Apache
/
lucene
mirror of https://github.com/apache/lucene.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

LUCENE-2568: fix AutomatonTestUtil.RandomAcceptedString to not return invalid (containing code points in the UTF16 surrogates range) UTF32 

git-svn-id: https://svn.apache.org/repos/asf/lucene/dev/trunk@979639 13f79535-47bb-0310-9956-ffa450edef68
This commit is contained in:
Michael McCandless 2010-07-27 10:56:04 +00:00
parent c76e15a33d
commit 96a6898ce7
1 changed files with 37 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

43
lucene/src/test/org/apache/lucene/util/automaton/AutomatonTestUtil.java
View File

@ -77,12 +77,44 @@ public class AutomatonTestUtil {
return new String(buffer, 0, end);
}
// picks a random int code point that this transition
// accepts, avoiding the surrogates range since they are
// "defined" in UTF32. Don't call this on a transition
// that only accepts UTF16 surrogate values!!
// picks a random int code point, avoiding surrogates;
// throws IllegalArgumentException if this transition only
// accepts surrogates
private static int getRandomCodePoint(final Random r, final Transition t) {
return t.min+r.nextInt(t.max-t.min+1);
final int code;
if (t.max < UnicodeUtil.UNI_SUR_HIGH_START ||
t.min > UnicodeUtil.UNI_SUR_HIGH_END) {
// easy: entire range is before or after surrogates
code = t.min+r.nextInt(t.max-t.min+1);
} else if (t.min >= UnicodeUtil.UNI_SUR_HIGH_START) {
if (t.max > UnicodeUtil.UNI_SUR_LOW_END) {
// after surrogates
code = 1+UnicodeUtil.UNI_SUR_LOW_END+r.nextInt(t.max-UnicodeUtil.UNI_SUR_LOW_END+1);
} else {
throw new IllegalArgumentException("transition accepts only surrogates: " + t);
}
} else if (t.max <= UnicodeUtil.UNI_SUR_LOW_END) {
if (t.min < UnicodeUtil.UNI_SUR_HIGH_START) {
// before surrogates
code = t.min + r.nextInt(UnicodeUtil.UNI_SUR_HIGH_START - t.min);
} else {
throw new IllegalArgumentException("transition accepts only surrogates: " + t);
}
} else {
// range includes all surrogates
int gap1 = UnicodeUtil.UNI_SUR_HIGH_START - t.min;
int gap2 = t.max - UnicodeUtil.UNI_SUR_LOW_END;
int c = r.nextInt(gap1+gap2);
if (c < gap1) {
code = t.min + c;
} else {
code = UnicodeUtil.UNI_SUR_LOW_END + c - gap1 + 1;
}
}
assert code >= t.min && code <= t.max && (code < UnicodeUtil.UNI_SUR_HIGH_START || code > UnicodeUtil.UNI_SUR_LOW_END):
"code=" + code + " min=" + t.min + " max=" + t.max;
return code;
}
public static class RandomAcceptedStrings {
@ -206,7 +238,6 @@ public class AutomatonTestUtil {
} else {
t = s.transitionsArray[r.nextInt(s.numTransitions)];
}
soFar.add(getRandomCodePoint(r, t));
s = t.to;
}