#general
@ronak: Is regex_like function supported on map column? e.g query ```SELECT mapValue(mapDim1__KEYS, 'k1', mapDim1__VALUES) FROM FeatureTest1 WHERE REGEXP_LIKE(mapValue(mapDim1__KEYS, 'k1', mapDim1__VALUES), '^p.*') LIMIT 5``` Tried similar query on (pinot 0.7.1) and getting exception.
@mayanks: We are moving towards JSON data type, instead of MAP/List etc. For JSON data type, you can set a JSON index as well as a text index and do TEXT_MATCH queries. However, the text index is on the entire json blob treated as text, as opposed to individual fields.
@laabidi.raissi: @laabidi.raissi has joined the channel
@nesrullayev.ali: @nesrullayev.ali has joined the channel
@orbit: @orbit has left the channel
@orbit: @orbit has joined the channel
#random
@laabidi.raissi: @laabidi.raissi has joined the channel
@nesrullayev.ali: @nesrullayev.ali has joined the channel
#pinot-helix
@chinmay.cerebro: @chinmay.cerebro has joined the channel
@hristo: @hristo has joined the channel
@hristo: hi @chinmay.cerebro
#order-by
@shish: @shish has joined the channel
#troubleshooting
@laabidi.raissi: @laabidi.raissi has joined the channel
@nesrullayev.ali: @nesrullayev.ali has joined the channel
@orbit: @orbit has left the channel
@orbit: @orbit has joined the channel
#pinot-s3
@shish: @shish has joined the channel
#onboarding
@orbit: @orbit has joined the channel
#pinot-dev
@shish: @shish has joined the channel
@atri.sharma: Any challenges seen in building Pinot master on M1? I run into dependency issues where GRPC and Node dependencies are not found for the platform
@orbit: @orbit has left the channel
@orbit: @orbit has joined the channel
#announcements
@shish: @shish has joined the channel
#thirdeye-pinot
@shish: @shish has joined the channel
#getting-started
@raluca.lazar: @raluca.lazar has joined the channel
@nesrullayev.ali: @nesrullayev.ali has joined the channel
@orbit: @orbit has left the channel
@orbit: @orbit has joined the channel
@bagi.priyank: Can someone please point me to documentation for `enableDefaultStarTree` and `enableDynamicStarTreeCreation` fields in the table confi? I want to understand what does a default / dynamic star-tree index mean.
@bagi.priyank: Also what happens when I update star-tree index configs in scenarios like - adding a new dimension to `dimensionsSplitOrder` or even removing one - what happens to the index and the segments? same for `functionColumnPairs`. I am thinking of editing as adding a new one and dropping the old one.
#flink-pinot-connector
@shish: @shish has joined the channel
#pinot-trino
@shish: @shish has joined the channel
#udf-type-matching
@xiangfu0: @xiangfu0 has joined the channel
@yupeng: @yupeng has joined the channel
@xiangfu0: set the channel topic:
@xiangfu0: set the channel description:
@yupeng: thanks
@jackie.jxt: @jackie.jxt has joined the channel
@christil: @christil has joined the channel
@amrish.k.lal: @amrish.k.lal has joined the channel
@yupeng: folks, created a channel to discuss
@yupeng: we see more and more users run into this issue
@yupeng: and it's a concerning one, as it could return incorrect result silently
@yupeng: any suggestions on having a stop-gap sooner?
@amrish.k.lal: I agree that this is concerning, although I am not quite sure if there is a good enough stop-gap fix that is available. Seems to me that any stop-gap fix would still suffer from some of the type issues that we see now. So, at the very least it would be good to know what our final solution would be to fix this issue before we start adding stop-gap fixes.
@yupeng: makes sense. we can discuss the ideal state, and then see if there's opportunity to have some easy solution to approximate
@g.kishore: @g.kishore has joined the channel
@chinmay.cerebro: @chinmay.cerebro has joined the channel
@yupeng: add @g.kishore @chinmay.cerebro too, since i brought this up a while ago to them
@yupeng: my thoughts is to have a typing hierarchy from high precision to low precision. so any high precision will short-circuit and map to its corresponding implementation
@jackie.jxt: Ideally we should pick the same way as how java performs the type matching (high to low precision, and implicit type conversion)
@yupeng: good point
@yupeng: we need to implement this implicit cast rule
@amrish.k.lal: we would also need to detect if implicit conversion is lossy. For example, while java allows implicit conversion of `long` to `double`, that conversion is lossy. ```public static void main(String[] args) { long l = 8223372036854775802l; double d = l; long f = (long) d; System.out.println("Original Long Value : " + l); System.out.println("Intermediate Double Value: " + d); System.out.println("Final Long Value : " + f); System.out.println( l == f ? "Not a lossy conversion" : "Lossy conversion"); }``` If we convert a `long` value to `double` and then convert the resulting `double` value back to `long`, then the final `long` value may not be same as initial `long` value due to lossy conversion of `long` to `double`. Currently the output type of all functions in `ArithmeticFunctions.java` is set to `double`.
@richard892: @richard892 has joined the channel
@yupeng: interesting
@yupeng: also add @richard892 to this channel, as there was a similar conv on a PR review
@amrish.k.lal: Adding @steotia as well, since he had a few thoughts regarding scalar functions using templates.
@steotia: @steotia has joined the channel
--------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@pinot.apache.org For additional commands, e-mail: dev-h...@pinot.apache.org
