[
https://issues.apache.org/jira/browse/SOLR-14916?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Joel Bernstein updated SOLR-14916:
----------------------------------
Description:
Currently the *timeseries* function only supports aggregations across the time
dimension. This ticket will add the *split* parameter which will add a top
level split by categorical field, to produce time lines per each split. The
limit parameter will also be added and be applied only to split. The sort will
always be desc on the first aggregation in the first iteration to keep the
parameters simple.
Sample syntax:
{code}
timeseries(collection1,
q="*:*",
field="timefield",
split="company",
limit=10,
gap="+1DAY",
format="YYYY-dd-MM" ,
avg(price_f))
{code}
The output of this can be easily pivoted into a matrix and correlated or
clustered like the output of the *facet2D* function. The *diff* and
*minMaxScale* functions already support operations over matrix rows so it's
very easy to perform clustering etc.. on this output.
was:
Currently the *timeseries* function only supports aggregations across the time
dimension. This ticket will add the *split* parameter which will add a top
level split by categorical field, to produce time lines per each split. The
limit parameter will also be added and be applied only to split. The sort will
always be desc on the first aggregation in the first iteration to keep the
parameters simple.
Sample syntax:
{code}
timeseries(collection1,
q="*:*",
field="timefield",
split="company",
limit=10,
gap="+1DAY",
format="YYYY-dd-MM" ,
avg(price_f))
{code}
The output of this can be easily pivoted into a matrix and correlated or
clustered like the output of the *facet2D* function. The *diff* and
*minMaxScale* functions already support operations over matrix rows so it's
very easy to perform clustering etc.. on this output.
> Add split parameter to timeseries Streaming Expression
> ------------------------------------------------------
>
> Key: SOLR-14916
> URL: https://issues.apache.org/jira/browse/SOLR-14916
> Project: Solr
> Issue Type: Improvement
> Reporter: Joel Bernstein
> Priority: Major
>
> Currently the *timeseries* function only supports aggregations across the
> time dimension. This ticket will add the *split* parameter which will add a
> top level split by categorical field, to produce time lines per each split.
> The limit parameter will also be added and be applied only to split. The sort
> will always be desc on the first aggregation in the first iteration to keep
> the parameters simple.
> Sample syntax:
> {code}
> timeseries(collection1,
> q="*:*",
> field="timefield",
> split="company",
> limit=10,
> gap="+1DAY",
> format="YYYY-dd-MM" ,
> avg(price_f))
> {code}
> The output of this can be easily pivoted into a matrix and correlated or
> clustered like the output of the *facet2D* function. The *diff* and
> *minMaxScale* functions already support operations over matrix rows so it's
> very easy to perform clustering etc.. on this output.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@solr.apache.org
For additional commands, e-mail: issues-h...@solr.apache.org
