Get results with exact match - lucene

Question

I want to do a query like that : "banana apple cherry" on a "fruit" field.
All the fruits in the desserts needs to be in the query, but not all the fruits in the query needs to be in the desserts..
Here's an example..
NAME FRUIT
Dessert1 banana apple OK (we got banana and apple in the query)
Dessert2 cherry apple banana OK(the order doesn't matter)
Dessert3 cherry apple banana melon NO (melon is missing in the query)
public class ArrayStringFieldBridge implements TwoWayFieldBridge{
#Override
public Object get(String name, Document document) {
IndexableField[] fields = document.getFields(name);
String[] values = new String[fields.length];
for (int i=0; i<fields.length; i++) {
values[i] = fields[i].stringValue();
}
return values;
}
#Override
public String objectToString(Object value) {
return StringUtils.join((String[])value, " ");
}
#Override
public void set(String name, Object value, Document document, LuceneOptions luceneOptions) {
String newString = StringUtils.join((String[])value, " ");
Field field = new Field(name, newString, luceneOptions.getStore(), luceneOptions.getIndex(), luceneOptions.getTermVector());
field.setBoost(luceneOptions.getBoost());
document.add(field);
}
}
#Indexed
#AnalyzerDef(name = "customanalyzer",
tokenizer = #TokenizerDef(factory = StandardTokenizerFactory.class))
public class Dessert {
#Analyzer(definition="customanalyzer")
#Field(name = "equipment", index=Index.YES, analyze = Analyze.YES, store=Store.YES)
#FieldBridge(impl=ArrayStringFieldBridge.class)
public String[] fruits = new String[]{};
}
Even if you are not using hibernate-search, every suggestions about the theory to handle that would be great... Thank you

Answer 1

Step 1 : Fire lucene query "fruit:banana OR fruit:apple OR fruit:cherry"
Step 2 : Gather all matched dessert documents
Step 3 : Post process your match dessert document with query
convert match document to array of terms matchDocArr : {banana, apple}
convert query terms to array - queryArr : {banana, apple, cherry}
iterate over matchDocArr and make sure each term of matchDocArr is found in queryArr by array, if NOT (melon use case) knockout this matched document
Here is an example function which needs to be called for every matched doc
public static boolean isDocInterested(String query, String matchDoc)
{
List<String> matchDocArr = new ArrayList<String>();
matchDocArr = Arrays.asList(matchDoc.split(" "));
List<String> queryArr = new ArrayList<String>();
queryArr = Arrays.asList(query.split(" "));
int matchCounter = 0;
for(int i=0; i<matchDocArr.size(); i++)
{
if (queryArr.contains(matchDocArr.get(i)))
matchCounter++;
}
if (matchCounter == matchDocArr.size())
return true;
return false;
}
if function returns TRUE we are interested in doc/dessert, if it returns FALSE ignore this doc/dessert.
of course this function can be written in many different ways but I think you get the point.