/**
* finds the next match in the current input, appends it to <code>out</code> and returns the
* {@link FaAction} associated with the match. Input is read until a match is found, {@link
* #maxCopy} is reached or EOF is hit. Non-matching input is handled according to {@link
* #setOnFailedMatch setOnFailedMatch()}. In particular:
*
* <dl>
* <dt>{@link #UNMATCHED_COPY}
* <dd>will append up to {@link #maxCopy} non-matching characters in front of the match. If
* <code>maxCopy</code> is reached before the match, <b>no matching text is returned</b>,
* only the non-matching characters. In this case the return value is <code>null</code>, and
* should <code>maxCopy</code> be ≤ 1, then 1 character is always delivered. If a
* match is found before <code>maxCopy</code> is reached, the match is appended to <code>out
* </code>. To find out where the match actually starts, call {@link #matchStart()}.
* <dt>{@link #UNMATCHED_DROP}
* <dd>will drop (delete) unmatched text. In this case the matching text is the only text
* appended to <code>out</code>.
* <dt>{@link #UNMATCHED_THROW}
* <dd>causes a {@link monq.jfa.NomatchException} to be thrown. No text will be appended to
* <code>out</code> and the offenting text will still be available in the {@link CharSource}
* serving as input to <code>this</code>.
* </dl>
*
* <p><b>Hint:</b> Use this method if you are interested only in a simple tokenization of the
* input. The actions returned may serve as the token type. If you however want to apply the
* actions returned immediately to the match, then rather use one of the <code>read</code> or
* <code>filter</code> methods. If you find yourself using <code>if</code> statements on the
* <code>FaAction</code> returned, you are definitively doing something wrong.
*
* @return
* <dl>
* <dt>eofAction
* <dd>When EOF is hit the first time and the <code>Dfa</code> operated has a action set for
* EOF which is not <code>null</code> this is returned (see {@link Nfa#compile
* Nfa.compile()}).
* <dt>{@link #EOF}
* <dd>if EOF is hit and <code>eofAction</code> was already delivered or is <code>null
* </code>. The output may have non-matching input that was found just before EOF.
* <dt><code>null</code>
* <dd>if <code>UNMATCHED_COPY</code> is active and <code>maxCopy</code> non-matching
* characters where found before a match was encountered.
* <dt>an action
* <dd>found for a match.
* </dl>
*/
public FaAction next(StringBuilder out) throws java.io.IOException {
matchStart = out.length();
FaAction a = dfa.match(in, out, smd);
if (a == null) {
// There was no match, so we have to search for the first
// match. Note: there is always at least one character available as
// long as not Dfa.EOF is returned by dfa.match()
if (onFailedMatch == UNMATCHED_COPY) {
int unmatched = 0;
do {
out.append((char) (in.read()));
unmatched += 1;
a = dfa.match(in, out, smd);
} while (a == null && unmatched < maxCopy);
matchStart += unmatched;
} else if (onFailedMatch == UNMATCHED_DROP) {
do {
in.read();
a = dfa.match(in, out, smd);
} while (a == null);
} else {
// everything else is a failure
String emsg = lookahead();
throw new NomatchException(
"no matching regular expression " + "when looking at `" + emsg + "'");
}
}
// We handle EOF and eofAction as if we have found a match
if (a == EOF && dfa.eofAction != null && eofArmed) {
eofArmed = false;
return dfa.eofAction;
}
return a;
}
/**
* fetch a bit of lookahead for use in messages for exceptions. The lookahead is pushed back into
* the input afterwards.
*/
private String lookahead() {
// Read up to 30 chars for a decent error message
StringBuilder sb = new StringBuilder(30);
int i;
try {
for (i = 0; i < 30; i++) {
int ch = in.read();
if (ch == -1) break;
sb.append((char) ch);
}
} catch (java.io.IOException e) {
in.pushBack(sb, 0);
return "IOException when trying to generate context info";
}
String result;
if (i == 30) result = sb.substring(0, 27) + "...";
else result = sb.toString() + "[EOF]";
in.pushBack(sb, 0);
return result;
}
/**
* reads one character immediately from the input source and returns it without filtering. If
* filtered characters are already available because of a previous {@link #read()}, these are not
* touched and will be used in the next call to one of the <code>read()</code> functions.
*/
public int skip() throws java.io.IOException {
return in.read();
}