I think grouping by a regex of a tag value does make sense, and could be a powerful tool. While a SUBSTRING() function would also perhaps work, functions aren't currently supported outside the SELECT clause, there are no text manipulation functions, and functions are only valid on field values, not tags. So the regex approach is much less overall change for the system.
Can you open a feature request <https://github.com/influxdata/influxdb/issues/new> describing your use case? On Mon, Sep 26, 2016 at 3:01 PM, Sam Hatchett <[email protected]> wrote: > In trying to combine InfluxDB with Grafana's World Map plugin for > displaying time series data spatially, I've identified the need to > spatially subsample because I just have too many geographical points to > display all at once - too much lag time to return the result set. The > location's geohash provides a great way to do this (encoded as a tag value, > so I can group-by geohash to return some aggregate of potentially hundreds > of location's values), and then I can select the max, mean, min, etc., for > each spatial region, but the level of spatial aggregation has to be decided > prior to writing data to Influx. > > What I'm wondering about are the relative merits of workarounds, or of > finding some way to decide on the geohash length when the query is executed. > > So for instance, have multiple tags: (geo8, geo9, geo10 ...) for each > series so that I can decide on spatial subsampling when querying and choose > the right length: > select "value" from "temperature" where ... group by "geo8" > > But what I really want to do is store the full 12-character geohash for > each location and group by a substring of that geohash: > select "value" from "temperature" where ... group by substr("geohash",0,8) > > It's also useful to note that I can currently limit the locations that are > queried by using a regex: > select "value" from "temperature" where "geohash" =~ /^gbsuv7s/ > > ... but can't use a substring of the hash to do the group-by, which would > take me the rest of the way there. > > Any thoughts or comments would be most appreciated. I don't think that > there's a really straightforward solution right now, but does using a > substring (or other arbitrary transformation) of a tag value for grouping > make sense to anyone else? Seems like there could be many more uses of this > type of approach. > > -- > Remember to include the InfluxDB version number with all issue reports > --- > You received this message because you are subscribed to the Google Groups > "InfluxDB" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To post to this group, send email to [email protected]. > Visit this group at https://groups.google.com/group/influxdb. > To view this discussion on the web visit https://groups.google.com/d/ > msgid/influxdb/2ea6b448-16a6-45e9-85c1-afff36b1a4d4%40googlegroups.com > <https://groups.google.com/d/msgid/influxdb/2ea6b448-16a6-45e9-85c1-afff36b1a4d4%40googlegroups.com?utm_medium=email&utm_source=footer> > . > For more options, visit https://groups.google.com/d/optout. > -- Sean Beckett Director of Support and Professional Services InfluxDB -- Remember to include the InfluxDB version number with all issue reports --- You received this message because you are subscribed to the Google Groups "InfluxDB" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at https://groups.google.com/group/influxdb. To view this discussion on the web visit https://groups.google.com/d/msgid/influxdb/CALGqCvPnzaSYODzZJ6UVS5Vc21tx5Df3r%2BWmVHdG_AmHvg7Usg%40mail.gmail.com. For more options, visit https://groups.google.com/d/optout.
