Array columns error in Jupyter notebook and lab

On astropy version: 4.3.dev1429+ga832159 and 4.3.dev1484+g3f8cf27b8, in a jupyter lab environment, putting a particular table as the only item in a Jupyter Lab cell results in the following traceback and printed out table (MWE):

import numpy as np
from astropy.table import Table
Table([np.ones([10,10]), np.arange(10)])

results in both

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
~/anaconda3/lib/python3.7/site-packages/IPython/core/formatters.py in __call__(self, obj)
    343             method = get_real_method(obj, self.print_method)
    344             if method is not None:
--> 345                 return method()
    346             return None
    347         else:

~/repos/astropy/astropy/table/table.py in _repr_latex_(self)
   1549         latexdict = self.meta.get('latexdict', {})
   1550         lTable = ascii.latex.Latex(latexdict=latexdict)
-> 1551         lines = lTable.write(table=self)
   1552         self.meta['latexdict'] = lTable.latex
   1553 

~/repos/astropy/astropy/io/ascii/latex.py in write(self, table)
    338         self.header.start_line = None
    339         self.data.start_line = None
--> 340         return core.BaseReader.write(self, table=table)
    341 
    342 

~/repos/astropy/astropy/io/ascii/core.py in write(self, table)
   1474         # Check that table column dimensions are supported by this format class.
   1475         # Most formats support only 1-d columns, but some like ECSV support N-d.
-> 1476         self._check_multidim_table(table)
   1477 
   1478         # Now use altered columns

~/repos/astropy/astropy/io/ascii/core.py in _check_multidim_table(self, table)
   1282             If any column exceeds the number of allowed dimensions
   1283         """
-> 1284         _check_multidim_table(table, self.max_ndim)
   1285 
   1286     def read(self, table):

~/repos/astropy/astropy/io/ascii/core.py in _check_multidim_table(table, max_ndim)
     53     nd_names = [col.info.name for col in table.itercols() if len(col.shape) > max_ndim]
     54     if nd_names:
---> 55         raise ValueError(f'column(s) with dimension > {max_ndim} '
     56                          "cannot be be written with this format, try using 'ecsv' "
     57                          "(Enhanced CSV) format")

ValueError: column(s) with dimension > 1 cannot be be written with this format, try using 'ecsv' (Enhanced CSV) format

and

col0 [10] | col1
-- | --

1.0 .. 1.0 | 0
1.0 .. 1.0 | 1
1.0 .. 1.0 | 2
1.0 .. 1.0 | 3
1.0 .. 1.0 | 4
1.0 .. 1.0 | 5
1.0 .. 1.0 | 6
1.0 .. 1.0 | 7
1.0 .. 1.0 | 8
1.0 .. 1.0 | 9

Why is it using _repr_latex_ instead of _repr_html_, is that a config option in Jupyter ?

@mwcraig

What is your jupyter --version? When it comes to Lab, maybe @mwcraig or @eteq know something.

$ jupyter --version
jupyter core     : 4.7.0
jupyter-notebook : 6.1.6
qtconsole        : 4.7.7
ipython          : 7.19.0
ipykernel        : 5.1.4
jupyter client   : 6.1.7
jupyter lab      : 2.2.6
nbconvert        : 5.6.1
ipywidgets       : 7.6.1
nbformat         : 5.0.8
traitlets        : 4.3.3

and

$ jupyter --version
jupyter core     : 4.7.1
jupyter-notebook : 6.2.0
qtconsole        : 5.0.2
ipython          : 7.21.0
ipykernel        : 5.1.2
jupyter client   : 6.1.7
jupyter lab      : 1.2.6
nbconvert        : 6.0.7
ipywidgets       : 7.6.3
nbformat         : 5.1.2
traitlets        : 5.0.5

(I tried on both machines)

But it works in a "notebook" right?

No, the same issue occurs in a pure Notebook too (I hadn't tried that before, but I just tested and verified the result is the same)

Yeah, I tried too. Same results. Works on IPython session though. 🤷

@taldcroft

Oh indeed, I get the error too on master. Bisect points to b40b22b (cc @taldcroft)

The root problem is really using _repr_latex_ instead of _repr_html_. I noticed a problem in the VS code Jupyter extension where it is doing the same thing and displaying the latex for a table. I believe if you take out the check from b40b22b then you'll get a different problem that it shows a latex table not the desired HTML. This seems like an upstream problem where it isn't choosing the cell output type properly.

I was wondering if _repr_html_ was somehow failing for a multidimensional column now and Jupyter was falling through to _repr_latex_, but seems not since the following works OK:

from IPython.display import HTML
HTML(Table([np.ones([10,10]), np.arange(10)])._repr_html_())

So I just don't understand Jupyter notebook. I took out the new check that causes problems and Jupyter runs _repr_latex_ (now successfully) but just throws away that output and moves on to _repr_html_.

Anyway, fix coming.

Navigation Menu

Repository navigation

Array columns error in Jupyter notebook and lab #11695

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

Array columns error in Jupyter notebook and lab #11695

Description

Activity

saimn commented on May 4, 2021

pllim commented on May 4, 2021

keflavich commented on May 4, 2021

pllim commented on May 4, 2021

keflavich commented on May 4, 2021

pllim commented on May 4, 2021

saimn commented on May 4, 2021

taldcroft commented on May 5, 2021

taldcroft commented on May 5, 2021

taldcroft commented on May 5, 2021

Metadata

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

Issue actions