nbconvert: Transformer tests #3914
Conversation
| @@ -79,7 +80,7 @@ def transform_cell(self, cell, resources, cell_index): | |||
| else: | |||
| data = data.encode("UTF-8") | |||
|
|
|||
| #Build a figure name | |||
| #Build a output name | |||
|
@minrk , second pass? |
|
This contains a couple fixes for the existing transformers, should I open a separate PR targeting 1.0 to merge those fixes? |
|
bump 🚗 |
|
Looks good to me. we have a few month to refine. |
| """Build an empty resources dictionary.""" | ||
|
|
||
| res = ResourcesDict() | ||
| res['metadata'] = ResourcesDict() |
There was a problem hiding this comment.
Does metadata need to be a ResourceDict ? Or should we also test the fact that when it not a resource dict thing works ? (you probably know nbconvert more than me now)
There was a problem hiding this comment.
Yes, just so we can do conversions if something doesn't exist (like notebook name). Otherwise Jinja will crash when parsing if exporting without a notebook name.
|
After carefull re-reading I have some concern of the use of ResourceDict as then test might actually be testing that attributes are (read or written) as we want to test that they are (written) |
| def build_notebook(self): | ||
| """Build a notebook in memory for use with transformer tests""" | ||
|
|
||
| outputs = [nbformat.new_output(output_type="stream", stream="stdout", output_text="a"), |
There was a problem hiding this comment.
You could also use IPython.nbformat.v3.tests.nbexamples for an example notebook
nbconvert: Transformer tests
nbconvert: Backport fixes cherry-picked from ipython#3914 and ipython#3923, which include tests that will not be merged into 1.0.
nbconvert: Transformer tests
[not for 1.0]
Adds tests for transformers.