Repository navigation
Commit 96c7f07
committed
test: ignore the nested BINARY invalid UTF-8 tests on DataFusion 56
DataFusion 56 coerces nested BINARY file fields to Utf8 before the parquet
reader, which skips UTF-8 validation for unannotated BINARY columns, so the
native scan panics in debug builds instead of decoding to U+FFFD. Ignore the
Rust and JVM tests from #7106 until apache/datafusion#27065 is fixed (#7469).1 parent 56a88c9 commit 96c7f07
2 files changed
Lines changed: 6 additions & 1 deletion
File tree
- native/core/src/parquet
- spark/src/test/scala/org/apache/comet/parquet
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
841 | 841 | | |
842 | 842 | | |
843 | 843 | | |
| 844 | + | |
844 | 845 | | |
845 | 846 | | |
846 | 847 | | |
| |||
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
147 | 147 | | |
148 | 148 | | |
149 | 149 | | |
150 | | - | |
| 150 | + | |
| 151 | + | |
| 152 | + | |
| 153 | + | |
| 154 | + | |
151 | 155 | | |
152 | 156 | | |
153 | 157 | | |
| |||
0 commit comments