-//\r
-// This software is now distributed according to\r
-// the Lesser Gnu Public License. Please see\r
-// http://www.gnu.org/copyleft/lesser.txt for\r
-// the details.\r
-// -- Happy Computing!\r
-//\r
-package com.stevesoft.pat;\r
-\r
-import java.util.*;\r
-\r
-/**\r
- Shareware: package pat\r
- <a href="copyright.html">Copyright 2001, Steven R. Brandt</a>\r
- */\r
-/**\r
- The RegexTokenizer is similar to the StringTokenizer class\r
- provided with java, but allows one to tokenize using\r
- regular expressions, rather than a simple list of characters.\r
- Tokens are any strings between the supplied regular expression,\r
- as well as any backreferences (things in parenthesis)\r
- contained within the regular expression. */\r
-public class RegexTokenizer\r
- implements Enumeration\r
-{\r
- String toParse;\r
- Regex r;\r
- int count = 0;\r
- Vector v = new Vector();\r
- Vector vi = new Vector();\r
- int pos = 0;\r
-\r
- int offset = 1;\r
- void getMore()\r
- {\r
- String s = r.right();\r
- if (r.searchFrom(toParse, pos))\r
- {\r
- v.addElement(r.left().substring(pos));\r
- vi.addElement(new Integer(r.matchFrom() +\r
- r.charsMatched()));\r
- for (int i = 0; i < r.numSubs(); i++)\r
- {\r
- if (r.substring() != null)\r
- {\r
- v.addElement(r.substring(i + offset));\r
- vi.addElement(\r
- new Integer(r.matchFrom(i + offset) +\r
- r.charsMatched(i + offset)));\r
- }\r
- }\r
- pos = r.matchFrom() + r.charsMatched();\r
- }\r
- else if (s != null)\r
- {\r
- v.addElement(s);\r
- }\r
- }\r
-\r
- /** Initialize the tokenizer with a string of text and a pattern */\r
- public RegexTokenizer(String txt, String ptrn)\r
- {\r
- toParse = txt;\r
- r = new Regex(ptrn);\r
- offset = Regex.BackRefOffset;\r
- getMore();\r
- }\r
-\r
- /** Initialize the tokenizer with a Regex object. */\r
- public RegexTokenizer(String txt, Regex r)\r
- {\r
- toParse = txt;\r
- this.r = r;\r
- offset = Regex.BackRefOffset;\r
- getMore();\r
- }\r
-\r
- /** This should always be cast to a String, as in StringTokenizer,\r
- and as in StringTokenizer one can do this by calling\r
- nextString(). */\r
- public Object nextElement()\r
- {\r
- if (count >= v.size())\r
- {\r
- getMore();\r
- }\r
- return v.elementAt(count++);\r
- }\r
-\r
- /** This is the equivalent (String)nextElement(). */\r
- public String nextToken()\r
- {\r
- return (String) nextElement();\r
- }\r
-\r
- /** This asks for the next token, and changes the pattern\r
- being used at the same time. */\r
- public String nextToken(String newpat)\r
- {\r
- try\r
- {\r
- r.compile(newpat);\r
- }\r
- catch (RegSyntax r_)\r
- {}\r
- return nextToken(r);\r
- }\r
-\r
- /** This asks for the next token, and changes the pattern\r
- being used at the same time. */\r
- public String nextToken(Regex nr)\r
- {\r
- r = nr;\r
- if (vi.size() > count)\r
- {\r
- pos = ( (Integer) vi.elementAt(count)).intValue();\r
- v.setSize(count);\r
- vi.setSize(count);\r
- }\r
- getMore();\r
- return nextToken();\r
- }\r
-\r
- /** Tells whether there are more tokens in the pattern. */\r
- public boolean hasMoreElements()\r
- {\r
- if (count >= v.size())\r
- {\r
- getMore();\r
- }\r
- return count < v.size();\r
- }\r
-\r
- /** Tells whether there are more tokens in the pattern, but\r
- in the fashion of StringTokenizer. */\r
- public boolean hasMoreTokens()\r
- {\r
- return hasMoreElements();\r
- }\r
-\r
- /** Determines the # of remaining tokens */\r
- public int countTokens()\r
- {\r
- int _count = count;\r
- while (hasMoreTokens())\r
- {\r
- nextToken();\r
- }\r
- count = _count;\r
- return v.size() - count;\r
- }\r
-\r
- /** Returns all tokens in the String */\r
- public String[] allTokens()\r
- {\r
- countTokens();\r
- String[] ret = new String[v.size()];\r
- v.copyInto(ret);\r
- return ret;\r
- }\r
-};\r
+//
+// This software is now distributed according to
+// the Lesser Gnu Public License. Please see
+// http://www.gnu.org/copyleft/lesser.txt for
+// the details.
+// -- Happy Computing!
+//
+package com.stevesoft.pat;
+
+import java.util.Enumeration;
+import java.util.Vector;
+
+/**
+ Shareware: package pat
+ <a href="copyright.html">Copyright 2001, Steven R. Brandt</a>
+ */
+/**
+ * The RegexTokenizer is similar to the StringTokenizer class provided with
+ * java, but allows one to tokenize using regular expressions, rather than a
+ * simple list of characters. Tokens are any strings between the supplied
+ * regular expression, as well as any backreferences (things in parenthesis)
+ * contained within the regular expression.
+ */
+public class RegexTokenizer implements Enumeration
+{
+ String toParse;
+
+ Regex r;
+
+ int count = 0;
+
+ Vector v = new Vector();
+
+ Vector vi = new Vector();
+
+ int pos = 0;
+
+ int offset = 1;
+
+ void getMore()
+ {
+ String s = r.right();
+ if (r.searchFrom(toParse, pos))
+ {
+ v.addElement(r.left().substring(pos));
+ vi.addElement(new Integer(r.matchFrom() + r.charsMatched()));
+ for (int i = 0; i < r.numSubs(); i++)
+ {
+ if (r.substring() != null)
+ {
+ v.addElement(r.substring(i + offset));
+ vi.addElement(new Integer(r.matchFrom(i + offset)
+ + r.charsMatched(i + offset)));
+ }
+ }
+ pos = r.matchFrom() + r.charsMatched();
+ }
+ else if (s != null)
+ {
+ v.addElement(s);
+ }
+ }
+
+ /** Initialize the tokenizer with a string of text and a pattern */
+ public RegexTokenizer(String txt, String ptrn)
+ {
+ toParse = txt;
+ r = new Regex(ptrn);
+ offset = Regex.BackRefOffset;
+ getMore();
+ }
+
+ /** Initialize the tokenizer with a Regex object. */
+ public RegexTokenizer(String txt, Regex r)
+ {
+ toParse = txt;
+ this.r = r;
+ offset = Regex.BackRefOffset;
+ getMore();
+ }
+
+ /**
+ * This should always be cast to a String, as in StringTokenizer, and as in
+ * StringTokenizer one can do this by calling nextString().
+ */
+ public Object nextElement()
+ {
+ if (count >= v.size())
+ {
+ getMore();
+ }
+ return v.elementAt(count++);
+ }
+
+ /** This is the equivalent (String)nextElement(). */
+ public String nextToken()
+ {
+ return (String) nextElement();
+ }
+
+ /**
+ * This asks for the next token, and changes the pattern being used at the
+ * same time.
+ */
+ public String nextToken(String newpat)
+ {
+ try
+ {
+ r.compile(newpat);
+ } catch (RegSyntax r_)
+ {
+ }
+ return nextToken(r);
+ }
+
+ /**
+ * This asks for the next token, and changes the pattern being used at the
+ * same time.
+ */
+ public String nextToken(Regex nr)
+ {
+ r = nr;
+ if (vi.size() > count)
+ {
+ pos = ((Integer) vi.elementAt(count)).intValue();
+ v.setSize(count);
+ vi.setSize(count);
+ }
+ getMore();
+ return nextToken();
+ }
+
+ /** Tells whether there are more tokens in the pattern. */
+ public boolean hasMoreElements()
+ {
+ if (count >= v.size())
+ {
+ getMore();
+ }
+ return count < v.size();
+ }
+
+ /**
+ * Tells whether there are more tokens in the pattern, but in the fashion of
+ * StringTokenizer.
+ */
+ public boolean hasMoreTokens()
+ {
+ return hasMoreElements();
+ }
+
+ /** Determines the # of remaining tokens */
+ public int countTokens()
+ {
+ int _count = count;
+ while (hasMoreTokens())
+ {
+ nextToken();
+ }
+ count = _count;
+ return v.size() - count;
+ }
+
+ /** Returns all tokens in the String */
+ public String[] allTokens()
+ {
+ countTokens();
+ String[] ret = new String[v.size()];
+ v.copyInto(ret);
+ return ret;
+ }
+};