It looks like a bug in codetools. It handles a number of functions
specially; data() is one, binomial() is another. I think there are some
implicit assumptions in it that stats and utils are on the search path,
but in the new tests, they aren't.
Duncan Murdoch
On 14/07/2015 3:40 AM, peter dalga
Dear Duncan,
Thank you for tracking down the source of this error.
Best,
John
On Tue, 14 Jul 2015 07:22:21 -0400
Duncan Murdoch wrote:
> On 13/07/2015 9:42 PM, John Fox wrote:
> > Dear Duncan,
> >
> > On Mon, 13 Jul 2015 20:00:02 -0400
> > Duncan Murdoch wrote:
> >> On 13/07/2015 7:35 PM,
On 13/07/2015 9:42 PM, John Fox wrote:
> Dear Duncan,
>
> On Mon, 13 Jul 2015 20:00:02 -0400
> Duncan Murdoch wrote:
>> On 13/07/2015 7:35 PM, John Fox wrote:
>>> Dear Duncan,
>>>
-Original Message-
From: Duncan Murdoch [mailto:murdoch.dun...@gmail.com]
Sent: July-13-15 7:
The danger of minimal examples is that you risk working around only one aspect.
The ones Alexandra had (from the sensR package) didn't (all?) involve data(). A
couple of them were like the one below, in which the FORTRANesque style of
naming the function the same as the variable with the eventu
