-
Notifications
You must be signed in to change notification settings - Fork 1.5k
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
1.1.0-rc1: count selector at root ignores given value #3840
Comments
Hi @mehdiym can you provide a sample data set and exact steps you did starting from a fresh cluster that led to the result you're seeing? |
I'll send you a dump of my database on slack. |
Great, looking forward to your message on Slack. |
I can reproduce the issue with the database p, w, and zw dirs you shared. I suspect this issue is due to how your integration tests load data into Dgraph. I can't reproduce this with live loader, bulk loader, or constructing my own sample data set by loading via mutations. If you can share the Dgraph data loading steps that's done by your tests we should be able to reproduce this issue. |
In Dgraph v1.1.0-rc1 we are not updating the count index for delete mutations. Specifically, the count index for reverse edges. |
I'm glad you found it out. |
@mangalaman93 ran
|
We cannot use txn.cache.GetFromDeltas for reverse mutations becasue cache will not have deltas corresponding to reverse key. Fixes #3840
Working great now, all my integration tests pass, thx. |
What version of Dgraph are you using?
1.1.0-RC1
Steps to reproduce the issue (command/config used to run Dgraph).
Set in the schema a predicate with
@reverse
and@count
, eg:Add nodes and edges with that predicate, then do a query based on the count, eg:
I would expect results having only one reverse edge
~game_answer
.My result:
Oddly, if I add that selector as a post filter, it works, eg:
Gives:
The text was updated successfully, but these errors were encountered: