Hi,
On Tuesday 14 September 2010, Prasenjit Kapat wrote:
> On my end, R 2.11.1 and lattice 0.19-11 (r-cran-lattice form deb
> repo). Before upgrading lattice, could you try to remove rkward and
> then re-install
>
> Rscript -e "remove.packages ('rkward')"
> make install
>
> and see if it helps?
On Tue, Sep 14, 2010 at 4:22 AM, Thomas Friedrichsmeier
wrote:
> On Tuesday 14 September 2010, Prasenjit Kapat wrote:
>> I tried it twice now. Didn't see any mismatch.. But I'll keep an eye on it.
>
> Interesting. I keep getting this:
>
> mark 1
> Current plot does not match with plots[[2]]
> mark
On Tuesday 14 September 2010, Prasenjit Kapat wrote:
> I tried it twice now. Didn't see any mismatch.. But I'll keep an eye on it.
Interesting. I keep getting this:
mark 1
Current plot does not match with plots[[2]]
mark 2
mark 3
mark 4
mark 5
mark 6
mark 7
mark 8
mark 8a
mark 9
mark 10
R 2.11.1
HI,
On Mon, Sep 13, 2010 at 7:45 AM, Thomas Friedrichsmeier
wrote:
> Hi,
>
> On Sunday 12 September 2010, Prasenjit Kapat wrote:
>> Yeah, the lattice plots are tracked through trellis.last.object () and
>> not recordPlot () and for that reason rk.activate.device () should be
>> used instead of de
Hi,
On Sunday 12 September 2010, Prasenjit Kapat wrote:
> Yeah, the lattice plots are tracked through trellis.last.object () and
> not recordPlot () and for that reason rk.activate.device () should be
> used instead of dev.set (). I'll update the test code later, after
> adding a few more tests.
Hi,
On Monday 13 September 2010, Prasenjit Kapat wrote:
> so I had to use rk.get.tempfile.name ("image", "jpg"). But the
> function IS defined with defaults!
in the test-framework, rk.get.tempfile.name() is redefined to always return a
fixed result (for static test outcomes). In redefining it, I
Hi,
On Sun, Sep 12, 2010 at 1:36 PM, Thomas Friedrichsmeier
wrote:
>
> Another thing: I just added a test to rkward_application_tests.R. Basically
> this opens two devices and calls some of the history actions. Then it checks
> whether the current plots are the expected ones. Somehow, when I try
Hi,
On Sun, Sep 12, 2010 at 3:41 PM, Prasenjit Kapat wrote:
> On Sun, Sep 12, 2010 at 2:41 PM, Prasenjit Kapat wrote:
>>
>>
>> > Another thing: I just added a test to rkward_application_tests.R. Basically
>>
>> Thanks.
>>
>> > this opens two devices and calls some of the history actions. Then it
On Sun, Sep 12, 2010 at 2:41 PM, Prasenjit Kapat wrote:
>
>
> > Another thing: I just added a test to rkward_application_tests.R. Basically
>
> Thanks.
>
> > this opens two devices and calls some of the history actions. Then it checks
> > whether the current plots are the expected ones. Somehow, w
Hi,
On Sun, Sep 12, 2010 at 1:36 PM, Thomas Friedrichsmeier
wrote:
> Hi,
>
> On Sunday 12 September 2010, Prasenjit Kapat wrote:
>> The "Append this plot" is supposed to be a safe guard action: append
>> _any_ displayed plot no matter what its state / type is, using
>> recordPlot (). For example,
Hi,
On Sunday 12 September 2010, Prasenjit Kapat wrote:
> It should not get lost! Do you see 'NA'? It should at least use the
> old behavior of 'X: xlab, Y: ylab, main'. Now the code may not always
> be able to find xlab / ylab / main from the recorded plot. For
> example, the structure of a recor
Hi,
On Sun, Sep 12, 2010 at 9:08 AM, Thomas Friedrichsmeier
wrote:
> Hi,
>
> BTW, I encountered some surprising behavior:
> When you add a plot using "Append this plot", the plot-call appears to be
> lost. Appending in any other way makes the plot-call is shown as expected. Is
> this intentional
Hi,
On Saturday 11 September 2010, Prasenjit Kapat wrote:
> None of the static messages will show such an error. But say if you call:
> xyplot (0~0,
> panel = function (...) {
> panel.xyplot (...)
> panel.abline (h=0,v=0)
> })
>
> Then, the showInfo box will be a few line
HI,
On Sat, Sep 11, 2010 at 1:03 PM, Thomas Friedrichsmeier
wrote:
> Hi,
>
> On Friday 10 September 2010, Prasenjit Kapat wrote:
>
>> 2. Are the message boxes of fixed height but variable width? Some of
>> the multiline text gets eaten up after 5/6 lines. But I guess, one
>> should not narrate st
Hi,
I've just started playing with the new history, so no feedback on that, yet.
But some answers to these questions:
On Friday 10 September 2010, Prasenjit Kapat wrote:
> @ Thomas:
> 1. There was some hickup while committing (I hadn't updated in a
> while). I hope none of your commits are screw
15 matches
Mail list logo
