Hi,
I have a use-case where a single Document in Lucene contains a single
NumericField that could potentially have a 100s of 1000s of values.
Values are being added to a document instance like so:
List<Fieldable> fields = .... // get fields, possibly 100s of 1000s
with the same name, but a different value
for (Fieldable field : fields) {
doc.add(field);
}
Internally, doc adds the field to an ArrayList of Fieldables: List<Fieldable> .
This means there is quite a significant memory overhead for very large amounts
of fields.
Has anyone toyed with the idea of storing the fields in Document in a map,
keyed off Field with the map-values being a list of values?
Or a new MultiValueField type, which contains an array of field values, rather
than a single value per Field ?
Are there any other known solutions or work-arounds for this problem ?
Cheers,
Nick
---------------------------------------------------------------------
To unsubscribe, e-mail: java-user-unsubscr...@lucene.apache.org
For additional commands, e-mail: java-user-h...@lucene.apache.org
