Closed kevinmarks closed 7 years ago
The spec says:
If more than one URL is specified, it is expected that the publisher pings each of these URLs, so the subscriber may subscribe to one or more of these.
It was a side effect of this implementation that the second one in the list is used. I changed it to use the first in the list, so at least that should be less unexpected.
It might make sense to test both and report on each - if other subscription implementations vary which they choose it could make problems with the other one.
On Thu, 1 Dec 2016, 09:10 Aaron Parecki, notifications@github.com wrote:
The spec says:
If more than one URL is specified, it is expected that the publisher pings each of these URLs, so the subscriber may subscribe to one or more of these.
It was a side effect of this implementation that the second one in the list is used. I changed it to use the first in the list, so at least that should be less unexpected.
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/aaronpk/websub.rocks/issues/6#issuecomment-264232500, or mute the thread https://github.com/notifications/unsubscribe-auth/AAGCwKMjAZizxieSNJJUCBXu_khUXfWYks5rDv9agaJpZM4K_Vad .
Also, when testing subscribers, see which they choose from multiple choices and document that. If they are consistent about first/last could change the spec to match.
On Thu, 1 Dec 2016, 09:17 Kevin Marks, kevinmarks@gmail.com wrote:
It might make sense to test both and report on each - if other subscription implementations vary which they choose it could make problems with the other one.
On Thu, 1 Dec 2016, 09:10 Aaron Parecki, notifications@github.com wrote:
The spec says:
If more than one URL is specified, it is expected that the publisher pings each of these URLs, so the subscriber may subscribe to one or more of these.
It was a side effect of this implementation that the second one in the list is used. I changed it to use the first in the list, so at least that should be less unexpected.
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/aaronpk/websub.rocks/issues/6#issuecomment-264232500, or mute the thread https://github.com/notifications/unsubscribe-auth/AAGCwKMjAZizxieSNJJUCBXu_khUXfWYks5rDv9agaJpZM4K_Vad .
I'll make it show an option to choose which hub to test when it finds more than one
Good idea: #8
https://mastodon.social/users/kevinmarks.atom has
and only the superfeedr one is seen.