Request for new PyPI trove classifier for Jupyter Framework #210
Comments
|
Original comment by Richard Jones (Bitbucket: r1chardj0n3s, GitHub: r1chardj0n3s): I don't tend to add classifiers before there is a clear demand for them. Currently searching for "Jupyter" returns very few results, and those appear to be core components, not extensions (which appear to be the target of this exercise). |
|
Original comment by blink1073 (Bitbucket: blink1073, GitHub: blink1073): Richard, that is because they have not officially changed the name of the main project yet. However, Doug, myself, and others are already building tools that target Jupyter. IPython is a major project, and we wanted the classifier to be in place before the switch, to help with the transition. |
|
Original comment by Douglas Blank (Bitbucket: dblank, GitHub: dblank): Richard, just adding more details to what Steven said: people have been working on IPython 3.0.0 (aka Jupyter) tools for many months, but they are unable to identify those as Jupyter because of course there is no Jupyter classifier. We are working on building tools (for example, listing the available kernels for Jupyter). However, without a Jupyter classifier, there is no good way to do that. All of the following should be listed as Jupyter (rather than IPython): julia-kernel, octave-kernel, calico-scheme, chatbot-kernel, calysto-scheme, r-kernel, bash-kernel, echo-kernel, eval-kernel, hy-kernel, metakernel, calysto, processing-kernel, prolog-kernel, and probably many others. And more are on their way! Jupyter is planned to be released probably in the next month or so. It would be very helpful if we could get all of the above packages organized before release. |
|
Original comment by Matthias Bussonnier (Bitbucket: Carreau, GitHub: Carreau): Technically here are a few project that should move classifier from IPython to Jupyter, bash_kernel
There are probably a few others, but if you prefer to wait for us to be closer to release It's fine. |
|
Original comment by Brian Granger (Bitbucket: bgranger, GitHub: bgranger): Hi, one of the core IPython devs and founders of Project Jupyter here... IPython is already one of the largest python projects. We are in the middle of renaming most of the project to Jupyter. Thus, sometime in 2015, Jupyter will be one of the biggest Python projects around. Part of the renaming work involves splitting our code into multiple subprojects, creating repos for those, Python packages, etc. Having the Jupyter classifiers would be a huge help for us and the larger python community during and after this transition. Our own project alone will have 10-20 projects on pypi that would use these classifiers. Thanks! |
|
Original comment by Scott Sanderson (Bitbucket: scott_sanderson, GitHub: Unknown): I was looking for a Jupyter classifier today when releasing a new version of https://github.com/quantopian/pgcontents and was surprised to find that it wasn't available. |
|
Hi there ! We now have more than 60 packages that use the jupyter prefix, it would be great to have that in the trove classifier ! Thanks ! |
|
Pitching in from Jupyter here: our ecosystem is growing quite a bit, and it would be tremendously helpful for our users and the entire project to be able to easily find, list, track, etc., tools designed to work with Jupyter-related tools and ideas. If there's anything we can do from our side to help, please let us know, we'd be happy to. |
|
I think the main pain point for the PyPI maintainers is that there's no UI to add new classifiers so the only way to add or update them is to write raw SQL queries (they can correct me if I'm wrong) My plan was to create a UI to this make process much easier for Warehouse, but I don't really have much free time these days :( |
|
@fperez
[edit] # changes needed in warehouse to make it easier to fulfill trove requests
# ?/views.py
def add_classifier(request):
classifier = Classifier.from_string("For :: Example :: Jupyter :: 1.2"))
session.add(classifier)
# session.flush() / session.commit()
# classifiers/models.py
class Classifier( ):
def from_string(self, classifier):
""" TODO: see https://github.com/pypa/pypi-legacy/blob/master/trove.py """
attrs = {}
parts = [p.strip() for p in classifier.split("::")]
for i, part in enumerate(parts[0:], 0): # 1:] ?
attrs['l' + i] = part
return Classifier(**attrs) |
|
... are these like URIs? [edit] {"name": "This :: Is :: The :: Name",
"url": "#this/is/the/name" } |
AS requested in pypi/legacy#210
|
added |
Thanks you ! |
|
Thanks!
…On Thursday, April 6, 2017, Matthias Bussonnier ***@***.***> wrote:
added Framework :: Jupyter
Thanks you !
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#210 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AADGy-E-wN54au_MaVAg7-UYUILPUJmOks5rtW69gaJpZM4MJarT>
.
|
|
Don't want to be noisy here, but this is worth more than just an emoji 👍 ... We're very grateful this was done, as it will really help us organize our ecosystem and track its growth. Kudos to the team for making this happen!! |
Originally reported by: Douglas Blank (Bitbucket: dblank, GitHub: dblank)
IPython has developed a new project called Jupyter. This takes the Python-specific parts and makes a framework that is language agnostic, but still written in Python. IPython would make the Jupyter system available under this, and we would like to be able to classify components for Jupyter which are written in Python under this classifier.
Thanks!
References:
https://github.com/jupyter/
The text was updated successfully, but these errors were encountered: