Jupyter Weekly Summary (2016, Week 48)
Matthias Bussonnier
Hi all,
This is an attempt to capture the weekly meeting in a written summary.
I’ll try to do that each Tuesday after each meeting. Though I have to
cap the time I spend on writing this summary to 30 minutes. Apologies
if it is a bit rough.
I hope that going through written notes is easier than listening to 45
minute video on youtube.
If this is useful to you, please say so. If you are open to helping in
making these notes more readable help is extremely appreciated.
Classic notebook:
We are closing on a 4.3 release (long due). If you have some last
minutes requests or want to give a hand, now is the time.
JupyterLab
Long time requested feature in progress, restoring the sate of
Jupyterlab:
There’s a large, open PR on state management and application
restoration on page refreshes that should be finished within a day
or so. If you want to follow along, there’s an overview of the
restoration lifecycle in the PR description:
https://github.com/jupyterlab/jupyterlab/pull/1291. If you buy into
this system with your extension you should get state restoration for
free on page reload.JupyterLab 0.11 should be release in the next day or so. As usual
for any last minutes critical bugfixes it is now.JupyterLab was demoed at Gateways 2016, NYU, PlotCon and multiple
other venues. It was very well received, lots of interest for
potential collaborations. Special thanks to Matt Rocklin & Luke
Canavan for great Dask demo.
NBformat
- Discussion on draft spec for capturing a notebook environment is
reaching a consensus soon,
(https://github.com/jupyter/nbformat/pull/60) if you have anythong
to say speak now.
JupyterHub
- Jupyterhub 0.7 is due this week, one more documentation pass and it
should be ready. A couple of eyes on
https://github.com/jupyterhub/jupyterhub/pull/886 would be
appreciated (Expand traitlet documentation for spawner base class).
nbdime (NoteBook DIff and MErge)
- Should get a 0.1 release this week.
- Last pass on documentation.
Project is still on early stages and its the right time to get
involved.
QtConsole
- 4.3 likely going to be released this week.
IPywidgets
Ongoing work for a 6.0 release. Two betas released, including a
styling overhaul and fixing many small bugs. We’re concentrating
particularly on easing the transition from the 5.x releases.
Ongoing work on the embedding of widgets with formal json spec for
widget state + sphinx extension. (Widget state can be fully generated
from python backend now, which will enable pure-python sphinx
extensions)
There will be some small backwards-incompatible changes with 5.x in
JupyterLab For example, in the javascript, the way to specify widget
defaults is slightly changed, but the changes are fairly
straightforward. We also have ongoing discussion about some possible
changes to the Layout widget’s display functionality (see
ipython/ipywidgets#919, for example).
The changes should (IIUC) mostly affect you if you are writing widgets
for JupyterLab.
GitHub automation
We are working on GitHub automation: A bot that Backport PRs, Greet
users, migrate issue. Code not public yet but if you are interested to
participate let us know.
New Book on the website
We added the Mastering IPython 4.0 book on IPython.org :
http://ipython.org/books.html
We know have a policy on adding new books:
http://ipython.org/books_policy.html#books-policy
Feedback and link to new books welcomed.
Pycon Talk Proposal
We have a Pycon Tutorial proposal due today.
https://github.com/ipython/ipython-in-depth/pull/38 It’s a little late
to request feedback, we’ll try to do better next time.
Writing the Talk proposal about migrating to Python 3 only.
https://github.com/python3statement/pycon-2017/blob/master/proposal.md
Collaboration welcomed.
Considering making a Jupyter Talk Proposal, if anyone is interested in
helping writing one, (and co-presenting ?) if you plan to attend. Help
welcomed.
Nbconvert
- progress on header id filtering (the headers ids have someissues)
- trying to figure out what exactly pandoc is doing, haskell is…fun
- need to use regex (not re) in python to get access to unicode
character property classes do any one have good or bad experience
with it ? Do you have anything better to suggest ? - trying to find good comparable javascript library, XRegExp seems
like a good candidate, known issues?
Quantitative analysis of notebooks: proof of concept (Sam Penrose)
Sam Penrose and Connor Ameres are starting to be regularly involved,
they are working on the following:
The wrote a notebook feature exrtactor
https://github.com/cameres/notebook-feature-extractor it extract
cell-wise and notebook-wise feature extraction from IPython notebooks
to a Dataframe.
This make them able to cluster Mozilla’s notebooks into is / not used
for distributed data processing.
The goal is to understand notebooks as distributed data processing
IDE.
To help then they need :
- Point them at notebook collections (working / messy, not
presentation / tutorial) - Download and run their extractor your own notebooks:
- Add feature extractors: typically 1-5 line Python functions
- send PRs !
How can we make this a useful resource for the community?
Ideas for a shared place to collect analyses? gist.github.com or ?
That’s it I’m out of time to clean these notes, only 2 minutes left. I of course got some things wrong, please correct me.
See you next week.
Cheers,
Carol Willing
--
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 jupyter+u...@googlegroups.com.
To post to this group, send email to jup...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/jupyter/a037057a-a760-40ca-aae7-e08cf36d64d5%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
Big Stone
Damián Avila
--
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 jupyter+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/jupyter/b6f65183-4924-46a9-a667-9a5739fbbc02%40googlegroups.com.
Jason Grout
Hi Mathias,Thanks for the summary, reading through hackpads is not a pleasant task.Request:- would it be possible NOT to propose the "terminal" in Jupyterlab when it's not possibly working (aka not WSL Windows behind) ?