lucene-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Sujit Pal (Created) (JIRA)" <>
Subject [jira] [Created] (SOLR-3163) FacetComponent returns list of alternate name, value instead of map for counts in JSON responses
Date Sat, 25 Feb 2012 17:39:48 GMT
FacetComponent returns list of alternate name, value instead of map for counts in JSON responses

                 Key: SOLR-3163
             Project: Solr
          Issue Type: Bug
          Components: SearchComponents - other
    Affects Versions: 3.5
            Reporter: Sujit Pal
            Priority: Minor

Facet counts for facet.field come back in JSON format like this (a list of name and values,
where even positions are keys and odd positions are values).


unlike facet.query facet counts, which come back like this (a map of name value pairs).

      "u_reviewdate:[NOW-6MONTH TO NOW]":42,
      "u_reviewdate:[NOW-1YEAR TO NOW-6MONTHS]":1561,
      "u_reviewdate:[NOW-2YEAR TO NOW-1YEAR]":3017,
      "u_reviewdate:[NOW-5YEAR TO NOW-2YEAR]":1538,
      "u_reviewdate:[NOW-100YEAR TO NOW-5YEAR]":47},

This is only for wt=json. With wt=xml, the two formats look similar.

<lst name="facet_queries">
  <int name="u_reviewdate:[NOW-6MONTH TO NOW]">42</int>
  <int name="u_reviewdate:[NOW-1YEAR TO NOW-6MONTHS]">1561</int>
  <int name="u_reviewdate:[NOW-2YEAR TO NOW-1YEAR]">3017</int>
  <int name="u_reviewdate:[NOW-5YEAR TO NOW-2YEAR]">1538</int>
  <int name="u_reviewdate:[NOW-100YEAR TO NOW-5YEAR]">47</int>

<lst name="facet_fields">
  <lst name="u_idx">
    <int name="section">39663</int>
    <int name="adam">6211</int>
    <int name="image">3840</int>
    <int name="slideshow">2480</int>

I believe this is because of the use of NamedList for this data structure in FacetCountsComponent,
SimpleOrderedMap should be used instead. This will produce results of facet_field counts like
a map in JSON format (and keep the XML output the same).


This makes it consistent across different kinds of facets and is easier to consume on the

There is the concern about backward compatibility (ie users are already consuming the incorrect/inconsistent
format so we can't change it), but if that is not as important as consistency and ease of
use, then I can supply a patch for this.

This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:!default.jspa
For more information on JIRA, see:


To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message