Skip to content
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

adjust neurosift service endpoint URL, passing additional info #1706

Merged
merged 5 commits into from
Nov 22, 2023

Conversation

magland
Copy link
Contributor

@magland magland commented Oct 13, 2023

This adjusts the external endpoint URL for neurosift such that it:

  • always provides the version of the asset URL that includes that asset ID (rather than just doing this for embargoed dandisets)
  • provides the dandiset ID and version

This is information I would like to have access to on the neurosift end.

I tried to do this in the most straightforward way I could -- if adjustments are needed in the future I think it wouldn't be a problem since it only affects the one file. But if there is a better way @waxlamp, feel free to suggest.

The other external services should work the same as they did before.

I realize that this is related to #1655 but I wanted to start from the latest commit that includes the other related changes from #1692

@waxlamp
Copy link
Member

waxlamp commented Nov 2, 2023

@magland, would you mind enabling me to push commits to your branch? I have a few minor updates to make this tighter that I'd like to add.

@magland
Copy link
Contributor Author

magland commented Nov 2, 2023

@magland, would you mind enabling me to push commits to your branch? I have a few minor updates to make this tighter that I'd like to add.

Sure, I have added you to https://github.com/magland/dandi-archive

@yarikoptic
Copy link
Member

yarikoptic commented Nov 2, 2023

I also see
image
on the right, so @waxlamp should be able to push to your branch @magland without you adding him explicitly AFAIK... even I am allowed to commit to it ;) (I removed that comment since then)
image

@waxlamp
Copy link
Member

waxlamp commented Nov 3, 2023

I also see image on the right, so @waxlamp should be able to push to your branch @magland without you adding him explicitly AFAIK... even I am allowed to commit to it ;) (I removed that comment since then) image

Yes, I apologize for being rather unclear in my original request, @magland -- I just wanted you to enable the option to allow maintainers to collaborate on your PR branch.

@waxlamp
Copy link
Member

waxlamp commented Nov 3, 2023

@magland I made a few modifications that I'd like your feedback on. Basically, I made the "API" less abstract by removing the base API information, and supplying more explicit asset URLs (one of which simplifies your Neurosift template changes significantly).

I also removed the asset ID from the transmitted data because your use case no longer needs it and I want to make sure we don't overplay this new "API" without a serious discussion. I'm going to convene such a discussion ASAP so that we have a chance to weigh in on this without delaying merging this PR (when it's ready). I'll include you in that discussion, @magland.

In the meantime, let me know what you think of my changes (and @AlmightyYakob, please weigh in as well).

Thanks, @magland, for sending in this PR, and sorry for the delay in getting to this review.

@magland
Copy link
Contributor Author

magland commented Nov 3, 2023

@waxlamp This looks great to me. You are right that neurosift does not need the asset ID if it has the asset_dandi_url. And I do prefer the asset_dandi_url rather than the asset_url even though it requires an extra step on page load because then it is handling things in the same way whether or not it's an embargoed dataset. So what you have there looks perfect.

@yarikoptic I didn't know about that capability - that's helpful. Is there anything you want me to do at this point to enable something?

@yarikoptic
Copy link
Member

@yarikoptic I didn't know about that capability - that's helpful. Is there anything you want me to do at this point to enable something?

nope, AFAIK all is good here and in #1666 in that regard.

@yarikoptic
Copy link
Member

@waxlamp ping on this, so we could progress forward (there are PRs, #1755, coming which would conflict already)

@magland
Copy link
Contributor Author

magland commented Nov 22, 2023

@waxlamp @yarikoptic It would be great if this could be merged/released soon. I ran into an issue with an embargoed dataset where the neurosift link didn't work because it was the s3 bucket link (see below, although you prob don't have access). I thought we had solved that with the previous PR #1692. But in any event, I think this PR here would certainly do the trick because then it would always pass the api URL to neurosift.

https://dandiarchive.org/dandiset/000620/draft/files?location=sub-Elgar&page=1

Copy link
Member

@waxlamp waxlamp left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Sorry for the delay on this. Looks good to me!

@jjnesbitt jjnesbitt added patch Increment the patch version when merged release Create a release when this pr is merged labels Nov 22, 2023
@jjnesbitt jjnesbitt merged commit d6cd592 into dandi:master Nov 22, 2023
3 checks passed
@dandibot
Copy link
Member

🚀 PR was released in v0.3.64 🚀

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
patch Increment the patch version when merged release Create a release when this pr is merged released This issue/pull request has been released.
Projects
None yet
Development

Successfully merging this pull request may close these issues.

5 participants