You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Per PEP 621 a Python package can be distributed with a license metadata field set to either file: link_to_license_file or text: some_license_text.
Packages that set file instead of text can end up with very long content. This seems to be valid, as demonstrated by the long version content you can see on Bokeh's PyPI page, this project setting file indeed.
One issue here is that there is nothing in the JSON response from PyPI (e.g: https://pypi.org/pypi/bokeh/json ) that explicitly says whether the contents of license is file or text so we'll have to use some judgement. I'll get a PR up shortly..
chris48s
added
bug
Bugs in badges and the frontend
service-badge
Accepted and actionable changes, features, and bugs
and removed
question
Support questions, usage questions, unconfirmed bugs, discussions, ideas
labels
Dec 7, 2022
Are you experiencing an issue with...
shields.io
🐞 Description
Per PEP 621 a Python package can be distributed with a
license
metadata field set to eitherfile: link_to_license_file
ortext: some_license_text
.Packages that set
file
instead oftext
can end up with very long content. This seems to be valid, as demonstrated by the long version content you can see on Bokeh's PyPI page, this project settingfile
indeed.🔗 Link to the badge
This leads to a very long badge :)
![image](https://user-images.githubusercontent.com/35924738/205895964-0f24420b-6cd8-4e72-9572-cd8bb8eac970.png)
https://img.shields.io/pypi/l/bokeh
💡 Possible Solution
Shouldn't the badges be built from the License classifier instead of the core
license
metadata?The text was updated successfully, but these errors were encountered: