: However, I am doing a DisMax query and when I debug it shows that it is
: falling back on the default field in the schema, not the qf fields. Looking
: at the code in SolrHighlighter confirms this. Is it a bug or should I change
: the wiki (or both)?
it is suppose to work that way. In 1.2 you can see this code arround line
362 of DisMaxRequestHandler.java...
/* * * Highlighting/Summarizing * * */
if(HighlightingUtils.isHighlightingEnabled(req) && parsedUserQuery !=
null) {
String[] highFields = queryFields.keySet().toArray(new String[0]);
NamedList sumData =
HighlightingUtils.doHighlighting(
results.docList,
parsedUserQuery.rewrite(req.getSearcher().getReader()),
req,
highFields);
when dismaxed was changed to a subclass of standard request handler, the
field selection seems to have been migrated to
DisMaxQParserPlugin.getDefaultHighlightFields() ... which at first glance
seems like it should work (assuming the HighlightComponent it's getting
ResponseBuilder that has the QParser set appropriately)
i haven't tested it to verify that it isn't working as you say -- but
assuming you are correct, it's definitely bug.
-Hoss
