Fix bug, aggregate value var works with blank node in upsert

[mangalaman93]

Fixes #4712

---

This change is 

[harshil-goel]

Reviewable status: 0 of 2 files reviewed, all discussions resolved (waiting on @harshil-goel, @manishrjain, and @martinmr)

[MichelDiz]

Reviewable status: 0 of 2 files reviewed, 1 unresolved discussion (waiting on @mangalaman93, @manishrjain, and @martinmr)

---

dgraph/cmd/alpha/upsert_test.go, line 2818 at r1 (raw file):

    me() {
      sVerIncr as sum(val(VerIncr))

This PR has been tested and works perfectly. However, I have a concern about this usage. We should eliminate the empty block from this procedure, as it was not created for this and can lead to future problems.

However, it is not a blocking for me. It's just a concern. We can merge it. But it would not be a definitive solution to issue 4767. In terms of practicality and future issues.

[manishrjain]

Reviewed 2 of 2 files at r1.
Reviewable status: all files reviewed, 2 unresolved discussions (waiting on @mangalaman93 and @martinmr)

---

edgraph/server.go, line 476 at r1 (raw file):

// If val(variable) exists in a query, but the values are not there for the variable,
// it will ignore the mutation silently.
func updateValInNQuads(nquads []*api.NQuad, qc *queryContext, isSet bool) []*api.NQuad {

Don't need isSet, unless Dgraph would complain about this.

[mangalaman93]

Reviewable status: all files reviewed, 2 unresolved discussions (waiting on @manishrjain, @martinmr, and @MichelDiz)

---

dgraph/cmd/alpha/upsert_test.go, line 2818 at r1 (raw file):

Previously, MichelDiz (Michel Conrado) wrote…

This PR has been tested and works perfectly. However, I have a concern about this usage. We should eliminate the empty block from this procedure, as it was not created for this and can lead to future problems.

However, it is not a blocking for me. It's just a concern. We can merge it. But it would not be a definitive solution to issue 4767. In terms of practicality and future issues.

It doesn't solve the whole problem, agreed.

---

edgraph/server.go, line 476 at r1 (raw file):

Previously, manishrjain (Manish R Jain) wrote…

Don't need isSet, unless Dgraph would complain about this.

Turns out Dgraph complains.

[MichelDiz]

Reviewed 2 of 2 files at r1.
Reviewable status: all files reviewed, 1 unresolved discussion (waiting on @manishrjain and @martinmr)