Possible bug in test_loads.py

[mfeurer]

Hi, I think I stumbled upon a bug in test_loads.py in the method test_format_correct. This test should not pass in my opinion because the only line of data is the same line which contains @data.

This also raises the question if a line which denotes @data should be checked with a u_row.startswith(). I propose to use a u_row.startswith() followed by a check if there is something else in the same line.

[mfeurer]

@renatopp can you please have a look at this?

[renatopp]

Hey, @mfeurer, despite the ARFF documentation, I always test all our ARFF examples within Weka. Thus, if Weka opens a given file, I think we should open too.

For instance, this example ARFF can be opened with Weka (3.7+) so I guess there is no problem with this format.

For reference, the example we are discussing here is:

@relation "software metric"
@attribute number_of_files numeric
@attribute "lines of code" numeric
@attribute 'defect density' numeric
@data 10,10,10

[mfeurer]

I posted an issue over at the wekalist. Let's see what happens.

[renatopp]

Hey @mfeurer, any news on this issue?

[mfeurer]

They confirmed it's a bug. I wrote again (2 hours ago) to ask whether they are going to change the behaviour. Let's see what happens.

[mfeurer]

It'll be fixed in the next release.

[mfeurer]

I just check this with the latest release and WEKA still parses this one :(