[BUG] NumericRangeIterator_OnReopen is not implemented
#3171
Conversation
oshadmi
changed the title
NumericRangeIterator_OnReopen is nmot implementedNumericRangeIterator_OnReopen is not implemented
| if (ir->gcMarker == ir->idx->gcMarker) { | ||
| // no GC - we just go to the same offset we were at | ||
| size_t offset = ir->br.pos; | ||
| ir->br = NewBufferReader(&ir->idx->blocks[ir->currentBlock].buf); |
There was a problem hiding this comment.
I believe the ir->idx itself might be already freed here
Codecov ReportBase: 82.55% // Head: 82.49% // Decreases project coverage by
Additional details and impacted files@@ Coverage Diff @@
## master #3171 +/- ##
==========================================
- Coverage 82.55% 82.49% -0.06%
==========================================
Files 181 181
Lines 30119 30142 +23
==========================================
+ Hits 24864 24866 +2
- Misses 5255 5276 +21
Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here. ☔ View full report at Codecov. |
| env.expect('HSET', 'doc1', 'test', '1').equal(1) | ||
| env.expect('HSET', 'doc2', 'test', '2').equal(1) | ||
| conn.execute_command('HSET', 'doc1', 'test', '1') | ||
| conn.execute_command('HSET', 'doc2', 'test', '2') |
There was a problem hiding this comment.
Why changing that?
There was a problem hiding this comment.
You are skipping on cluster so not sure why you change it...
There was a problem hiding this comment.
@MeirShpilraien
It is a new test and I often use the conn. Syntax to avoid confusion.
Would you like me to switch?
There was a problem hiding this comment.
No, just wonder why you changed it.
(cherry picked from commit b3075a5)
This causes a crash if a numeric tree has a split of a leaf node which invalidates the inverted index held by the
IndexReader.MOD-4255
MOD-4296