In a similar line though, BoostingFunctionTermQuery doesn't really fit with BoostingNearQuery. I see part of why its not called BoostingTermQuery is because BoostingTermQuery is deprecated - but why can't the BoostingFunctionTermQuery impl replace BoostingTermQuery with average as the default? (which is now how the deprecated BoostingTermQuery is implemented) There doesn't seem a good reason we abandoned it. And as all these types will take functions, there is no reason to have that in the name if BoostingNearQuery doesn't need it.
I'll commit the JavaDocs fixes now (BoostingFunctionTermQuery has the wrong javadoc as well), but I think more should be addressed here. - Mark Mark Miller wrote: > My fault - it does! I was just checking JavaDocs and this was misleading: > > Payload scores are averaged across term occurrences in the document. > > I'll update to be less ambiguous eg it takes a function that defaults to > average. > > > - Mark > > > Simon Willnauer wrote: > >> +1 >> >> On Thu, Aug 20, 2009 at 8:21 PM, Michael >> McCandless<[email protected]> wrote: >> >> >>> +1 >>> >>> Mike >>> >>> On Thu, Aug 20, 2009 at 11:08 AM, Mark Miller<[email protected]> wrote: >>> >>> >>>> BoostingNearQuery averages payloads - shouldn't it take a >>>> PayloadFunction as well? >>>> >>>> -- >>>> - Mark >>>> >>>> http://www.lucidimagination.com >>>> >>>> >>>> >>>> >>>> --------------------------------------------------------------------- >>>> To unsubscribe, e-mail: [email protected] >>>> For additional commands, e-mail: [email protected] >>>> >>>> >>>> >>>> >>> --------------------------------------------------------------------- >>> To unsubscribe, e-mail: [email protected] >>> For additional commands, e-mail: [email protected] >>> >>> >>> >>> >> --------------------------------------------------------------------- >> To unsubscribe, e-mail: [email protected] >> For additional commands, e-mail: [email protected] >> >> >> > > > -- - Mark http://www.lucidimagination.com --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
