planner: fix bug when pruning columns for TableDual #11054
Conversation
|
/run-all-tests |
winoros
changed the title
|
/run-common-test |
Codecov Report
@@ Coverage Diff @@
## master #11054 +/- ##
===========================================
Coverage 81.2821% 81.2821%
===========================================
Files 423 423
Lines 90133 90133
===========================================
Hits 73262 73262
Misses 11571 11571
Partials 5300 5300 |
|
/run-all-tests tidb-test=pr/847 |
| @@ -262,8 +262,15 @@ func (p *LogicalTableDual) PruneColumns(parentUsedCols []*expression.Column) err | |||
| } | |||
| } | |||
| for k, cols := range p.schema.TblID2Handle { | |||
| if p.schema.ColumnIndex(cols[0]) == -1 { | |||
| for i := len(cols) - 1; i >= 0; i-- { | |||
There was a problem hiding this comment.
The content of TblID2Handle looks weird for the above test case, i.e, t1.a and t2.a are all stored in single bucket of TblID2Handle[70] in my instance for example. Though t1 and t2 are two aliases for same table t, actually they should be 2 separate tables in the query. Would it have risk in executor when using TblID2Handle like this?
There was a problem hiding this comment.
Is any situation that the handle has not only one column? I think TblID2Handle should be the map<id, *Column>.
There was a problem hiding this comment.
Anyway. This is one of the chaos of Schema. Maybe temporarily fix bugs is OK.
There was a problem hiding this comment.
@lzmhhh123 @eurekaka
The only case that one table id maps to multiple Column is multi-table delete.
lzmhhh123
added
status/LGT2
|
@lzmhhh123 Currently, no. We can move this thing out of schema ASAP. |
What problem does this PR solve?
Dual may not be from only one data source.
So check
cols[0]is not enough.What is changed and how it works?
Check the full slice.
Check List
Tests
Related changes