Decouple parquet fuzz tests from converter (#1661) #2386
Conversation
| 20, | ||
| ConvertedType::NONE, | ||
| None, | ||
| &converter, | ||
| |vals| Arc::new(converter.convert(vals.to_vec()).unwrap()), |
There was a problem hiding this comment.
Until #2318 these types still use the converter plumbing=
| BinaryArray, BinaryBuilder, LargeStringArray, LargeStringBuilder, StringArray, | ||
| StringBuilder, | ||
| }; | ||
| use arrow::array::{StringArray, StringBuilder}; |
There was a problem hiding this comment.
Sadly we need to keep StringArrrayConverter for now as it used by the ComplexObjectArrayReader tests
tustvold
force-pushed
the
fuzz-test-converter
branch
from
1bce846
to
db944fb
Compare
| FixedSizeArrayConverter, | ||
| RandFixedLenGen, | ||
| >( | ||
| run_single_column_reader_tests::<FixedLenByteArrayType, _, RandFixedLenGen>( |
There was a problem hiding this comment.
It took me a while to grok this, so I figured I would note it for any other reviewers:
The reduction of generic arguments isn't a change in coverage (the same types are still covered) it is simply fewer generics required to express the same test (in this case on FixedLenByteArrayType 👍 )
|
Benchmark runs are scheduled for baseline = b6b1ffd and contender = 5127490. 5127490 is a master commit associated with this PR. Results will be available as each benchmark for each run completes. |
Which issue does this PR close?
Part of #1661
Rationale for this change
Part of moving away from ComplexObjectArrayReader is migrating the tests away from using its implementation details. This does this for the parquet fuzz tests
What changes are included in this PR?
Are there any user-facing changes?
No