Phil Elson
@pypelson
Proud parent of conda-forge, cartopy, Iris and 2 real-world kids. Former Au Pair of matplotlib. Working on accelerator controls at CERN. @pelson.bsky.social
Vous pourriez aimer
Retweets welcome on this... Ideally would like to rule out low Twitter engagement when determining the general utility of the project in this discussion (I also know I have a stronger conda audience 😉).
I posted about a PyPI-like interface for any PEP-503 compliant package repository at discuss.python.org/t/a-pypi-like-…. I'm looking for feedback before deciding whether to commit more effort in generalisation and open-development - your input is very welcome!
I posted about a PyPI-like interface for any PEP-503 compliant package repository at discuss.python.org/t/a-pypi-like-…. I'm looking for feedback before deciding whether to commit more effort in generalisation and open-development - your input is very welcome!
Excited to be sharing the powerful tooling that we've developed around PEP-503++ package indexing... I'll be presenting ideas which I believe expedite the adoption of recent index standards, and have a demo of a PyPI-like interface which can be used on any index (incl. file://)
Check out Phil Elson & Ivan Sinkarenko's talk at #EuroPython2023, in their words: The PyPI like Python package repository at CERN takes a modular approach, offering powerful flexibility to adapt to evolving needs, standards, and security threats. ep2023.europython.eu/session/the-py… 🐍
"Running a successful open source project is just Good Will Hunting in reverse, where you start out as a respected genius and end up being a janitor who gets into fights." Quote attributed to @cra, and I don't think I've ever seen anything more true posted.
A dask worker autoscaler which adapts not based on amount of work that the scheduler has, but rather on the workload of the underlying machine (e.g. imagine a cluster running on a collection of colleague's desktop machines). #doesitexistalready @dask_dev
As I'm a new face on python discourse, and given my fairly lengthy input on the PEP-691 discussion, is there a risk that the invading alien is me... 😂👽
Check out Phil Elson & Ivan Sinkarenko's talk at #EuroPython2023, in their words: The PyPI like Python package repository at CERN takes a modular approach, offering powerful flexibility to adapt to evolving needs, standards, and security threats. ep2023.europython.eu/session/the-py… 🐍
Christoph's sage advice (along with that of @cournape) were pivotal for getting @condaforge off the ground when it came to Windows builds.
Check out Phil Elson & Ivan Sinkarenko's talk at #EuroPython2023, in their words: The PyPI like Python package repository at CERN takes a modular approach, offering powerful flexibility to adapt to evolving needs, standards, and security threats. ep2023.europython.eu/session/the-py… 🐍
Now that I'm Geneva based this is really close, but not sure I'm going to be able to make it (🚼). Who do I know who is planning to attend?
We are happy to announce the CfP for EuroSciPy 2022 is open until 30 May. EuroSciPy will be back as in person event in Basel in August. euroscipy.org/2022/
TIL: When reviewing some code, I totally just learnt that you can pass arbitrary keywords to the class base list in Python. After all these years (15+), I'm clearly still learning #Python! peps.python.org/pep-3115/#spec…
Does anybody know of a tool to generate a UML diagram from type-stubs? Asking because I like the functionality of pyreverse but want to generate for native modules as well as pure Python. Stubgen does all the hard work of figuring out the interface, just need to vizualise it...
What would you say the collective name for the data structures which are natively represented by JSON was called? It is to name a method which produces something that can be JSON-ified. I see a lot of to_dict/from_dict, but that isn't really representative IMO.
Just wondering if anybody has tried making #conda capable of installing pure-python wheel files directly? That simple activity alone would unlock thousands of Python packages that aren't likely to ever be packaged in conda. The hardest part is then indexing them all... (doable)
TIL: You can insert a list into another list at a desired index using indexing... >>> l = [1, 2, 3] >>> l[0:0] = [-1, 0] >>> l [-1, 0, 1, 2, 3] Source: Reading some of the sphinx source code.
Reviewing a colleague's concurrent.futures code I asked for clarification on what was expected when the future was destroyed... The response was "I'm still quite new to futures and wasn't sure if this was a genuine question or some kind of joke" (it was genuine) #signofthetimes
United States Tendances
- 1. Cowboys 69.5K posts
- 2. Nick Smith 16.2K posts
- 3. Kawhi 4,521 posts
- 4. Cardinals 31.3K posts
- 5. #LakeShow 3,552 posts
- 6. Jerry 46K posts
- 7. #WWERaw 63.9K posts
- 8. #WeTVAlwaysMore2026 720K posts
- 9. Kyler 8,713 posts
- 10. Blazers 8,307 posts
- 11. Jonathan Bailey 29.6K posts
- 12. Logan Paul 10.5K posts
- 13. No Luka 3,813 posts
- 14. Valka 5,024 posts
- 15. Dalex 2,670 posts
- 16. Jacoby Brissett 5,801 posts
- 17. Pickens 6,726 posts
- 18. Pacers 13.5K posts
- 19. Javonte 4,449 posts
- 20. Bronny 15K posts
Vous pourriez aimer
-
Matthew Rocklin
@mrocklin -
Quansight
@quansightai -
Dask
@dask_dev -
Coiled
@CoiledHQ -
prefix.dev
@prefix_dev -
Aaron Meurer
@asmeurer -
Wolf Vollprecht
@wolfvollprecht -
Paul “π” Ivanov - @[email protected]
@ivanov -
polars data
@DataPolars -
Stephan Hoyer
@shoyer -
Conda Forge
@condaforge -
Joe Hamman
@_jhamman -
Hannah
@story645
Something went wrong.
Something went wrong.