You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
From what I can see, Julia randomly fails to define some methods in DataFrames correctly. Below are two runs of the current DataFrames master branch test suite in sequence: the first fails and the second succeeds even though nothing has changed in the underlying code in between. This has been happening for a long time now, but the randomness has made it hard for me to pin down.
What seems to happen is that line 402 of dataarray.jl sometimes get defined correctly and other times winds up getting defined as if it returned nothing. Perhaps there's some race condition in the method definitions? I honestly have no idea how to make sense of this. I used to think it had something to do with the data sets we generated using an RNG, but that seems less likely now.
jmwmacbookpro:DataFrames johnmyleswhite$ julia run_tests.jl
Running tests:
* test/dataarray.jl
* test/extras.jl
* test/pooleddataarray.jl
* test/data.jl
* test/index.jl
* test/dataframe.jl
* test/operators.jl
* test/io.jl
* test/constructors.jl
* test/abstractarray.jl
* test/booleans.jl
* test/indexing.jl
* test/sort.jl
* test/iteration.jl
* test/colfuncs.jl
* test/duplicates.jl
* test/padding.jl
* test/tabulation.jl
* test/datamatrix.jl
ERROR: no method *(DataArray{Float64,1},Nothing)
in include_from_node1 at loading.jl:91
in anonymous at no file:35
in include_from_node1 at loading.jl:91
in process_options at client.jl:274
in _start at client.jl:347
at /Users/johnmyleswhite/.julia/DataFrames/test/datamatrix.jl:45
at /Users/johnmyleswhite/.julia/DataFrames/run_tests.jl:36
jmwmacbookpro:DataFrames johnmyleswhite$ julia run_tests.jl
Running tests:
* test/dataarray.jl
* test/extras.jl
* test/pooleddataarray.jl
* test/data.jl
* test/index.jl
* test/dataframe.jl
* test/operators.jl
* test/io.jl
* test/constructors.jl
* test/abstractarray.jl
* test/booleans.jl
* test/indexing.jl
* test/sort.jl
* test/iteration.jl
* test/colfuncs.jl
* test/duplicates.jl
* test/padding.jl
* test/tabulation.jl
* test/datamatrix.jl
Sorry for raising such a strange issue.
The text was updated successfully, but these errors were encountered:
From what I can see, Julia randomly fails to define some methods in DataFrames correctly. Below are two runs of the current DataFrames master branch test suite in sequence: the first fails and the second succeeds even though nothing has changed in the underlying code in between. This has been happening for a long time now, but the randomness has made it hard for me to pin down.
What seems to happen is that line 402 of
dataarray.jl
sometimes get defined correctly and other times winds up getting defined as if it returnednothing
. Perhaps there's some race condition in the method definitions? I honestly have no idea how to make sense of this. I used to think it had something to do with the data sets we generated using an RNG, but that seems less likely now.Sorry for raising such a strange issue.
The text was updated successfully, but these errors were encountered: