Re: Using BasicAuth with SolrJ Code

Ok, thank you. Regards, Edwin On 15 April 2017 at 08:05, Noble Paul wrote: > I'll test with this and let you know > > On Apr 13, 2017 23:06, "Zheng Lin Edwin Yeo" wrote: > > > The security.json which I'm using is the default one that is available >

Re: Long GC pauses while reading Solr docs using Cursor approach

Chetas Joshi wrote: > Thanks for the insights into the memory requirements. Looks like cursor > approach is going to require a lot of memory for millions of documents. Sorry, that is a premature conclusion from your observations. > If I run a query that returns only 500K

Re: Need help with auto-suggester

I see, thanks. So I"m just using a string field to store the JSON. On Sat, Apr 15, 2017 at 11:15 PM, Walter Underwood wrote: > Sorry, that was formatted. The quotes are actually escaped, like this: > > {"term":"microsoft office","weight":14,"payload":"{\"count\": >

Re: Need help with auto-suggester

Sorry, that was formatted. The quotes are actually escaped, like this: {"term":"microsoft office","weight":14,"payload":"{\"count\": 1534255, \"id\": \"microsoft office\"}”} wunder Walter Underwood wun...@wunderwood.org http://observer.wunderwood.org/ (my blog) > On Apr 15, 2017, at 10:40

Re: Need help with auto-suggester

JSON does not have a binary data type, so true BLOBs are not possible in JSON. Sorry, I wasn’t clear. The payload I use is JSON in a string. It looks like this: suggest: { skill_names_infix: { m: { numFound: 10, suggestions: [ { term: "microsoft office", weight: 14, payload: "{"count": 1534255,

Re: Need help with auto-suggester

Hi - just wondering, what would be the difference between using a blob / binary field to store the JSON rather than simply using a string field? Thanks On Sat, Apr 15, 2017 at 2:50 AM, Walter Underwood wrote: > We recently needed multiple values in the payload, so I put a

Re: Return all docs with same last-value when sorting by non-unique-value

I don't think Solr supports this. Maybe you can do that in a custom component by cutting off in Solr at max-limit and then cutting again in the search component afterwards down to limit. Or just get more documents and deal with this on the middle-ware side. Regards, Alex.

Sharding strategy for optimal NRT performance

Hi! I am looking for some advice on an sharding strategy that will produce optimal performance in the NRT search case for my setup. I have come up with a strategy that I think will work based on my experience, testing, and reading of similar questions on the mailing list, but I was hoping to run

Re: Return all docs with same last-value when sorting by non-unique-value

Say order_by=likes descending, limit(4). And the likes are::: 10,9,8,7,7,7,4,2. Then we'd get back all 10-7 documents, so 6 docs. The same thing if they sort in the middle. It can also have a max-limit, so we don't get too many docs returned. Makes sense ? On Sat, Apr 15, 2017 at 8:24 AM,

Re: Return all docs with same last-value when sorting by non-unique-value

Not really making sense, no. Could you show an example? Also, you seem to imply that after sorting only the documents sorted at the end may have same values. What if they have the same values but sort into the middle? Regards, Alex http://www.solr-start.com/ - Resources for Solr users,