go.dev: search can show "zero results" on an exact package match

[hunterloftis]

What did you do?

I typed a full package url into the search box at go.dev, hoping to go to a particular package:

What did you expect to see?

I expected to either go directly to that package, or for that package to be a link in the search results.

What did you see instead?

[dmitshur]

/cc @julieqiu

[julieqiu]

@hunterloftis I wasn't able to reproduce that bug - visiting https://pkg.go.dev/search?q=github.com%2Fhunterloftis%2Fstorj%2Frelay redirects me to https://pkg.go.dev/github.com/hunterloftis/storj/relay?tab=doc.

Would you mind verifying if we are looking at the same URL?

[findleyr]

It looks like the first version of this package was only indexed a few hours ago.
https://index.golang.org/index?since=2020-02-07T00:00:00.000000Z
https://pkg.go.dev/github.com/hunterloftis/storj/relay?tab=versions

So it's likely that when this was first searched, we didn't have it.

[hunterloftis]

Thanks for trying to reproduce! I guess the search system index is a little behind the actual archive. It’s not a big deal but I wouldn’t be surprised if others run into the same issue as they are publishing packages and trying to verify them on go.dev.

On Fri, Feb 7, 2020 at 2:03 PM findleyr ***@***.***> wrote: It looks like the first version of this package was only indexed a few hours ago. https://index.golang.org/index?since=2020-02-07T00:00:00.000000Z https://pkg.go.dev/github.com/hunterloftis/storj/relay?tab=versions So it's likely that when this was first searched, we didn't have it. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#37111?email_source=notifications&email_token=AACY7VMHF3URXHEWA5T23N3RBWWAJA5CNFSM4KRQZV32YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOELEF56I#issuecomment-583556857>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AACY7VPHW72GMFZ2QPSBCU3RBWWAJANCNFSM4KRQZV3Q> .

-- - from mobile

[julieqiu]

Closing this issue as a duplicate of #37002.