handles integer vector stored as matrix in i, closes #826

[jangorecki]

Simple and I believe completed solution to close #826

[arunsrinivasan]

Could you please also add an item to NEWS, and add about this feature in the i argument of ?data.table? Thanks.

[jangorecki]

The ?data.table is already covered by sentence:

Integer and logical vectors work the same way they do in [.data.frame.

So no update required I believe. It is a special case of integer vector with second dimension artificially created. Description of it can be more confusing. I will update NEWS and put that to into bugfixes as it is bugfix comparing to data.frame behaviour.

[arunsrinivasan]

I'm not following. Where does it mention 1-col matrix? Why do you say it's artificially created? It's more confusing to see other matrices error, but not 1-col matrices.

[jangorecki]

Readme, Rd and code prepared. Some sequence number (62:63 in bugfix and 1549 in tests) skipped to avoid conflicts with other PR.

[jangorecki]

Looks like PR is right in time, see http://stackoverflow.com/questions/32509803/creating-a-data-partition-using-caret-and-data-table

[arunsrinivasan]

@jangorecki will take a look asap. You might want to resolve the conflicts in the meanwhile?

[jangorecki]

@arunsrinivasan will try today/tomorrow

[arunsrinivasan]

sure.

[jangorecki]

Finally learned some rebase :) once travis pass I think it is ready to merge.