OWTabletoTimeseries: Accept meta attributes

[ajdapretnar]

Issue

Fixes #51.

Description of changes

Accept meta attributes in As Timeseries widget.

Includes

• Code changes
• Tests
• Documentation

[kernc]

Only accepts TimeVariables from metas? The two lines above list also other continuous variables. Like this:

Suggested change

	[var for var in data.domain.metas if isinstance(var, TimeVariable)]
	[var for var in data.domain.metas if var.is_time] + \
	[var for var in data.domain.metas if var.is_continuous]

[ajdapretnar]

Yes, I wasn't sure whether to include numeric or not. Easy fix.

[ajdapretnar]

Can I ask why does the first two checks use isinstance? [var for var in data.domain.variables if isinstance(var, TimeVariable)]
Also, would it be nicer to:
[var for var in data.domain.metas if var.is_time or var.is_continuous]?
To avoid one extra loop?

[kernc]

The reason for the extra loop was maybe that continuous and time variables become grouped. I'm actually 👍 for that.

[kernc]

https://travis-ci.org/biolab/orange3-timeseries/builds/451775988#L708-L720

[ajdapretnar]

Dude, I was literally testing with this same data set this afternoon. Where did it go? 😮 And what is a good simple alternative dataset to test with?

[ajdapretnar]

@kernc Fixed.

[codecov-io]

Codecov Report

Merging #61 into master will not change coverage.
The diff coverage is n/a.

@@           Coverage Diff           @@
##           master      #61   +/-   ##
=======================================
  Coverage   67.06%   67.06%           
=======================================
  Files           7        7           
  Lines         671      671           
  Branches      102      102           
=======================================
  Hits          450      450           
  Misses        175      175           
  Partials       46       46

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update b987e1a...ec3c15e. Read the comment docs.

[kernc]

and not var.is_time? Can also var.is_time on the instance check above.

[ajdapretnar]

If TimeVariable is a ContinuousVariable, could we not simply

vars = [var for var in data.domain.variables if var.is_continuous] + \
                   [var for var in data.domain.metas if var.is_continuous]

[kernc]

Yes. This was to make variables appear grouped.

[ajdapretnar]

Ah, right. I'll stick to that. I think it is nice, having them grouped.

[kernc]

TimeVariable is a ContinuousVariable, so var.is_continuous check should be enough.