S3 bucket does not exist anymore

[giorgiosironi]

https://github.com/codeforscience/sciencefair-elife/blob/master/sync.js#L9 points to an S3 bucket, which does not exist anymore.

[blahah]

how strange! @gmaciocci any idea what's happened here? did eLife move to a new bucket?

[giorgiosironi]

As part of the migration to eLife 2.0 this bucket was migrated to one that is exposed through a CDN instead. Now I understand you're doing a sort of crawling so neither the API nor the CDN would cut it as they don't provide an easy way to list all the files.

If you want to get all published XML/figures/PDF ever, point at the http://prod-elife-published.s3.amazonaws.com/ bucket. It's not as stable in time as the CDN URLs, but it's the most similar option (notice there is an additional /articles prefix).

[blahah]

Ah, sorry @giorgiosironi I didn't realise at first that you work for eLife :)

Thanks for the explanation. Will my existing credentials work for the new bucket, or will I need new ones?

[giorgiosironi]

This bucket is public, so you shouldn't need credentials.

[blahah]

Public to access over HTTP, but not to do syncing over the S3 API I think? With the last bucket I needed credentials to be able to do reasonable sync operations.

[giorgiosironi]

I found the IAM policy for the user RichardSmithUnna, it is updated for prod-elife-published now.