[jira] [Updated] (SOLR-11075) Refactor handling of params in CloudSolrStream and FacetStream

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

[jira] [Updated] (SOLR-11075) Refactor handling of params in CloudSolrStream and FacetStream

JIRA jira@apache.org

     [ https://issues.apache.org/jira/browse/SOLR-11075?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Erick Erickson updated SOLR-11075:
----------------------------------
    Attachment: SOLR-11075.patch

straw-man patch, [~joel.bernstein][~dpgove] any comments?

It would be possible to just make the calls to these classes use ModifiableSolrParams, but that would involve changing the c'tors which is a back-compat issue. Plus I kind of like having a local copy.

> Refactor handling of params in CloudSolrStream and FacetStream
> --------------------------------------------------------------
>
>                 Key: SOLR-11075
>                 URL: https://issues.apache.org/jira/browse/SOLR-11075
>             Project: Solr
>          Issue Type: Improvement
>      Security Level: Public(Default Security Level. Issues are Public)
>            Reporter: Erick Erickson
>            Assignee: Erick Erickson
>         Attachments: SOLR-11075.patch
>
>
> We started to look more closely at how toExpression is used in these classes and the more we look the more puzzled we became (Varun and I that is).
> Is there any reason other than history why the params are pulled apart then reconstructed into comma-separated lists when there are more than one of any particular parameter? I suspect than when I worked on SOLR-8467 I didn't delve deeply enough here.
> [~dpgove][~joel.bernstein] [~risdenk] in particular we'd like your opinion. Arguably this is going to lead to anomalies, i.e. differences in what streaming selects .vs. what standard Solr would select.
> For instance, let's say the user puts two "q" parameters in. Standard Solr parsing uses the first one encountered. what happens when we get q=clause1,clause2 as a result of the toExpression is anybody's guess. It just shouldn't be different than straight-up Solr IMO.
> "fl" parameters on the other hand are all honored, as are "fq" clauses.
> Multiple "sort" clauses it appears first one wins.
> So my question is whether it makes sense to just add the parameter multiple times, presumably reflecting the actual query.
> Assigning to myself but someone else should feel free to take it



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Loading...