-
Notifications
You must be signed in to change notification settings - Fork 1k
Table columns in "Download Files" section are disproportionate #3453
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Comments
@emddudley Thanks for the bug report, especially for the specific example case of NumPy. I see how this is especially annoying in releases with, say, five or more packages. I want to check: When you are scanning a list of packages, are you pretty much always looking for the same metadata (that we could perhaps present in a different column), such as the platform the wheel was built for? |
I look for the ABI ( I sometimes also check the "local version label", for example the Sometimes I check the version to confirm the absence of a local version label. I typically look for the file extensions. I trust that more than the inferred file types. Here is another case where I check for the above information: |
Local versions aren't allowed on PyPI, if someone managed to upload one, that's a bug. |
I got that filename from a third-party site. On PyPI I check for the absence of the label. Things don't work if I get the wrong wheel. But now that I know PyPI doesn't allow local version labels, I suppose it wasn't a reliable way to check... |
@emddudley I appreciate the further data -- thank you! This will help our designer think about how to solve this. She'll look at this issue within the next few business days. Thanks again. |
thanks for this feedback @emddudley - we definitely need to do something about this. I will add this point to our upcoming user tests - it will be interesting to see how others are reading this table and what data they think is important :) |
@nlhkabu A user pointed me to #929 which also demonstrated concern about this. I like the screenshot in #929 (comment) but am not sure whether it was a mockup or not :) @techtonik could you clarify? |
@emddudley In a few minutes this will be live on test.pypi.org and pypi.org so you can see whether the improved table style solves this problem for you. Please let us know! Thanks. |
@brainwane: I cleared my browser cache but I am not seeing any difference. |
Likely hasn't fallen out of the cache yet. |
It's better and I could work with it in its current state. I do still have some nitpicks though. The remaining issues I see are the table cell padding and the handling of very long filenames. The table cells are padded such that an entire line of text could fit between rows. I think this slightly hinders visual scanning. I do understand that the padding may be consistent with other parts of the site, and that there may be other usability considerations. Still, my preference is less padding--my eye is better able to compare the shape of the text without actually reading it. Compare the following. Secondly, very long filenames still seem like a problem. Again, they disrupt scanning. The only time I see these long names are with OS X wheels, and I am guessing that it is rare that anyone reads the entire OS X platform string. Just as an idea, what if very long filenames were abbreviated with an ellipsis, and you could view the full name by hovering the cursor and viewing a tooltip? |
@emddudley Sorry for the delay. @yeraydiazdiaz is this something you could address in #3783? |
The table columns in the "Download Files" section of a project page are not proportioned well. Filename, Size & Hash are all crammed into a single tiny column, while three less important things (File Type, Python Version, and Upload Date) take up 75% of the width.
I am not able to visually scan for the files I want because the filenames are so wrapped.
Case in point: https://pypi.org/project/numpy/#files
Compare to: https://pypi.python.org/pypi/numpy
One potential solution could be to put data on two lines, something like this:
The text was updated successfully, but these errors were encountered: