Hi, On Saturday 24 April 2010, mat wrote: > /home/matvir/Téléchargements/rkward-0.5.3-pre3/tests/import_export_plugins > _testfile.sav: position 1: Variable name begins with invalid character > > I tried to convert (latin nd ut8), is it still another encoding?
it should be the ISO-8859-15, IIRC. However the plugin does the encoding
conversion *after* the import, so selecting this will not chnage anything, I'm
afraid. Perhaps we could try setting a specific LC_CTYPE *before* doing the
import (using Sys.setlocale()). Anyway, while it's a good thing that the test
turned up this issue, I think we can live with it for the time being (we need
some good way to document known test failures, probably another page in the
wiki; any volunteer to make a start?).
> -I have a new problem: in the output, the run again is no more a link...
> the code appears instead:
>
> Rerun code:
>
> rk.call.plugin ("rkward::dixon_test", descriptives.state="0",
> length.state="1", opposite.state="FALSE", two_sided.state="TRUE",
> type.string="0", x.available="a", submit.mode="submit")
This means you're working with the workspace from the automated tests. If this
is in the RKWard-session from make plugintests: Simply quit and start a new
session. If it's a different session, you probably saved the workspace, and now
it was restored, automatically (which is another feature that needs fixing some
day). In this case, you should move the file .RData out of the way.
> -Finally, running the make plugintest, one is asked to enter once "enter
> for next plot", is this normal?
Yes it's normal, although we should really fix that somehow.
Regards
Thomas
signature.asc
Description: This is a digitally signed message part.
------------------------------------------------------------------------------
_______________________________________________ RKWard-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/rkward-devel
