You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@lucene.apache.org by "Shalin Shekhar Mangar (JIRA)" <ji...@apache.org> on 2015/07/01 22:08:06 UTC
[jira] [Commented] (SOLR-6352) Let Stats Hang off of Range Facets
[ https://issues.apache.org/jira/browse/SOLR-6352?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14610926#comment-14610926 ]
Shalin Shekhar Mangar commented on SOLR-6352:
---------------------------------------------
An example of the current response format for range facets is as follows:
{code}
curl 'http://localhost:8983/solr/test/select' --data-binary 'q=price:[0 TO 100]&facet=true&facet.range=price&facet.range.start=0&facet.range.end=1000&facet.range.gap=100&wt=json&indent=on&stats=on&stats.field=price&facet.range.other=all'
{code}
{code}
"facet_counts":{
"facet_queries":{},
"facet_fields":{},
"facet_dates":{},
"facet_ranges":{
"price":{
"counts":[
"0.0",7,
"100.0",2,
"200.0",1,
"300.0",3,
"400.0",1,
"500.0",0,
"600.0",1,
"700.0",0,
"800.0",0,
"900.0",0],
"gap":100.0,
"start":0.0,
"end":1000.0,
"before":0,
"after":1,
"between":15}},
"facet_intervals":{},
"facet_heatmaps":{}},
"stats":{
"stats_fields":{
"price":{
"min":0.0,
"max":2199.0,
"count":16,
"missing":16,
"sum":5251.270030975342,
"sumOfSquares":6038619.175900028,
"mean":328.20437693595886,
"stddev":536.3536996709846}}}}
{code}
In order to hang stats off of range facets, I propose the following response format:
{code}
curl 'http://localhost:8983/solr/test/select' --data-binary 'q=price:[0 TO 100]&facet=true&facet.range={!stats=s1}price&facet.range.start=0&facet.range.end=1000&facet.range.gap=100&wt=json&indent=on&stats=on&stats.field={!tag=s1}price&facet.range.other=all'
{code}
{code}
"facet_counts":{
"facet_queries":{},
"facet_fields":{},
"facet_dates":{},
"facet_ranges":{
"price":{
"counts":[
{
"start" : "0.0",
"count" : "7"
"stats":{
"stats_fields":{
"price":{
"min":0.0,
"max":92.0,
"count":7,
"missing":0,
"sum":198.43999862670898,
"sumOfSquares":14617.752310049444,
"mean":28.348571232387,
"stddev":38.7131800847063
}
}
}
},
{
"start" : "100.0"
"count" : "2"
"stats":{
"stats_fields":{
"price":{
"min":179.99000549316406,
"max":185.0,
"count":2,
"missing":0,
"sum":364.99000549316406,
"sumOfSquares":66621.40207742923,
"mean":182.49500274658203,
"stddev":3.5426010894910442
}
}
}
},
...
],
"gap":100.0,
"start":0.0,
"end":1000.0,
"before": {
"count" : "0",
"stats":{
"stats_fields":{
"price":{
"min":0,
"max":0,
"count":0,
"missing":0,
"sum":0,
"sumOfSquares":0,
"mean":0,
"stddev":0
}
}
}
},
"after":{
"count" : "1",
"stats":{
"stats_fields":{
"price":{
"min":2199.0,
"max":2199.0,
"count":1,
"missing":0,
"sum":2199.0,
"sumOfSquares":4835601.0,
"mean":2199.0,
"stddev":0.0
}
}
}
},
"between":{
"count" : "15",
"stats":{
"stats_fields":{
"price":{
"min":0.0,
"max":649.989990234375,
"count":15,
"missing":0,
"sum":3052.270030975342,
"sumOfSquares":1203018.175900028,
"mean":203.48466873168945,
"stddev":203.8781183115027
}
}
}
}}},
"facet_intervals":{},
"facet_heatmaps":{}},
"stats":{
"stats_fields":{
"price":{
"min":0.0,
"max":2199.0,
"count":16,
"missing":16,
"sum":5251.270030975342,
"sumOfSquares":6038619.175900028,
"mean":328.20437693595886,
"stddev":536.3536996709846}}}}
{code}
Thoughts?
> Let Stats Hang off of Range Facets
> ----------------------------------
>
> Key: SOLR-6352
> URL: https://issues.apache.org/jira/browse/SOLR-6352
> Project: Solr
> Issue Type: Sub-task
> Reporter: Hoss Man
> Assignee: Shalin Shekhar Mangar
>
> The same basic idea as the sibling pivot issue: ask the RangeFacet code to compute some stats X for each range it generates. We can do this with the existing {{stats.field}} params, but we'll leverage the {{tag}} local param of the {{stats.field}} instances to be able to associate which stats we want hanging off of which {{facet.range}} instance
> Example...
> {noformat}
> facet.range={!stats=s1}timestamp
> facet.range={!stats=s2}bytes_sent
> stats.field={!key=sum_bytes tag=s1 sum=true}bytes_sent
> stats.field={!tag=s1 percentiles=99}request_duration
> stats.field={!tag=s2 mean=true percentiles=99}request_duration
> {noformat}
> ...with the request above, (hypothetically over an index of web log traffic) each range bucket over the "timestamp" field will (in addition to simple facet count) also include the sum of total bytes_sent, and the 99th percentile of request_duration for all docs in that time range; and each range bucket over the "bytes_sent" field will (in addition to simple facet count) also include the mean and 99th percentile of request_duration.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org