The old patch somewhat works.
http://trac.sagemath.org/sage_trac/attachment/ticket/11436/trac_11436.patch

It didn't apply automatically because the source in sage-5.0
looks a bit different now:
---------
    def _read_in_file_command(self, filename):
        r"""
        Return the R command (as a string) to read in a file named
        filename into the R interpreter.

        EXAMPLES::

            sage: r._read_in_file_command('file.txt')
            'file=file("file.txt",open="r")\nsource(file)'
        """
        return 'file=file("%s",open="r")\nsource(file)'%filename
---------

Changing to
>        return 
> 'file=file("%s",open="r")\nsource(file,print.eval=TRUE)'%filename
doesn't do the trick, however reverting back to the old way, does:
>        return 'source("%s",print.eval=TRUE)'%filename

The error message in the notebook prints full %filename, however.

- MS


On 21 Maj, 15:17, kcrisman <kcris...@gmail.com> wrote:
> On May 20, 5:38 pm, Maciek Sykulski <macie...@gmail.com> wrote:
>
>
>
>
>
>
>
>
>
> > Thanks Nils.
> > It turns out the problem concerns any text output from R, not only
> > error messages.
> > In each case R code is being evaluated (ex. values of variables
> > change),
> > however no sign of feedback from R if the input is too long.
>
> > sage: Ri.eval(('#1')**100+"\n print(1)")
> > '\n[1] 1'
> > sage: Ri.eval(('#1')**1000+"\n print(1)")
> > ''
>
> > For now, I'll stick with this change in local/lib/python/site-packages/
> > sage/interfaces/r.py
> > line 1045 from:
> > <        return Expect.eval(self, code, synchronize=synchronize,
> > *args, **kwds)
> > to:
>
> > >        return Expect.eval(self, code, synchronize=synchronize, 
> > > allow_use_file=False, *args, **kwds)
>
> Update:
>
> Apparently I solved this a year ago, and forgot.  Try the solution 
> athttp://trac.sagemath.org/sage_trac/ticket/11436and see if it is what
> you are looking for.

-- 
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to 
sage-devel+unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

Reply via email to