[rkward-users] Console printing behavior

Thomas Friedrichsmeier thomas.friedrichsmeier at ruhr-uni-bochum.de
Wed Nov 10 08:40:44 UTC 2010


Hi,

On Tuesday 09 November 2010, Gilles San Martin wrote:
> It could maybe be an easy temporary solution to add a button/command
> that do the whole stuf : copy from the script, paste in the console and
> add a "carriage return" to be sure that the last line of code is executed.

yes, this might be a short-term option. I'll try to remember to have another 
look, soon, and see, whether it can find a way to fix this, properly.
 
> When the execution of one line (command) is long, rkward print the
> following command(s) in green and without the prompt as if it was an R
> print/error message etc... instead of the normal syntaxic coloration
> with the prompt. But these "green" commands seems to be nevertheless
> executed. Some times each of these commands is executed 2 times (but I
> was not able to reproduce this artificially).

Yes, indeed. This will need some reworking.

About the commands being executed several times: It happens when submitting a 
construct like this in a line-by-line fashion:
	{
		print (1);
	}; print (2); {
		print (3);
	}
What throws us off is the third line: When submitting the command up to this 
point, the parser tells us the statement is still incomplete, but actually, 
some of it was a complete statement, and has already been executed. It's a bit 
embarassing, but fortunately this is not triggered all too often (and it's 
another reason why Shift + F8 and Shift + F9 run the command a bit 
differently).

I am in the middle of a large rework of the internals of the interface to R. 
This should also make it possible to fix this (but don't hold your breath).

Regards
Thomas
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: This is a digitally signed message part.
URL: <http://mail.kde.org/pipermail/rkward-users/attachments/20101110/a88da2a6/attachment.sig>


More information about the Rkward-users mailing list