Alex Chi Z.
Alex Chi Z.
> To achieve this goal, I will refactor some codes when creating a table, add two filed in BoundCreateTable like 'TableCatalog,Inner', get the two filed data from stmt after parsing,...
> ok, if so, I will only keep one vector in tableCatalog, and forbiden that syntax from binder Sounds good to me.
IIRC, all three should have been implemented in https://github.com/risinglightdb/risinglight/tree/main/src/optimizer/rules (filter join && filter scan). Let's change the issue title to "add tests for predicate pushdown"?
Except the last one, I'm not sure if it has been implemented yet...
``` > explain select a, b from t, s where a > 10 and b < 20; +-----------------------------------------------------------------------------------------------------------------------------------+ | PhysicalProjection: exprs [InputRef #0, InputRef #1] | | PhysicalNestedLoopJoin: op Inner,...
Should be the same as `count(*)` in the current framework. Only in edge cases will it be different from `count(*)` (after we have sub-query)
From the discussion in Slack... https://risingwave-community.slack.com/archives/C02UZDEE4AC/p1645885910667749 ``` array arrow mul/4096 time: [4.2330 ms 4.2809 ms 4.3336 ms] array arrow mul/65536 time: [4.2511 ms 4.3411 ms 4.4390 ms] array arrow mul/1048576...
https://github.com/risinglightdb/risinglight/blob/main/benches/array.rs This is our micro benchmark code. The arrow ones are in arrow2 repo.
cc @Shmiwy will take over the issue for a month. He will work on various parts of the system to gain full sort key support.
Storage now supports range filter scan. Need add support to other layers.