Details
-
Bug
-
Status: Resolved
-
Major
-
Resolution: Fixed
-
None
Description
It seems this is due to precompiled code being contaminated with undesired headers
For example, #include <iostream> in arrow/compare.h causes:
[ RUN ] TestDecimal.TestCastFunctions ../../src/gandiva/tests/decimal_test.cc:478: Failure Value of: (array_dec)->Equals(outputs[2], arrow::EqualOptions().nans_equal(true)) Actual: false Expected: true expected array: [ 1.23, 1.58, -1.23, -1.58 ] actual array: [ 0.00, 0.00, 0.00, 0.00 ] ../../src/gandiva/tests/decimal_test.cc:481: Failure Value of: (array_dec)->Equals(outputs[2], arrow::EqualOptions().nans_equal(true)) Actual: false Expected: true expected array: [ 1.23, 1.58, -1.23, -1.58 ] actual array: [ 0.00, 0.00, 0.00, 0.00 ] ../../src/gandiva/tests/decimal_test.cc:484: Failure Value of: (array_dec)->Equals(outputs[3], arrow::EqualOptions().nans_equal(true)) Actual: false Expected: true expected array: [ 1.23, 1.58, -1.23, -1.58 ] actual array: [ 0.00, 0.00, 0.00, 0.00 ] ../../src/gandiva/tests/decimal_test.cc:497: Failure Value of: (array_float64)->Equals(outputs[6], arrow::EqualOptions().nans_equal(true)) Actual: false Expected: true expected array: [ 1.23, 1.58, -1.23, -1.58 ] actual array: [ inf, inf, -inf, -inf ] [ FAILED ] TestDecimal.TestCastFunctions (134 ms)
Attachments
Issue Links
- links to