fix: wildcards in numeric range queries #75
Conversation
When the `*` symbol is used in ranges, the parsing dies with a `NullPointerException`. That is because input strings with the `null` value are not handled -- they should be treated as infinity with the respective sign (depending on the side of range where the wildcard is used).
| if (isNumber(lower) && isNumber(upper)) { | ||
| NumericRangeQuery.newDoubleRange(field, 8, lower.toDouble, | ||
| upper.toDouble, startInclusive, endInclusive) | ||
| val lb = Option(lower).getOrElse("-Infinity") |
There was a problem hiding this comment.
this appears to expand the syntax to mean any input that the parser yielded null for will be a valid open range, which boxes us in for future changes and I think is unintentional?
There was a problem hiding this comment.
I would prefer to see the NPE handled. If it occurs in;
private def isNumber(str: String): Boolean = {
fpRegex.matcher(str).matches()
}
then we should just change isNumber to return false if str is null.
There was a problem hiding this comment.
The null values in the second and third parameters of QueryParser/getRangeQuery() have a special meaning. It denotes the start and end of open ranges as suggested by the sources of Lucene 4.6.1:
boolean startOpen=false;
boolean endOpen=false;
if (goop1.kind == RANGE_QUOTED) {
goop1.image = goop1.image.substring(1, goop1.image.length()-1);
} else if ("*".equals(goop1.image)) {
startOpen=true;
}
if (goop2.kind == RANGE_QUOTED) {
goop2.image = goop2.image.substring(1, goop2.image.length()-1);
} else if ("*".equals(goop2.image)) {
endOpen=true;
}
q = getRangeQuery(field, startOpen ? null : discardEscapeChar(goop1.image), endOpen ? null : discardEscapeChar(goop2.image), startInc, endInc);You can also see in the toString() method of TermRangeQuery that nulls are uniquely translated back to stars:
buffer.append(includeLower ? '[' : '{');
buffer.append(lowerTerm != null ? ("*".equals(Term.toString(lowerTerm)) ? "\\*" : Term.toString(lowerTerm)) : "*");
buffer.append(" TO ");
buffer.append(upperTerm != null ? ("*".equals(Term.toString(upperTerm)) ? "\\*" : Term.toString(upperTerm)) : "*");
buffer.append(includeUpper ? ']' : '}');There was a problem hiding this comment.
aha, great, thanks for digging in.
When the
*(wildcard) symbol is used in ranges, the parsing dies with aNullPointerException. That is because input strings with thenullvalue are not handled -- they should be treated as infinity with the respective sign (depending on the side of range where the wildcard is used).