KiteSQL
KiteSQL copied to clipboard
KipData
SQL as a Function for Rust
## Feature Request - max-min-eliminate - ref: https://docs.pingcap.com/zh/tidb/stable/max-min-eliminate - correlated-subquery-optimization - ref: https://docs.pingcap.com/zh/tidb/stable/subquery-optimization - predicate-pushdown - outer -> inner, ref: [link](https://cn.pingcap.com/blog/tidb-source-code-reading-7/#:~:text=%E9%A6%96%E5%85%88%E4%BC%9A%E5%81%9A%E4%B8%80%E4%B8%AA%E7%AE%80%E5%8C%96%EF%BC%8C%E5%B0%86%E5%B7%A6%E5%A4%96%E8%BF%9E%E6%8E%A5%E5%92%8C%E5%8F%B3%E5%A4%96%E8%BF%9E%E6%8E%A5%E8%BD%AC%E5%8C%96%E4%B8%BA%E5%86%85%E8%BF%9E%E6%8E%A5%E3%80%82) - outer_join_eliminate - ref: https://cn.pingcap.com/blog/tidb-source-code-reading-21/ - aggregation_push_down - ref:...
## Feature Request If you want to contribute code to KiteSQL and find requirements, you can refer to the following list The corresponding test files are attached under the requirements...
It's kind of like a hierarchical lock and intention lock, but it is still a table level lock in most case, and multiple ddl on different table can execute concurrently....
## Feature Request `WITH TIME ZONE` on `CREATE TABLE TABLE_E141_07_10_01 ( A TIME WITH TIME ZONE DEFAULT CURRENT_TIMESTAMP )` I think this should create a new `DataValue` like `DataType::TimeWithZone` to...
## Bug Report Columns from different tables should be divided as much as possible on both sides so that they can be pushed down normally under the Join. 
The commit introduces a new rule for the optimization of LIKE operator in SQL queries. The LIKE operator expressions are rewritten to make use of binary operators such as GtEq...
## Feature Request Function - [ ] CURRENT_TIMESTAMP Create Table - [x] `PRIMARY KEY (column_1, column_2, ...)` - [x] `UNIQUE (column_1, column_2, ...)` - Options - [x] Default e.g. `CREATE...
## Feature Request When performing `Join Reorder` in CBO, `SortMergeJoin` may perform better than `HashJoin` under the influence of `interesting order`.
## Feature Request reason as https://github.com/KipData/FnckSQL/issues/219
## Feature Request When there is too much data to exist, the sorted data is temporarily stored in the hard disk.