Calling point on an iterator of 3d points raises an AttributeError

[seblabbe]

Drawing a list of 2d and 3d points works :

sage: point([(i,i) for i in range(10)])
sage: point([(i,i,i) for i in range(10)])

Drawing an iterator of 2d points works :

sage: point((i,i) for i in range(10))

Drawing an iterator of 3d points is broken :

sage: point((i,i,i) for i in range(10))
Traceback (most recent call last):
...
AttributeError: 'int' object has no attribute '_set_extra_kwds'

The same bug is obtained for empty iterator :

sage: point([])
sage: point(iter([]))
Traceback (most recent call last):
...
AttributeError: 'int' object has no attribute '_set_extra_kwds'

Moreover, I wonder if the line 1022 of sage/plot/plot3d/shapes2.pyc in
point3d(v, size, **kwds)

A = sum([Point(z, size, **kwds) for z in v])

should be

A = sum([Point(z, size, **kwds) for z in v], Graphics())

or something?

CC: @kcrisman

Component: graphics

Keywords: sd75

Author: Sébastien Labbé

Issue created by migration from https://trac.sagemath.org/ticket/10948

[kcrisman]

comment:1

Attachment: trac_10948_point3d-sl.patch.gz

See also #11426.

[seblabbe]

comment:2

I think I left this ticket as "new" because I was not sure about my fix. My question was : Is it necessary to expand the iterator as a list to fix it?

The problem was that in the function point, trying point2d on the iterator consumes it. Maybe point2d could do a copy of the iterator with itertools.tee before consuming it? I think such a fix would be better. What do you think?

Sébastien

[kcrisman]

comment:3

Hmm, that is very subtle!

sage: A = ((i,i) for i in range(10))
sage: type(A)
<type 'generator'>
sage: point(A)
<plots nicely>
sage: A.next()
--------------------------------------------------------------------
StopIteration: 

Yeah, that's not a bad idea. Usually people wouldn't do this, but I can certainly see it happening.

On the other hand, if it slows the process of getting this fix in, we should probably open a different ticket for that.

So it's up to you. If you feel motivated to add this right now, then do it and the whole thing can be reviewed. Otherwise, set this to 'needs review' and open another ticket for that.

[kcrisman]

Author: Sébastien Labbé

[seblabbe]

comment:8

I suggest to close this ticket as a duplicate of #14490.

[seblabbe]

Changed keywords from none to sd75

[seblabbe]

comment:9

I am cleaning my old trac tickets during Sage Days 75.

[embray]

comment:12

Determined to be invalid/duplicate/wontfix (closing as "wontfix" as a catch-all resolution).