[rkward-devel] Second call for testing: RKWard 0.5.3-pre2

Prasenjit Kapat kapatp at gmail.com
Wed Apr 21 09:37:40 UTC 2010


On Tue, Apr 20, 2010 at 4:52 PM, mat <matthieu.stigler at gmail.com> wrote:
>
> Thomas Friedrichsmeier a écrit :
>>
>> Hi,
>>
>> On Tuesday 20 April 2010, you wrote:
>>
>>>
>>>  From make plugintests:
>>>
>>> import_spss                   match          MISMATCH
>>> MISMATCH       ERROR          FAIL
>>> import_stata                  match          MISMATCH
>>> MISMATCH       ERROR          FAIL
>>> shapiro_wilk_test             match          MISMATCH       match
>>> (empty)  no             FAIL
>>> lillie_test                   match          MISMATCH       match
>>> (empty)  no             FAIL
>>>
>>
>> could you post/attach the difference listings for these tests?

I've attached mine as well... This is using the updated code (using sink).

>>>
>>> Otherwise, still need to install some packages, do you have a list of
>>> all packages that are needed to run plugintests? Or should I make one
>>> manually?
>>>
>>
>> I don't have a list, you'll have to make your own for now (but in the
>> future it might be a good idea to have the test-framework print a list of
>> the packages that were missing).
>>
>>
>
> so this would be the list:
> install.packages(c("R2HTML", "tseries", "nortest", "outliers",
> "exactRankTests", "moments", "car", "hdrcde", "qcc", "xtable", "eRm",
> "ltm"))
>
> maybe you could incude it in the compiling/testing website and/or in the
> plugintest makefile itself?

It indeed would be good to have it documented (at least statically)
somewhere: may be the INSTALL file under "Automated tests" section?

The idea is that these are not "required" for rkward to run. It is
required only when the corresponding plugins are used. Ideally, it
would be nice if the plugins remain disabled in the absence of
required libraries.

>>> Is there actually a shortcut to switch from script editor to console and
>>> vice versa (like emacs C-x o)? Or should one use show/hide console?

I remember a mail long time, someone suggesting to bring in the emacs
style shortcuts.. may be it was on the kile list! That would be really
nice ;)

>> Not quite, but there are some options that come close: You can switch back
>> and forth using Alt+0 and Alt+5. You can press Alt+5 repeatedly, which will
>> show/hide the console. Or you can detach the console window and use Alt+Tab
>> to switch between that and the main window. I do intend to add an
>> Alt+Tab-like feature to switch between the most recent rkward windows,
>> sooner or later, but it's not trivial to do.
>>
>>
>
> ok thanks. I indeed used Alt+0/Alt+5, but definitely would be nice once to
> have direct switch
>>>
>>> I think this is not new but copying directly from console with ctrl+v is
>>> not possible, right?
>>>
>>
>> It is possible to copy from the console, but Ctrl+C is used for "interrupt
>> command" (which might not be a good idea, after all; what does everybody
>> else think?). There are two different copy-actions in the Edit-menu, and by
>> now you can also customize the shortcuts using Settings->Configure
>> shortcuts.
>>
>>
>
> I tried to used this. But as soon as I use "ctrl", brings me back to latest
> console line... About which should be the default, as rather "GUI user" and
> so linux geek, I would argue ctrl+c as copy is more intuitive...

I think it is intuitive in an editor environment, not at a command
prompt. Whether it is the native R prompt, Matlab prompt, or any shell
prompt - ctrl+c is to interrupt.

So, I strongly suggest that Ctrl+C should remain as abort/interrupt
shortcut in R Console. To select multiple lines from R Console one
_has_ to use the mouse anyway, so might as well, right click and copy.
Everywhere else Ctrl+C works. It would be nice to have the right click
menu on help pages and tab bar, though.


-- 
Prasenjit
-------------- next part --------------
A non-text attachment was scrubbed...
Name: make_plugintests.txt.tar.gz
Type: application/x-gzip
Size: 4560 bytes
Desc: not available
URL: <http://mail.kde.org/pipermail/rkward-devel/attachments/20100421/7208f733/attachment.gz>


More information about the Rkward-devel mailing list