Paul Ganssle
@pganssle
Followers
2K
Following
15K
Media
310
Statuses
6K
Programmer working @Google in NYC. Maintainer of python-dateutil, Python core developer and general FOSS contributor. Fediverse: @[email protected]
Boston, MA, USA
Joined June 2015
I think I should have used Cunningham's law instead of asking questions in this post: https://t.co/X6x0sGhuny I suspect I would have gotten more people weighing in. #python
discuss.python.org
TL;DR (Edited): I think that lazy imports should probably be allowed in with blocks*, and it’s important that that be in this PEP*, because the main use case is ease of backporting, which is defeated...
1
0
1
While I’ve hosted lightning talks at #PyConUS before, I’ve finally given one today! Enforce documentation coverage with `interrogate`! It’s got a cute logo 😍🦥 https://t.co/vFnfck5cW9
0
4
30
Dear #pycon presenters. Here are my accessibility requests for next year. 1. No colorized code on black backgrounds. Blue or purple or gray is really hard to read. 2. Use large text please. I cannot read your regular terminal font. PLEASE.
15
67
565
Hey, do you know the @ThePSF is hiring for a full-time position called the Deputy CPython Developer in Residence? It's a special role where you'd work with me and the rest of the core team on directly shaping the future of #Python. Want to know more? https://t.co/gA3dPUt6o9
lukasz.langa.pl
There is a job opening for the Deputy CPython Developer in Residence. You should consider it, it’s an adventure of a lifetime!
0
49
116
I tried looking for an labs at Harvard and MIT, but I am not seeing any researchers there doing work that seems like it would need research subjects.
1
0
0
A perfect illustration of why clinical trial registration needs an upgrade to a full #RegisteredReports model. @RegReports There are some logistical challenges, but they are solvable. The alternative is this👇 cc @eturnermd1
19% of 755 phase 3 cancer trials had changes to the primary end points in the paper compared to the registration. 70% of those were not disclosed in the paper. Changes were associated with likelihood of a positive outcome (OR=1.86). https://t.co/5Jh9qWyGmb
0
15
24
Long shot, but anyone with Android / Android Auto experience know what's going on here? https://t.co/MhmBh4i01a
#android #auto #androidauto
0
1
0
Carter's has this shirt for kids, but I can't find an adult version anywhere. Trying to tell Amazon or a search engine that it's important that the shirt have all three colors (much less in a specific order) seems impossible.
0
0
0
One day I hope ML technology advances to the point that I can describe a shirt and find out if someone is selling something like it.
1
0
0
PEP 658 has finally landed on PyPI! Wheels uploaded as of about 15 minutes ago now have the appropriate information served from the simple APIs, and METADATA files available on https://t.co/hVzzMojVbA.
2
16
71
This is great
Attractive nuisances in software design: https://t.co/5HokP2j1WN A common anti-pattern where a problem has a solution that is obvious, intuitive and wrong. #python #programming
0
1
2
It's relatively easy to make a drop-in replacement for these, but also we're deprecating them because they're conceptually the wrong thing to do, so it's best to migrate to using aware datetimes if possible:
blog.ganssle.io
A public service announcement about the dangers of utcnow and utcfromtimestamp and the benefits of using their replacements.
1
0
3
datetime.utcnow and datetime.utcfromtimestamp will be deprecated in #python 3.12: https://t.co/MdwgRgw5qp If you maintain a package, now is probably a good time to grep your source code for utcnow and utcfromtimestamp to get out ahead of the deprecation warnings. 📅🕐
github.com
Feature or enhancement Previously, we have documented that utcnow and utcfromtimestamp should not be used, but we didn't go so far as to actually deprecate them, and I wrote a whole article abo...
1
9
26
Bugs like `dt.replace(year=dt.year + 1)`, code that works with ordinal day-of-year, etc.
0
0
0
Psst. Next year is a leap year. Depending on your deployment story, now (or 3 months ago😅) might be a good time to start looking into logic that won't handle leap years correctly.
2
0
3