[yt-users] Flush_data_to_grids glitch

[Matthew Turk]

On Sun, Jul 6, 2014 at 6:43 PM, David Collins <dcollins4096 at gmail.com> wrote:
> Hi, everybody--
>
> In data_containers.py, around line 2664, there's a routine
> flush_data_to_grids.  It seems like the line that says
> "np = pointI[0].ravel().size"
>
> is an error, as np refers to the numpy package.  Does that seem right?  I
> expect that should be number_of_points, or something?

Yup!  Good catch.  Want to submit a quick pull request with this fix?

(Also, thanks for kindly not bringing up all the times my shorthand
variable name for "point index" has conflicted with a rather important
constant.)

-Matt

>
>     def _flush_data_to_grids(self, field, default_val, dtype='float32'):
>         """
>         A dangerous, thusly underscored, thing to do to a data object,
>         we can flush back any changes in a given field that have been made
>         with a default value for the rest of the grid.
>         """
>         i = 0
>         for grid in self._grids:
>             pointI = self._get_point_indices(grid)
>             np = pointI[0].ravel().size
>             if grid.has_key(field):
>                 new_field = grid[field]
>             else:
>                 new_field = np.ones(grid.ActiveDimensions, dtype=dtype) *
> default_val
>             new_field[pointI] = self[field][i:i+np]
>             grid[field] = new_field
>             i += np
>
>
> --
> -- Sent from a computer.
>
> _______________________________________________
> yt-users mailing list
> yt-users at lists.spacepope.org
> http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
>