpression index
* @since gnu.regexp 1.1.0
*/
public int getStartIndex (int sub)
{
if (sub >= start.length)
return -1;
int x = start[sub];
return (x == -1) ? x : (x >= 0) ? offset + x : offset + x + 1;
}
/**
* Returns the index within the input string used to generate this match
* where subexpression number sub ends, or -1 if
* the subexpression does not exist. The initial position is zero.
*
* @param sub Subexpression index
* @deprecated Use getEndIndex(int) instead
*/
public int getSubEndIndex (int sub)
{
if (sub >= start.length)
return -1;
int x = end[sub];
return (x == -1) ? x : (x >= 0) ? offset + x : offset + x + 1;
}
/**
* Returns the index within the input string used to generate this match
* where subexpression number sub ends, or -1 if
* the subexpression does not exist. The initial position is zero.
*
* @param sub Subexpression index
*/
public int getEndIndex (int sub)
{
if (sub >= start.length)
return -1;
int x = end[sub];
return (x == -1) ? x : (x >= 0) ? offset + x : offset + x + 1;
}
/**
* Substitute the results of this match to create a new string.
* This is patterned after PERL, so the tokens to watch out for are
* $0 through $9. $0 matches
* the full substring matched; $n matches
* subexpression number n.
* $10, $11, ... may match the 10th, 11th, ... subexpressions
* if such subexpressions exist.
*
* @param input A string consisting of literals and $n tokens.
*/
public String substituteInto (String input)
{
// a la Perl, $0 is whole thing, $1 - $9 are subexpressions
CPStringBuilder output = new CPStringBuilder ();
int pos;
for (pos = 0; pos < input.length () - 1; pos++)
{
if ((input.charAt (pos) == '$')
&& (Character.isDigit (input.charAt (pos + 1))))
{
int val = Character.digit (input.charAt (++pos), 10);
int pos1 = pos + 1;
while (pos1 < input.length () &&
Character.isDigit (input.charAt (pos1)))
{
int val1 =
val * 10 + Character.digit (input.charAt (pos1), 10);
if (val1 >= start.length)
break;
pos1++;
val = val1;
}
pos = pos1 - 1;
if (val < start.length)
{
output.append (toString (val));
}
}
else
output.append (input.charAt (pos));
}
if (pos < input.length ())
output.append (input.charAt (pos));
return output.toString ();
}
/* The following are used for debugging purpose
public static String d(REMatch m) {
if (m == null) return "null";
else return "[" + m.index + "]";
}
public String substringUptoIndex(CharIndexed input) {
StringBuffer sb = new StringBuffer();
for (int i = 0; i < index; i++) {
sb.append(input.charAt(i));
}
return sb.toString();
}
*/
}