How does MultiFieldQueryParser work? Any example showing its mechanism please

Hi,- i am trying to understand what this api does and it seems it decomposes the input string by spaces and OR's them and then depending on occur field setting the whole OR'ed expression is either under "+" (i.e,must) or just OR'ed again. Such as: input string abc def occur values are : m

Re: SQL OR in lucene : where ((term1=a and term2=b) OR (term3=a and term4=b)) and context in (2,3,4,5.....200)

Hi Khurram, My understanding is that once you add a MUST or FILTER clause to a boolean query, SHOULD clauses become entirely optional, ie. none of them have to match the document being returned, unless you also set the "minimum number should match" property to a value >= 1 on the query. This doe

Re: SQL OR in lucene : where ((term1=a and term2=b) OR (term3=a and term4=b)) and context in (2,3,4,5.....200)

# is a required, non-scoring term (Occur.FILTER) - semantically it functions like Occur.MUST. It looks to me as if the Shehzad's query is missing a clause, eg based on his description it should look like: +((+term1:a +term2:b) (+term3:a and +term4:b)) #context:2 4 7 ... 198 I agree w/Tomoko,

Re: SQL OR in lucene : where ((term1=a and term2=b) OR (term3=a and term4=b)) and context in (2,3,4,5.....200)

Hi Khurram, Lucene query parser's OR operator or Occur.Should are work as expected for me, I suspect you could miss some points. > roughly in lucene > (+term1:a +term2:b) (+term3:a and +term4:b) #context:2 4 7 ... 198 It does not look like an ordinary lucene query string to me ('#' is not an ope

SQL OR in lucene : where ((term1=a and term2=b) OR (term3=a and term4=b)) and context in (2,3,4,5.....200)

Hi, I have a requirement to replicate following SQL query logic containing OR condition as where ((term1=a and term2=b) OR (term3=a and term4=b)) and context in (2,3,4,5.200) roughly in lucene (+term1:a +term2:b) (+term3:a and +term4:b) #context:2 4 7 ... 198 It doesn't seem to me g

Re: Replacement of CollapsingTopDocsCollector

Hi, I'm wondering if you mean Elasticsearch's org.apache.lucene.search.grouping.CollapsingTopDocsCollector. Elasticsearch 6.4.0, the latest version, is shipped with this class. https://github.com/elastic/elasticsearch/tree/v6.4.0/server/src/main/java/org/apache/lucene/search/grouping I'm not sure

Re: Question about BytesRef and BinaryDocValues

Kevin, the sequence is the following: get terms for the field, get postings for a term and further get payload from the postings. Have a read a little about reverse index structure and it will be more clear to you. Your Query creates Weight, that must create a scorer in the method scorer(context).