Nbconvert doesn't have the same functionality for reading XML notebook
files. It was using the standard libary ElementTree module to parse HTML
for a couple of filters, and this was switched to defusedxml to prevent
attacks where a (JSON) notebook containing maliciously crafted HTML was
sent to
Maybe deprecate it for a release cycle and see if anybody complains once
they see the warning?
(also I am, in general, always in favor of simplifying things in nbconvert
:-) )
On Fri, Feb 15, 2019 at 5:56 PM Matthew Seal wrote:
> Do you think we should remove the functionality from nbconvert
Do you think we should remove the functionality from nbconvert to simplify
things there too?
On Thu, Feb 14, 2019 at 1:28 PM Samuel Lelièvre
wrote:
> Related: the latest version of nbconvert added defusedxml as a dependency.
>
> --
> You received this message because you are subscribed to the
Related: the latest version of nbconvert added defusedxml as a dependency.
--
You received this message because you are subscribed to the Google Groups
"Project Jupyter" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to
Thanks all, I've merged the PR.
Agreed, this doesn't preclude converting notebooks to XML, but we probably
wouldn't use the removed code for that anyway. And we can always resurrect
this code if we realise that it is useful for something.
On Tue, 12 Feb 2019 at 08:16, Fernando Perez wrote:
>
+1 for removal - I think that if/when we tackle the problem of publishers,
a cleaner approach would probably be to output XML versions of notebook
content strictly tailored to their data/metadata schemas (and thus not
meant for *ingestion* by Jupyter, only as output).
On Mon, Feb 11, 2019 at
The only major potential stakeholder that (I think) still cares about XML
is large-scale publishers, but I don't think there's been any official
adoption of Jupyter Notebooks there anyway, so it's probably fine. Thanks
for bringing this up Thomas! I'm +1 as well!
On Mon, Feb 11, 2019 at 2:27 PM
+1
On Mon, Feb 11, 2019 at 1:41 PM Matthew Seal wrote:
> I'm 100% for removing -- I can't think of any recent tools that even
> support xml. Thanks for making the PR!
>
> On Mon, Feb 11, 2019 at 7:01 AM Thomas Kluyver wrote:
>
>> Hi all,
>>
>> Way back in 2011, when the first version of the
I'm 100% for removing -- I can't think of any recent tools that even
support xml. Thanks for making the PR!
On Mon, Feb 11, 2019 at 7:01 AM Thomas Kluyver wrote:
> Hi all,
>
> Way back in 2011, when the first version of the IPython Notebook was being
> written, there was an option to store
Hi all,
Way back in 2011, when the first version of the IPython Notebook was being
written, there was an option to store notebooks as XML. JSON was chosen as
the default, and has always been the format all of our applications use.
However, the code to read XML files stayed around, and recently
10 matches
Mail list logo
