ValueError when calling assert_lines_of_type() with line_types='regression'

[marty-larocque]

If base.assert_lines_of_type() is called with line_types='regression' on a plot without a regression line, we would expect it to raise an AssertionError.

However, if we do this on a plot with only a one-to-one line, base.assert_lines_of_type() will raise a ValueError. This is because the method for determining if a regression line exist uses the xy-data for each line. One-to-one lines do not necessarily have specific xy-data.

We should rewrite this function to account for this case.

[lwasser]

can you please provide some example code that highlights this issue ?

[marty-larocque]

Here is a gist demonstrating this issue.
https://gist.github.com/ryla5068/0f404cc9a5fae4fb5981aa21dd473672

[lwasser]

@ryla5068 can you clarify what is going on with this issue?? if we can close it please let me know.

[marty-larocque]

@lwasser the description for this issue is a bit misleading. This was caused by a weird way were were checking the line. This is addressed in #239, and can be closed when that PR is merged.

[nkorinek]

This can be closed, fixed in #274 @lwasser

[lwasser]

closing!!